Experience of Graduate Counseling Students During COVID-19: Application for Group Counseling Training

Bilal Urkmez, Chanda Pinkney, Daniel Bonnah Amparbeng, Nanang Gunawan, Jennifer Ojiambo Isiko, Brandon Tomlinson, Christine Suniti Bhat

 

The COVID-19 pandemic resulted in many universities moving abruptly from face-to-face to online instruction. One group of students involved in this transition was master’s-level counseling students. Their experiential group counseling training (EGCT) program started in a face-to-face format and abruptly transitioned to an online format because of COVID-19. In this phenomenological study, we examined these students’ experiences of participating and leading in six face-to-face and four online EGCT groups. Two focus groups were conducted, and three major themes emerged: positive participation attributes, participation-inhibiting attributes, and suggestions for group counseling training. The findings point to additional learning and skill development through the online group experience as well as its utility as a safe space to process the novel experience brought about by COVID-19.

Keywords: experiential group counseling training, phenomenological, COVID-19, face-to-face, online format

 

Most of what is known about group counseling and the training of group counselors has been learned from groups that occur in face-to-face group environments (Kozlowski & Holmes, 2014). This includes seminal works on group counseling’s therapeutic factors, such as universality, altruism, instillation of hope, cohesiveness, existential factors, interpersonal learning, self-understanding, and catharsis (Yalom & Leszcz, 2005). Researchers have found positive contributions of group therapeutic factors toward therapy outcomes (Behenck et al., 2017), and they have explored the experiences of group members in face-to-face group counseling settings, including the interpersonal and intrapersonal processes of members (Holmes & Kozlowski, 2015; Krug, 2009; Murdock et al., 2012). By contrast, there is considerably less research on online group counseling (Kozlowski & Holmes, 2014) or group counselors’ training in online modalities (Kit et al., 2014; Kozlowski & Holmes, 2017).

In this qualitative study, we utilized the phenomenological method to explore and compare master’s-level students’ experiences of participating in and leading during six face-to-face and four online experiential group counseling training (EGCT) groups as part of an introductory group counseling course. The master’s-level counseling students began their EGCT in face-to-face groups, and because of the COVID-19 pandemic, they continued to meet in four online groups after their university decided to suspend all face-to-face instruction.

Experiential Groups in Counselor Education
     Group counseling training is one of the eight core areas of required training for counselors stipulated by the Council for the Accreditation of Counseling and Related Educational Programs (CACREP; 2015). In order to learn the complex group processes necessary for effective group counseling, master’s-level counseling students are required to participate in EGCT (Association for Specialists in Group Work [ASGW], 2007; CACREP, 2015). For CACREP-accredited master’s programs, at least 10 clock hours of group participation during one academic semester are required (CACREP, 2015). During this experiential training, students learn to be both group counseling participants and group counseling leaders (Ieva et al., 2009) and gain valuable experience in and insight into group dynamics, group processes, and catharsis (Ohrt et al., 2014).

Master’s-level counseling students “benefit a great deal when allowed to develop practical and relevant clinical skills” (Steen et al., 2014, p. 236). Experiential training in group counseling also promotes self-awareness, personal growth, and a greater understanding of vulnerability and self-disclosure in the learners (Yalom & Leszcz, 2005). The experiential component of group counseling training provides an environment for counseling students to experience vicarious modeling, self-disclosure, validation, and genuineness from their classmates (Kiweewa et al., 2013). Finally, these experiential opportunities promote students’ self-confidence (Ohrt et al., 2014; Shumaker et al., 2011; Steen et al., 2014).

Online Counseling
     Barak and Grohol (2011) defined online counseling as “a mental health intervention between a patient (or a group of patients) and a therapist, using technology as the modality of communication” (p. 157). Counselors are increasingly using more digital modalities in their practice (Anthony, 2015; Richards & Viganó, 2013), and it is being seen as a viable alternative to support clients (Hearn et al., 2017). Since the start of the COVID-19 pandemic, counselors have begun to use more online modalities to provide counseling services (Peng et al., 2020). Online counseling began to emerge as a potential solution for mental health services when providers were forced to discontinue or scale down in-person services and adjust to virtual formats during the pandemic (Békés & Aafjes-van Doorn, 2020; Peng et al., 2020; Wind et al., 2020). Peng et al. (2020) noted the effects COVID-19 have had on the delivery of mental health services in China. They mentioned the governmental and authorities’ support for preparedness and response and the multidisciplinary enhancement of remote intervention quality for clients. They also suggested that governments should integrate the mental health interventions related to COVID-19 into existing public mental health emergency preparedness and response structures.

Because of the growing importance of online counseling, it is essential to train counseling students to conduct online counseling, including online group counseling, effectively. Understanding master’s students’ experiences in online EGCT can help identify potential challenges they may face during their training. It is also important to explore students’ experiences in face-to-face and online EGCT groups to better understand possible future training needs and help counselor educators create an educational curriculum that addresses group counseling knowledge and skills for online groups. There is currently a lack of information about how to train counseling students in the delivery of online counseling (Kozlowski & Holmes, 2014), and specifically group counseling (Kit et al., 2014).

Professional and Accreditation Bodies’ Guidance on Technology
     The American Counseling Association (ACA) Code of Ethics states, “Counselors understand that the profession of counseling may no longer be limited to in-person, face-to-face interactions” (2014, p. 17). The ASGW Best Practices Guidelines require that “Group workers are aware of and responsive to technological changes as they affect society and the profession” (ASGW, 2007, p. 115, A.9). Similarly, CACREP (2015) indicates “students are to understand the impact of technology on the counseling profession” (2.F.1.j) as well as “the impact of technology on the counseling process” (2.F.5.e). CACREP also emphasized that students understand “ethical and culturally relevant strategies for establishing and maintaining in-person and technology-assisted relationships” (2.F.5.d). Additionally, the Association for Counselor Education and Supervision (ACES; 2018) provides guidelines for online instruction featuring descriptions regarding course quality, content, instructional support, faculty qualifications, course evaluation procedures and expected technology standards.

Online Group Counseling
     Textbooks on group counseling have mainly approached EGCT in face-to-face formats (e.g., G. Corey, 2016; Yalom & Leszcz, 2005). Given the growing interest and demand for online counseling in recent years (Holmes & Kozlowski, 2015; Kozlowski & Holmes, 2017), COVID-19 has highlighted the need for greater awareness and understanding of online group counseling training. However, there is limited research on online group counseling and counseling students’ training in online group counseling.

Kozlowski and Holmes (2014) explored master’s-level counseling students’ experience in an online process group, reporting themes of participants’ experiences of a linear discussion, role confusion, and feelings of being disconnected, isolated, and unheard. In 2015, Holmes and Kozlowski expanded on their work with a study on master’s-level counseling students’ experiences in face-to-face and online group counseling training. They found that the online group participants felt significantly less comfortable than participants in the face-to-face group. Further, participants in the study evaluated face-to-face groups as preferable for participation, social cohesion, and security (Holmes & Kozlowski, 2015). Lopresti (2010) compared students’ group therapy experiences between face-to-face and online group counseling methods using synchronous text-based software. This research involved six master’s-level students engaging in an 8-week, 60-minute, weekly online group counseling session using the WebCT chat system. Results indicated that in the online format, some participants reported self-disclosure more easily, but they also shared that it was easy to hide behind the screen and to censor themselves.

Effectiveness of Online Group Counseling
     Some researchers have observed the efficacy of online support groups (Darcy & Dooley, 2007; Freeman et al., 2008; Lieberman et al., 2010; Webb et al., 2008). Haberstroh and Moyer (2012) reported that professionally moderated online support groups could supplement face-to-face counseling, especially for clients who want regular daily support during the process of recovering from self-injury. They also found that online group interaction provided clients with opportunities to engage in healthy self-expression and reduce their sense of loneliness and isolation (Haberstroh & Moyer, 2012). King et al. (2009) examined the effectiveness of internet-based group counseling to treat clients with methadone substance abuse, reporting that internet-based group counseling could reduce resistance and non-adherence in clients. Clients expressed satisfaction with the process and reported convenience and higher levels of trust in confidentiality because they were able to participate from home.

Similarly, Gilkey et al. (2009) reported the advantages and disadvantages of synchronous videoconferencing (SVC) web-based interventions. This study involved families with children with traumatic brain injury. The results revealed that SVC had the potential for family-based therapy delivery. However, it required important factors such as client readiness to address their issues and patience with the technology’s imperfections. SVC could reduce barriers to treatment with motivated families from diverse backgrounds. Nevertheless, the online group experience is vulnerable to the impact of technology glitches, privacy issues, disruptions in connectivity, and personal detachment (Amulya, 2020). In online group therapy, Weinberg (2020) identified four obstacles: managing the frame of the treatment, the disembodied environment, the question of presence, and the transparent background.

Purpose of Study and Research Questions
     In March 2020, as a result of the pandemic, our university moved most face-to-face classes to virtual environments following statewide restrictions for in-person gatherings. This sudden change led to a unique experience for first-year master’s-level counseling students enrolled in an introductory group counseling course at a CACREP-accredited program in the Midwest. It was planned that students would participate in 10 face-to-face EGCT groups of 90 minutes each to fulfill the CACREP (2015) group counseling experiential training requirements. Doctoral students facilitated the first five group counseling experiences for the counselors-in-training. The plan was for two master’s students to lead face-to-face groups under the supervision of doctoral students for the remaining five groups (6–10). However, the university closed for 2 weeks after Session 6 was completed. As a result, when classes resumed, they were online. EGCT Sessions 7 through 10 were conducted online using Microsoft Teams with master’s students leading and doctoral students supervising. Thus, in a single semester, the master’s students had the experience of participating in and leading both face-to-face and online groups. Our study was guided by the following research question: What were master students’ experiences of participating and leading in both face-to-face and online EGCT groups?

Methods

Research Design
     Qualitative methodology was used to explore first-year master’s students’ experiences of participating and leading in both face-to-face and online formats of EGCT. Our aim was to build an understanding of their experience shifting to an online modality with a specific interest in their attitudes, learning, facilitating, and adaptation to these two environments. For this purpose, a phenomenological approach was appropriate for investigating students’ unique experiences in both versions of the EGCT groups. Moustakas (1994) defined phenomenology as an approach for “comprehending or having in-depth knowledge of a phenomenon or setting and . . . attained by first reflecting on one’s own experience” (p. 36). In a phenomenological study, the aim is to describe the essence of individuals’ experiences with a certain phenomenon (Creswell & Creswell, 2018).

Participants and Procedures
     IRB approval was obtained, and purposive sampling was implemented with a recruitment email. All participants were recruited from a CACREP-accredited counseling program in the Midwest  United States. Our inclusion criteria were that participants must be current master’s-level counseling students and must be enrolled in a group counseling course. In addition, each participant must have experienced both participating in and leading at least one EGCT session during the prior term.

The invitation to participate in a focus group was emailed to all students enrolled in the group counseling course in the prior term. It included information about the study, addressed voluntary participation, and explained the entirely separate nature of participation in the focus group from evaluation of performance in the group class that had concluded. This recruitment email was sent out a total of three times within a 3-week period before the study was conducted.

Nine students agreed to participate in the study, and written consent forms were sent to them via email to read and review. Of the nine participants, three self-identified as male and six self-identified as female. Seven participants identified as White and two identified as “other,” and the age range was 18–34 years old. Two participants were specializing in school counseling, three in clinical mental health counseling, three in clinical mental health/clinical rehabilitation counseling, and one in clinical mental health/school counseling.

Before the focus group, prospective participants were emailed a copy of the semi-structured interview questions to alleviate any anxiety or concerns about the questions that would be asked during the study. Prospective participants were also invited to ask any questions at the start of the focus group and were then invited to provide verbal consent. To secure confidentiality, participants were assigned a code consisting of letters and numbers to protect their identity. Participants’ identification codes, with corresponding names, were kept securely in the possession of the first author, Bilal Urkmez.

Focus Groups
     Focus groups were used because they allow students to share their experiences with EGCT groups and compare points of view (Krueger & Casey, 2014). Two online focus groups were held—one with five participants (one male, four females) and one with four participants (two males, two females). Participants received invitation links from the focus group facilitator via Microsoft Teams. All participants were familiar with Microsoft Teams because they had used it for their experiential groups and classes after moving to online instruction. Urkmez contacted the university’s IT department regarding the protocol of recording and securing the video and audio of the focus groups on Microsoft Teams.

Our fifth and sixth authors, Jennifer Ojiambo Isiko and Brandon Tomlinson, who led and supervised the original EGCT groups, conducted the focus groups. Care was taken to ensure that master’s students were not placed in a focus group led by the same doctoral student who had previously led and supervised their 10-session EGCT groups.

We used Krueger and Casey’s (2014) guidelines to create a semi-structured focus group protocol. Open-ended questions were built in for the focus group leaders to use as prompts to facilitate discussion when necessary. The online focus groups lasted approximately 60 minutes. All the conversations were recorded and then transcribed verbatim by the designated focus group facilitator.

Authors’ Characteristics and Reflectivity
     Our research team consisted of two counselor educators with experience teaching and facilitating group counseling courses and five counselor education doctoral students. All doctoral students were part of a single cohort, and all had prior experiences facilitating group counseling. The counselor educators were Urkmez, who self-identifies as a White male, and Christine Suniti Bhat, an Asian female. The doctoral students were Chanda Pinkney, an African American female; Daniel Bonnah Amparbeng, an African male; Nanang Gunawan, an Asian male; Isiko, an African female; and Tomlinson, a White male. Before data collection, we met to discuss focus group questions, explore biases and assumptions, and assign focus group leaders for the study.

Our team used multiple strategies to establish trustworthiness. As two of the researchers taught group counseling and five of the researchers had led and supervised the EGCT groups, it was necessary to discuss possible biases before and during the data analysis process to ensure that the resulting themes and subthemes emerged from participants’ responses (Bowen, 2008).

First, some of the researchers shared that they believe face-to-face group counseling is better than online group counseling because they do not personally like to take or teach online courses in their education. All research members taught, learned, and supervised EGCTs predominantly in face-to-face environments prior to the study and pandemic. Secondly, some of the researchers also mentioned their frustrations with learning and supervising online. These discussions were held to promote awareness of potential biases so as to avoid focusing on the negative experiences of the master’s students. Bracketing was implemented throughout the study to reduce researchers’ possible influence on participants of favoring face-to-face counseling environments (Chan et al., 2013). This measure helped ensure the validity of the study’s data collection and analysis by having the researchers put aside any negative experiences of online learning environments during the pandemic (Chan et al., 2013). Urkmez, Pinkney, Bonnah Amparbeng, Gunawan, Isiko, and Tomlinson analyzed the data first, fulfilling investigator triangulation (Patton, 2015). This same group then met several times to discuss their analyses of the transcripts and agree upon the significant statements and themes.

Experiential Group Counseling Training
     Twenty-eight first-year master’s students were enrolled in an introductory group counseling course in the spring 2020 academic semester. The EGCT groups were a required adjunct to the didactic portion of the course. EGCT sessions for the master’s students met weekly for 90 minutes and were set up so that the master’s students were participants for Sessions 1 through 5 (led by doctoral students) and were leaders for Sessions 6 through 10 (supervised by doctoral students). All 10 sessions were planned to be face-to-face sessions. Doctoral students were enrolled in an advanced group counseling course, and their participation was a required component of the course.

During the first five sessions, doctoral students’ responsibilities as leaders included facilitating meaningful interaction among the participants, promoting member–member learning, and encouraging participants to translate insights generated during the interaction into practical actions outside the group (G. Corey, 2016). For Sessions 6–10, in the role of supervisors, doctoral students’ responsibilities were to mentor and monitor the master’s students’ group leadership skills and provide verbal feedback immediately after the session. Doctoral students also provided written feedback to both the master’s students and group counseling course instructors. Additionally, the doctoral students engaged in peer supervision with each other under the tutelage of the advanced group counseling course instructor, discussing how EGCT could be supervised more effectively.

As stated previously, two master’s students started to co-lead the EGCT groups during Session 6, which was conducted face-to-face. After Session 6, in-person classes were canceled by the university in response to COVID-19, so the remaining four sessions of EGCT were conducted online on Microsoft Teams. The online groups were conducted synchronously on the same day and time as the face-to-face groups had been conducted in the earlier part of the semester.

Session 7 was the first synchronous online session of the EGCT and deserves special mention. Prior to Session 7, the doctoral students received brief training on Microsoft Teams. The master’s students had no previous exposure to Microsoft Teams. Thus, during Session 7, the doctoral students provided support by demonstrating how Microsoft Teams worked and processing the master’s students’ thoughts, feelings, and levels of wellness in relation to the sudden pandemic. Students resumed leading the online synchronous groups for Sessions 8, 9, and 10 under doctoral students’ supervision.

Data Analysis
     Isiko and Tomlinson led the two focus groups and transcribed the data collected from the participants who shared their experiences in the focus groups. We utilized the phenomenological data analysis method described by Moustakas (1994). Urkmez, Pinkney, Bonnah Amparbeng, Gunawan, Isiko, and Tomlinson conducted the data analysis while Bhat served as a peer debriefer because of her position of seniority in terms of expertise in not only qualitative methodology, but also group counseling research, as well as her experience of more than 15 years in teaching both master’s- and doctoral-level group counseling courses at the CACREP-accredited program. Her primary role was to read the transcripts, review the raw data and analysis, and scrutinize established themes to point out discrepancies (Creswell & Creswell, 2018).

Our research team (except for Bhat) met to discuss our potential biases and bracket our assumptions about the phenomenon under investigation. Then, each of us independently read all transcripts multiple times to become familiar with the data. Next, we reviewed the transcripts according to the horizontalization phase of analysis (Moustakas, 1994). Moustakas defined the horizontalization phase as the part of the analysis “in which specific statements are identified in the transcripts that provide information about the experiences of the participants” (Moustakas, 1994, p. 28). During this step, we independently reviewed each transcript and identified significant statements that reflected the participants’ interpretations of their experiences with the phenomenon. We identified these significant statements based on the number of times they were mentioned both within and across participants. From this point, we each independently created a list of significant statements.

Subsequently, we met to review our lists to establish coder consistency, create initial titles for the themes, and place data into thematic clusters (Moustakas, 1994). Each of our themes and related subthemes were similar in content and typically varied only in the titles used. Titles for themes and subthemes were discussed until consensus was obtained. We revisited the horizontalized statements and discussed our different perspectives. Next, we evaluated the most commonly occurring themes and created a composite summary of each theme from the participants’ experiences. After these steps, we arrived at a consensus about each theme’s essential meaning and decided on specific participant quotes that represented each theme.

Findings

We identified three main themes related to the participants’ experiences of taking part in and leading both face-to-face and online EGCT. The three main themes were positive participation attributes, participation-inhibiting attributes, and suggestions for group counseling training.

Positive Participation Attributes
     The central theme of positive participation attributes focused on exploring master’s students’ perceptions about what helped them actively participate in both online and face-to-face EGCT groups as a group member. Five subthemes were identified in the main theme of positive participation attributes: (a) knowing other group members, (b) physical presence, (c) comfortability of online sessions, (d) cohesiveness, and (e) leadership interventions.

Knowing Other Group Members
     The EGCT group involved graduate-level counseling students who knew each other for a semester before engaging in the EGCT. Study participants shared that seeing familiar faces provided a safe and supportive environment for them to participate in both face-to-face and online group sessions as a group member. One participant noted that “a part of it helped because it was many people I had already known,” and another participant stated that “it was easier to have face-to-face after we had already kind of met everybody in the semester and so I wasn’t worried about confidentiality. I wasn’t in this group with a whole bunch of strangers.” Participants noted that knowing other group members helped them to participate actively in EGCT. They reported that having familiar faces in the group made them feel comfortable and connected, and that it helped them engage more fully during the ECGT groups.

Physical Presence
     Study participants shared that group members’ physical presence during the face-to-face sessions enhanced their willingness to participate. The physical presence provided access and a better ability to understand group members’ content and emotion through their body language, eye contact, vocal tone, and other nonverbal cues during sessions. As one participant shared, “I feel so much more in touch and present with people when I can see them, but just kind of feel their physical presence rather than just watching the faces online.” Furthermore, the study participants shared that being physically present during the face-to-face sessions allowed for the incorporation of more icebreaker activities by both doctoral and master’s student group leaders, enhancing their participation in groups. One participant noted that “the small icebreakers, I just remember doing those at the beginning during our face-to-face sessions; those were a lot of fun.”

Comfortability of Online Sessions
     Participants reported that they felt comfortable engaging in online EGCT from their familiar surroundings at home. They appreciated the convenience of participating in ECGT groups from wherever they were. One participant reported that “people could be outside or eating or drinking or whatever, which I think is cool.” Another participant shared that before the state-issued quarantine, they already used online technology to communicate with friends, so it was easy to use Microsoft Teams for online experiential training groups. Another participant noted:

We were doing them (EGCT) from the comfort of our own home; it just increased how comfortable you were in general. We were all at home, rocking in sweatpants and not having to worry about stuff. I feel we were in our own comfortable, safe space, and that made the online easier for me.

Cohesiveness
     Participants reported they felt “anxious,” “lonely,” and “isolated” and experienced other difficulties during the COVID-19 pandemic. They noted that they actively engaged in online EGCT sessions because it provided them with the opportunity to connect, share, and process their thoughts and emotions. A group participant reported, “We all had to isolate. [It] made it exciting to be able to connect with everyone again, to talk about how it (COVID-19) was affecting us, to vent out our emotions and check in with others.” Additionally, another participant reported:

When we started these sessions [online], it was at the beginning of these COVID-19 issues, and I was feeling more stressful, and there was nothing to do. It was so difficult to adjust to this environment, even staying at home. This was like an opportunity for me to connect with classmates in the group and [it] helped me to reflect on my anxiety and how other people were thinking around these COVID-19 issues.

     As a result of the online EGCT groups, participants gained a means of personal interaction during isolation. The subthemes presented above capture the positive participation factors that helped participants to engage actively in both online and face-to-face sessions.

Leadership Interventions
     Participants shared leadership interventions that helped them to participate during face-to-face and online sessions. The sudden transition to online groups due to COVID-19 was characterized by trial and error and uncertainty for everyone. Participants noted that while working with the new online EGCT group and different processes than what they experienced before COVID-19, doctoral students and master’s student leaders demonstrated a sense of flexibility and adaptability to the prevailing situation and could steer the groups in the changing environment. Both the doctoral and master’s student leaders were aware of the effect of COVID-19 on the participants, and they allowed the participants to get support from each other before they could get into the session plan for the group. One participant mentioned that “we kind of partly used that [the group] as a social support group . . . and reflect on how we’re feeling during social isolation.” Another participant shared that “the facilitators were flexible. So, even if they had a topic or something like that, they would allow for flexibility, to check in [with participants], and be able to kind of shift focus to what we all needed.”

Participants explicitly mentioned that the doctoral and master’s students’ leadership interventions, such as encouraging, checking in, and being present, helped them engage in the EGCT groups. Participants highlighted the strength of the group leaders’ encouragement of reflection (“I appreciated that the leader really put emphasis on encouraging us to answer questions”) and overall presence and attention (“[The leader] was attending our behavior and was really good with reflecting”). The participants also found the aspect of “checking in” by the leaders as something that enhanced their participation: “The leaders were always pretty quick to check in on someone if something seemed off.”

Group leaders’ ability to coordinate and successfully facilitate group sessions can significantly influence group outcomes (G. Corey, 2016; Gladding, 2012). Study participants shared that group facilitators demonstrated leadership skills and techniques to facilitate meaningful discussions and participation among members in both face-to-face and online sessions: “Like she [group leader] was always there to answer questions if there is silence; like she didn’t want us to rely on her to do the entire conversation, so her encouragement was beneficial for me.”

Participation-Inhibiting Attributes
     For this main theme, we examined attributes that negatively influenced participation and leading in the online and face-to-face formats of the EGCT groups. Three subthemes were identified: (a) group dynamics, (b) challenges with online EGCT, and (c) technological obstacles for online EGCT. The most prominent subtheme that arose and spread across both group formats was that of the group dynamic. Friction within the group dynamic was one of the primary issues reported by participants. The remaining subthemes were related to challenges with online EGCT groups. These challenges include the importance of “being with” or physically present with the rest of the group, problems with missing nonverbal communication in the online meetings, difficulties navigating awkward silences and pauses in the group, and technical obstacles.

Group Dynamics
     Study participants shared that the group dynamics dictated how much of a connection developed among group members and significantly influenced the progression to the working phase in the groups. In the words of one participant, “I feel like that was definitely something with our group dynamic. . . . There was definitely still good conversations, but I think that impacted it.”

Some participants reported their initial concerns about fostering rapport with group mates chosen randomly for them. Participants expressed thoughts that personalities did not mesh well in their group and that there were issues of building good rapport. Some participants indicated that having a reserved personality made it hard to participate: “For me, it was more about a personal thing because I am an introverted personality, so I find it difficult to talk in groups anyway, so that’s what hindered my participation sometimes.” Another participant stated: “I felt like the others protect themselves by not talking, so why should I open myself and put myself into risk? I thought about that.”

Challenges With Online EGCT
     Participants in this study emphasized that one of the main difficulties of the online EGCT experience that affected their participation and leadership negatively was missing body language and physical cues. Participants shared that they could use nonverbal cues and body language to know when it was a good time to speak without interrupting other group members during the face-to-face ECGT. Because these were missing in online EGCT, the students did not have immediate awareness to participate in group conversation without interrupting other group members. For example, one participant noted the difficulties of “just not being able to read body language as well and not being able to see everyone at once.” As a result of these online environment limitations, study participants indicated they had a sense of “stepping on toes” while trying to participate in online EGCT: “I think that one of the biggest challenges with doing it [EGCT] online is that you want to be respectful and make sure that you are not gonna talk over somebody else.”

Kozlowski and Holmes (2014) previously noted that the unfamiliar environment of online counseling, the time delay because of technology, and the inability to utilize group members’ body language can all create a one-dimensional or “linear” experience in online group counseling environments. These factors appeared to hinder the natural growth and development of the EGCT groups in our study as well. In an effort to reduce the perception of being rude, there were times of awkward silence as participants avoided constant interruptions during the sessions; this difficulty gave the feeling of a linear environment.

One other factor the participants noted in the online format more so than the in-person group was what students described as an awkward silence. This occurrence serves as a subtheme of missed physical cues because the participants noted that the lack of said cues complicated determining when to speak and when to wait: “Online, the silence almost felt like it was much longer than what it really would have been if it was face-to-face.” Another participant stated that they “feel pretty comfortable with silences, but it’s a lot harder to gauge that when it’s online.” This issue presented itself in several circumstances, though one group did attempt to figure out a solution, per the report of one participant: “For our group . . . to help with people talking over each other, we had people type in a smiley face in the chat when they wanted to share.”

Notably, participants in this study also mentioned that there was some physical presence that they could not describe but found to be relevant to them in their connection with the group. Although students were unable to identify it precisely, several study participants agreed on its importance. One participant said that they “enjoy the voice and the video, but I feel like when we are talking, especially in a group dynamic and group processes, especially to grasp something important, I really need to be with this person in a physical space.”

The participants emphasized the importance of physical presence, from the ability to see and greet one another to having space to do activities that got them up and moving. Many participants mentioned some intangible quality they could not name but that was missing when the groups convened electronically instead of in person. A participant shared that “you can observe the body language—what is happening in the group actually, but in online sessions, it’s like you don’t know, you are just talking.”

As noted in other sections, the group members appreciated the space for doing activities together when they were in person. Master’s student group leaders reported that they felt anxious when facilitating icebreaker activities in their online EGCT sessions because of the missing physical presence and noted the loss of face-to-face icebreakers. Study participants lamented that the online format did not allow for these bonding and icebreaking exercises, which when utilized in the usual face-to face format tended to put them in a position to feel better equipped to share with their group members, almost like a metaphorical entryway to the group process: “Some of the exercises are not possible to execute [online] because we were doing some physical things in our group, like throwing balls to each other and stuff.” Without these social warm-ups, the group flow and process suffered; according to those in the focus group, leaders needed more assistance to run activities in online EGCT sessions. One participant added a similar sentiment: “How do we lead a group online with proximity activities or icebreakers we would use? We can’t really do [that] because of the virtual interaction, [it] can’t work.”

Overall, the online EGCT environment limited the interpersonal relationships of the EGCT members and group leaders. Group members could not use their nonverbal communication skills or participate in physical group activities. Lastly, online EGCT appears to provide added pressure on group leaders to keep members engaged during the session. Master’s students had to choose topics where all members felt comfortable enough to participate with minimal encouragement, which was a challenge.

Technological Obstacles for Online EGCT
     Participants reported some technological difficulties that inhibited their ability to participate and lead the online EGCT sessions. Some participants noted that when participants turned off their cameras, it exacerbated disengagement levels within the group and hampered group dynamics. Some speculated that technical difficulties might be an excuse to disengage from the group: “Like in online, I can be mute, I can turn off my camera, I can not talk, and I can accuse the technology for that.” This capacity to disengage negatively impacted the group for several of the focus group participants, who noted that they felt this closed off the group and circumvented the ability to engage with all members of the group.

The limitations of the university-sanctioned online platform used for the EGCT groups, Microsoft Teams, adversely affected engagement during the online sessions as it only allowed four members (at the time of the online EGCT sessions) to be seen on the screen at a time. As one participant stated, “I cannot see all the group members . . . my attention is not with all members. This was difficult. It was difficult to lead the group.” Several group members were vociferous in their dislike for this limitation of the platform. Further, internet connectivity issues were problematic: “Sometimes like a group member would disconnect [because of technology problems], and there would be several minutes before they could come back.” These types of interruptions were frustrating to all group members and group leaders. Master’s student group leaders had a difficult time leading with interruptions.

One focus group participant noted, and others agreed, that it was challenging to learn how to lead a group online because they were missing so many elements of the in-person process of leading a group, and they did not have previous group leadership experience in an online environment. A participant shared that “it’s hard [leading group online]. It’s maybe harder for leaders because they cannot observe what’s going on . . . like body language.” 

Suggestions for Group Counseling Training
     Participants were invited to share their concerns and ways to develop and improve face-to-face and online EGCT group experiences. Three subthemes were identified: (a) software issues and training, (b) identified group topics, and (c) preferred EGCT environment.

Software Issues and Training
     Participants shared common concerns about the software for their online experiential training groups. Specifically, they found Microsoft Teams’ display of only four people at one time prevented them from seeing all group members on the screen. Members who were not speaking were displayed at the bottom of the computer screen with their profile picture or initials, which was not conducive to interaction. One participant suggested that they should “probably just use Zoom instead . . . I like Zoom better, seriously, because I can see absolutely everyone.” Another participant agreed, “But for the reason, at least, in Zoom, I can see everyone’s faces, not, um, not just four.”

Another participant similarly emphasized the importance of seeing everyone on one screen during their meeting: “If you don’t see the faces [at one time], you’re just clueless. I mean, have to, like, awkwardly check in with this person all the time.” Participants also brought another suggestion about training on leading online experiential training groups. Participants shared their anxiety about leading groups using online software because it is a new and unique experience. Because of the sudden onset of COVID-19, the students did not have a chance to get training on how to lead online experiential training groups. A participant mentioned that having training where students could learn how to facilitate online groups before leading weekly sessions would help alleviate anxiety and build competence: “Perhaps allowing a small period where everyone kind of gets adjusted to it and becomes more familiar with it might help facilitate [online] group sessions better.”

Identified Group Topics
     Another suggestion by participants regarding their EGCT experience was using one selected topic for each group. For example, a participant shared: “I think part of what was hard about this that might be something to change is, like having the group just be all over the place in terms of topics from week to week.” Another participant added: “If the group was more, like, a little bit more specific and clearer about like, the goal, or something like that, that might be—might help it flow a little bit better.” Some participants also suggested allowing students to select which group they wanted to attend, instead of having groups pre-assigned to them. In other words, participants preferred to join a specific group based on their interests. A participant mentioned: “I think that would be like a really good option to give like a list of ten types of groups or topics in the groups.” Another participant similarly suggested “giving an opportunity to all students to choose one group. For example, like the one group would work specifically on self-esteem problems or the other one would work on grief problems.”

Some participants noted that they felt there was a lack of purpose for the group, indicating that they were not sure of the group’s goals or objectives and that this hindered their ability to participate fully. Some also shared having confusion about their role and the boundaries of the group and what they could or could not share. One participant noted: “In the first session when we were trying to set up our goals, it was difficult for us to find what the goals will be as a group leader candidate, or as a person.” The focus group participants suggested giving more concrete topics overall for the EGCT group to understand better how to participate. This notion spanned across the online and face-to-face format as a more general recommendation.

Preferred Training Environment
     Lastly, participants were asked about their preference for participation in a face-to-face or online EGCT experience, if given a choice. Even though participants reported a reasonably good experience with online EGCT groups, such as comfortability and cohesiveness, most of the participants voiced a preference for face-to-face sessions if they had to do the group counseling training over again. One participant stated: “Ultimately, face-to-face will probably still be better.” Another participant added: “Face-to-face for sure. I just think as like a profession, we all enjoy working with people. We would prefer to work with someone in person.” Similarly, another participant mentioned: “I would definitely choose face-to-face, but I was thankful that we had the opportunity to do it online.”

Asking the participants about their preferred experiential training group environment garnered the most reaction during the interviews. Most of the participants shared that they preferred face-to-face groups. Even though participants had personal connections in an online setting, they wanted to have face-to-face meetings to interact better. One participant mentioned that “we are doing online sessions right now. I wish that I [could] continue to do the group lab and connect with the group members, but if I have the opportunity to take face-to-face, absolutely, I would do that.” Lastly, another participant added: “Absolutely, it’s face-to-face, but if we are in a situation like this, COVID-19 issues, sometimes the online sessions can be helpful.”

Participants offered their perspectives on learning group counseling skills during the global COVID-19 pandemic. Despite the unprecedented circumstances, the students persevered and completed the course. Group leaders and professors encouraged the group members to participate to the best of their abilities. The concerns and suggestions shared in these focus groups could help counselor educators plan and develop for EGCT in both online and face-to-face formats.

Discussion

This study investigated the experiences of master’s students in an online and face-to-face EGCT group. EGCT is an essential aspect of novice counselors’ preparation and is required by CACREP (2015) standards. In this study, participants identified positive factors related to their EGCT group participation, such as knowing other group members, group leadership skills, physical presence, and connection with other group members. They also reported participation-inhibiting factors such as the complexities of group dynamics, missing physical cues, and technological challenges. Our research findings are similar to Kozlowski and Holmes’s (2014) study on online group counseling training. Their participants reported problems with the group feeling artificial, lacking attending skills, and difficulties with achieving cohesion and connectedness.

In the current study, course instructors and student leaders did not have control over the choice of an online platform. The limitations of Microsoft Teams, which at the time of the online EGCT sessions only allowed four participants to be visible on the screen at one time, added to difficulties with engaging and feeling connected. For participants to remain engaged, leaders and instructors should have access to online platforms that allow students to see all group members simultaneously on the screen. Setting ground rules requiring that cameras remain on during sessions and utilizing the chat feature or the hand-raising feature to facilitate discussions would also help create and maintain a sense of connection. Outlining contingency plans such as the alternatives for not being able to join the group with the camera on are important for successful group outcomes.

Participants in this study appreciated the convenience of participating in online ECGT groups. This is similar to the findings of King et al. (2009) about the convenience of access to online group counseling. In the same study by King et al. (2009), the participants shared that online counseling sessions allowed them to participate from the comfort of their homes, thus improving both convenience and privacy. One of the difficulties participants reported was that of awkward silence. This experience, coupled with interruptions (“stepping on toes”), resulted in students finding that the experience online was more linear and less organic compared to face-to-face interactions. These findings are similar to those of Kozlowski and Holmes (2014). Yalom and Leszcz (2005) noted that the group leader’s role is to design the group’s path, get it going, and keep it functional to achieve effectiveness. Presence, self-confidence, the courage to take risks, belief in the group process, inventiveness, and creativity are essential leadership traits in leading groups (G. Corey, 2016). However, these traits are for in-person groups. It is possible that effectively leading online groups requires other skills that have not yet been identified. The sudden change to online training in this instance did not allow for a planful design. It is necessary for group leaders to possess specific group leadership skills and appropriately perform them to help group members participate in groups (M. S. Corey et al., 2018). However, participants appreciated that the doctoral and master’s student leaders demonstrated flexibility, allowing for additional time to check in with group members and process their experiences and emotions related to the pandemic.

One interesting finding related to how COVID-19 impacted participants’ experiences in the ECGT groups was that group participants actively engaged in the online sessions when they were allowed to process their anxiety and stress due to COVID-19, as it served as a support group. This result is dissimilar to findings of previous studies in which participants felt unsafe during online group sessions and being on online platforms impeded participants’ emotional connection and trust levels (Fletcher-Tomenius & Vossler, 2009; Haberstroh et al., 2007; Kozlowski & Holmes, 2014).

Bellafiore et al. (2003) emphasized online group leaders’ roles as “shaping the group” and “setting the tone.” They also expressed that “establishing and maintaining a leadership style is important in keeping the group going” (p. 211). In the current study, first-year master’s students, many of whom were participating in or leading groups for the first time, had the unexpected and sudden additional layer of learning how to lead online. Further, the abrupt transition from face-to-face to online groups because of COVID-19 did not allow for extensive instructor planning and preparation. Leading groups online was challenging and anxiety-provoking for members, as they lacked experience and were unsure how to proceed. Master’s students need additional training on facilitating online groups, establishing a leadership style, and managing silence. This information corresponds with Cárdenas et al.’s (2008) findings that master’s-level counseling students felt more confident to provide online counseling services after training.

Implications

Although the findings from this study are not generalizable, there may be implications for designing and leading EGCT groups that merit consideration based on the experience of the counselor trainees described in this study. Part of the group design entailed assigning different topics to focus on for each session. The rationale for having different topics for each session should be clearly explained to the participants. Any questions regarding the identified topics should be addressed early to enhance the group facilitation process for both leaders and participants. Additionally, group leaders or course instructors need to explain roles clearly, and group members should understand the group’s boundaries and how they fit with their didactic course.

With online EGCT groups, it is essential to consider how participation is influenced by a lack of natural communication signals, such as body language and physical presence. Counselor educators and EGCT student leaders need to establish ground rules about online group interactions such as having all cameras remain on during sessions, having a private and quiet space from which to participate, and minimizing distractions from pets or relatives, all of which are necessary for successful groups. Further, utilizing technology that allows all members to be seen on the screen may help build connection and cohesiveness. Utilizing methods such as using the chat to insert a symbol or using the hand-raising icon can also help facilitate participation.

Overall, students reported feeling unprepared to lead online counseling groups. However, as counselor educators, we are responsible for preparing our students to engage in online counseling successfully, especially as the COVID-19 pandemic continues into its second year and will continue to affect how much virtual counseling will take place in the future. The recent normalization of online counseling (individual and group) may persuade educators and counselors to “increase their skills in terms of development, comfortability, and flexibility in the online environment” (International OCD Foundation, 2020, p. 1). Therefore, counselor educators should cover online-specific facilitation skills in their training programs.

Limitations and Future Research Directions

This study was the first step in attempting to understand and describe master’s-level students’ experiences of participating and leading in both face-to-face and online formats of EGCT. As with all research, limitations should be considered in interpreting the findings. Further, some of the limitations point to potential research directions.

COVID-19 created a situation where the transition from face-to-face to online formats was compulsory. It is therefore not clear what the experience would have been like if the transition was planned and did not have a situation like COVID-19 in the background pushing the transition, or if the group had been entirely online. Because of unplanned adjustment, course instructors and student leaders did not have control over the choice of an online platform. Outlining contingency plans, such as alternatives when a group member cannot join the group with their camera on, are essential for successful group outcomes, and a lack of familiarity with online platforms may have prevented instructors and student leaders from providing these contingencies and therefore impacted the experience for students.

Further, the EGCT groups were conducted with master’s-level students, and participants already had preexisting relationships with each other. This may have contributed to their strong support of face-to-face groups over online groups. In future research, studies with participants who do not already know each other may help us assess the appeal of online groups to participants. Further, researchers in the future may wish to examine the efficacy of online group counseling training for counseling students compared to in-person group training by comparing two equivalent experiential groups.

The current study recruited master’s-level counseling students from a CACREP-accredited counseling program in the Midwest United States; thus, results cannot be generalized to other institutions. The sample size was small in the current study. Therefore, we caution against generalizing our findings. During the focus groups, participants shared some apprehension about how much information to disclose in group counseling, and they verbalized some confusion on group purpose, direction, or goals. For many, these EGCT groups were the students’ first experience in group counseling training, and this could contribute to them questioning if their feelings and experiences were appropriate (Ohrt et al., 2014).

There are methodological considerations to improve future studies. Focus groups were conducted to collect the data from the participants. In-depth individual interviews would enhance a deeper conversation in understating and reflecting on the challenges and needs of master’s-level students. Participants may have censored some of their true feelings, as they were aware that their group leaders were also part of the research team, even though they did not run the focus groups. We acknowledge that the students knowing each other from previous classes may have influenced how much they shared in groups. Participants in this study expressed comfort with knowing each other from a previous semester. However, it is also possible that students may have disclosed minimal personal information so as not to effect public perception of themselves or effect future professional relationships.

Another area to expand on would be investigating counselors’ self-efficacy while facilitating online counseling groups. For example, exploring positive participation attributes that increase online groups’ participation from the leader’s perspective could be useful. This may allow researchers and practitioners to identify how group counseling can best be leveraged in an online environment.

Conclusion

The purpose of this study was to explore and compare first-year master’s-level counseling students’ experiences of participating and leading in both face-to-face and online formats of EGCT. In summary, students considered that the online format was challenging because it added a layer of learning to their fledgling group work skills beyond the face-to-face setting. Technological barriers that were outside the control of participants inhibited their participation, but on the other hand, the online groups served as a safe and supportive space for students to alleviate their stress and loneliness due to COVID-19. Regardless of the teaching environment, thoughtful and well-planned EGCT groups are essential for student development in this area, and skilled group leaders can manage group dynamics and model group counseling skills. COVID-19 has necessitated a focus on teletherapy and online counseling. The group counseling profession should be proactive in addressing this training need, as conducting online group counseling sessions is likely to continue to be a much-needed skill in a post-pandemic world.

 

Conflict of Interest and Funding Disclosure
The authors reported no conflict of interest
or funding contributions for the development
of this manuscript.

 

References

American Counseling Association. (2014). ACA code of ethics. https://www.counseling.org/docs/default-source/default-document-library/2014-code-of-ethics-finaladdress.pdf?sfvrsn=96b532c_2

Amulya, D. S. L. (2020). An experiment with online group counseling during COVID 19. In L. S. S. Manickam (Ed.), COVID-19 pandemic: Challenges and responses of psychologists from India (pp. 182–197).

Anthony, K. (2015). Training therapists to work effectively online and offline within digital culture. British Journal of Guidance & Counselling, 43(1), 36–42. https://doi.org/10.1080/03069885.2014.924617

Association for Counselor Education and Supervision. (2018). ACES guidelines for online learning – 2017. https://acesonline.net/knowledge-base/aces-guidelines-for-online-learning-2017-2

Association for Specialists in Group Work. (2007). Association for Specialists in Group Work: Best practice guidelines. https://www.researchgate.net/publication/247784312_Association_for_Specialists_in_Group_Work_Best_Practice_Guidelines_2007_Revisions

Barak, A., & Grohol, J. M. (2011). Current and future trends in internet-supported mental health interventions. Journal of Technology in Human Services, 29(3),155–196. https://doi.org/10.1080/15228835.2011.616939

Behenck, A., Wesner, A. C., Finkler, D., & Heldt, E. (2017). Contribution of group therapeutic factors to the outcome of cognitive–behavioral therapy for patients with panic disorder. Archives of Psychiatric Nursing, 31(2), 142–146. https://doi.org/10.1016/j.apnu.2016.09.001

Békés, V., & Aafjes-van Doorn, K. (2020). Psychotherapists’ attitudes toward online therapy during the COVID-19 pandemic. Journal of Psychotherapy Integration, 30(2), 238–247. https://doi.org/10.1037/int0000214

Bellafiore, D. R., Colon, Y., & Rosenberg, P. (2003). Online counseling groups. In R. Kraus, J. Zack, & G. Stricker (Eds.), Online counseling: A handbook for mental health professionals (pp. 197–216). Academic Press.

Bowen, G. A. (2008). Naturalistic inquiry and the saturation concept: A research note. Qualitative Research, 8(1), 137–152. https://doi.org/10.1177/1468794107085301

Burlingame, G. M., McClendon, D. T., & Yang, C. (2019). Cohesion in group therapy. In J. C. Norcross & M. J. Lambert (Eds.), Psychotherapy relationships that work: Evidence-based therapist contributions (pp. 205–244). Oxford University Press.

Cárdenas, G., Serrano, B., Flores, L. A., & De la Rosa, A. (2008). Etherapy: A training program for development of clinical skills in distance psychotherapy. Journal of Technology in Human Services, 26(2–4), 470–483. https://doi.org/10.1080/15228830802102180

Chan, Z. C., Fung, Y., & Chien, W. T. (2013). Bracketing in phenomenology: Only undertaken in the data collection and analysis process. The Qualitative Report, 18(30), 1–9.
https://doi.org/10.46743/2160-3715/2013.1486

Corey, G. (2016). Theory and practice of group counseling (9th ed.). Cengage.

Corey, M. S., Corey, G., & Corey, C. (2018). Groups: Process and practice (10th ed.). Cengage.

Council for the Accreditation of Counseling and Related Educational Programs. (2015). CACREP 2016 standards. http://www.cacrep.org/wp-content/uploads/2017/08/2016-Standards-with-citations.pdf

Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE.

Darcy, A. M., & Dooley, B. (2007). A clinical profile of participants in an online support group. European Eating Disorders Review, 15(3), 185–195. https://doi.org/10.1002/erv.775

Fletcher-Tomenius, L., & Vossler, A. (2009). Trust in online therapeutic relationships: The therapist’s experience. Counselling Psychology Review, 24(2), 24–33.

Freeman, E., Barker, C., & Pistrang, N. (2008). Outcome of an online mutual support group for college students with psychological problems. Cyberpsychology & Behavior, 11(5), 591–593.
https://doi.org/10.1089/cpb.2007.0133

Gilkey, S. L., Carey, J., & Wade, S. L. (2009). Families in crisis: Considerations for the use of web-based treatment models in family therapy. Families in Society, 90(1), 37–45. https://doi.org/10.1606/1044-3894.3843

Gladding, S. T. (2012). Groups: A counseling specialty (6th ed.). Pearson.

Haberstroh, S., Duffey, T., Evans, M. P., Gee, R., & Trepal, H. (2007). The experience of online counseling. Journal of Mental Health Counseling, 29(3), 269–282. https://doi.org/10.17744/mehc.29.3.j344651261w357v2

Haberstroh, S., & Moyer, M. (2012). Exploring an online self-injury support group: Perspectives from group members. The Journal for Specialists in Group Work, 37(2), 113–132.
https://doi.org/10.1080/01933922.2011.646088

Hearn, C. S., Donovan, C. L., Spence, S. H., & March, S. (2017). A worrying trend in social anxiety: To what degree are worry and its cognitive factors associated with youth social anxiety disorder? Journal of Affective Disorders, 208, 33–40. https://doi.org/10.1016/j.jad.2016.09.052

Holmes, C. M., & Kozlowski, K. A. (2015). A preliminary comparison of online and face-to-face process groups. Journal of Technology in Human Services, 33(3), 241–262. https://doi.org/10.1080/15228835.2015.1038376

Ieva, K. P., Ohrt, J. H., Swank, J. M., & Young, T. (2009). The impact of experiential groups on master students’ counselor and personal development: A qualitative investigation. The Journal for Specialists in Group Work, 3(4), 351–368. https://doi.org/10.1080/01933920903219078

International OCD Foundation. (2020, July 15). Teletherapy in the time of COVID-19. https://iocdf.org/covid19/teletherapy-in-the-time-of-covid-19

King, V. L., Stoller, K. B., Kidorf, M., Kindbom, K., Hursh, S., Brady, T., & Brooner, R. K. (2009). Assessing the effectiveness of an Internet-based videoconferencing platform for delivering intensified substance abuse counseling. Journal of Substance Abuse Treatment, 36(3), 331–338.

https://doi.org/10.1016/j.jsat.2008.06.011

Kit, P. L., Wong, S. S., D’Rozario, V., & Teo, C. T. (2014). Exploratory findings on novice group counselors’ initial co-facilitating experiences in in-class support groups with adjunct online support groups. The Journal for Specialists in Group Work, 39(4), 316–344. https://doi.org/10.1080/01933922.2014.954737

Kiweewa, J., Gilbride, D., Luke, M., & Seward, D. (2013). Endorsement of growth factors in experiential training groups. The Journal for Specialists in Group Work, 38(1), 68–93.
https://doi.org/10.1080/01933922.2012.745914

Kozlowski, K. A., & Holmes, C. M. (2014). Experiences in online process groups: A qualitative study. The Journal for Specialists in Group Work, 39(4), 276–300. https://doi.org/10.1080/01933922.2014.948235

Kozlowski, K. A., & Holmes, C. M. (2017). Teaching online group counseling skills in an on-campus group counseling course. Journal of Counselor Preparation and Supervision, 9(1).

Krueger, R. A., & Casey, M. (2014). Focus groups: A practical guide for applied research (5th ed.). SAGE.

Krug, O. T. (2009). James Bugental and Irvin Yalom: Two masters of existential therapy cultivate presence in the therapeutic encounter. Journal of Humanistic Psychology, 49(3), 329–354.
https://doi.org/10.1177/0022167809334001

Lieberman, M., Winzelberg, A., Golant, M., Wakahiro, M., DiMinno, M., Aminoff, M., & Christine, C. (2010). Online support groups for Parkinson’s patients: A pilot study of effectiveness. Social Work Health Care, 42(2), 23–38. https://doi.org/10.1300/J010v42n02_02

Lopresti, J. M. (2010). The process and experience of online group counseling for masters-level counseling students (Order No. 3451084). Available from ProQuest Dissertations & Theses A&I. (862058819).

Moustakas, C. (1994). Phenomenological research methods. SAGE.

Murdock, J., Williams, A., Becker, K., Bruce, M. A., & Young, S. (2012). Online versus on-campus: A comparison study of counseling skills courses. The Journal of Human Resource and Adult Learning, 8(1), 105–118.

Ohrt, J. H., Prochenko, Y., Stulmaker, H., Huffman, D., Fernando, D., & Swan, K. (2014). An exploration of group and member development in experiential groups. The Journal for Specialists in Group Work, 39(3), 212–235. https://doi.org/10.1080/01933922.2014.919047

Patton, M. Q. (2015). Qualitative research & evaluation methods: Integrating theory and practice (4th ed.). SAGE.

Peng, D., Wang, Z., & Xu, Y. (2020). Challenges and opportunities in mental health services during the COVID-19 pandemic. General Psychiatry, 33(5). https://doi.org/10.1136/gpsych-2020-100275

Richards, D., & Viganó, N. (2013). Online counseling: A narrative and critical review of the literature. Journal of Clinical Psychology, 69(9), 994–1011. https://doi.org/10.1002/jclp.21974

Shumaker, D., Ortiz, C., & Brenninkmeyer, L. (2011). Revisiting experiential group training in counselor education: A survey of master’s-level programs. The Journal for Specialists in Group Work, 36(2), 111–128. https://doi.org/10.1080/01933922.2011.562742

Steen, S., Vasserman-Stokes, E., & Vannatta, R. (2014). Group cohesion in experiential growth groups. The Journal for Specialists in Group Work, 39(3), 236–256. https://doi.org/10.1080/01933922.2014.924343

Webb, M., Burns, J., & Collin, P. (2008). Providing online support for young people with mental health difficulties: Challenges and opportunities explored. Early Intervention in Psychiatry, 2(2), 108–113. https://doi.org/10.1111/j.1751-7893.2008.00066.x

Weinberg, H. (2020). Online group psychotherapy: Challenges and possibilities during COVID-19—A practice review. Group Dynamics: Theory, Research, and Practice, 24(3), 201–211.
https://doi.org/10.1037/gdn0000140

Wind, T. R., Rijkeboer, M., Andersson, G., & Riper, H. (2020). The COVID-19 pandemic: The ‘black swan’ for mental health care and a turning point for e-health. Internet Interventions, 20.
https://doi.org/10.1016/j.invent.2020.100317

Yalom, I. D., & Leszcz, M. (2005). The theory and practice of group psychotherapy (5th ed.). Basic
Books.

 

Bilal Urkmez, PhD, LPC, CRC, is an assistant professor at Ohio University. Chanda Pinkney, MA, CT, is a doctoral student at Ohio University. Daniel Bonnah Amparbeng, MEd, NCC, LPC, is a doctoral student at Ohio University. Nanang Gunawan, MA, is a doctoral student at Ohio University. Jennifer Ojiambo Isiko, MA, is a doctoral student at Ohio University. Brandon Tomlinson, MA, NCC, LPC, is a doctoral student at Ohio University. Christine Suniti Bhat, PhD, LPC, LSC, is a professor at Ohio University. Correspondence may be addressed to Bilal Urkmez, Patton Hall 432P, Athens, OH 45701, urkmezbi@ohio.edu.

Enhancing Assessment Literacy in Professional Counseling: A Practical Overview of Factor Analysis

Michael T. Kalkbrenner

Assessment literacy is an essential competency area for professional counselors who administer tests and interpret the results of participants’ scores. Using factor analysis to demonstrate internal structure validity of test scores is a key element of assessment literacy. The underuse of psychometrically sound instrumentation in professional counseling is alarming, as a careful review and critique of the internal structure of test scores is vital for ensuring the integrity of clients’ results. A professional counselor’s utilization of instrumentation without evidence of the internal structure validity of scores can have a number of negative consequences for their clients, including misdiagnoses and inappropriate treatment planning. The extant literature includes a series of articles on the major types and extensions of factor analysis, including exploratory factor analysis, confirmatory factor analysis (CFA), higher-order CFA, and multiple-group CFA. However, reading multiple psychometric articles can be overwhelming for professional counselors who are looking for comparative guidelines to evaluate the validity evidence of scores on instruments before administering them to clients. This article provides an overview for the layperson of the major types and extensions of factor analysis and can serve as reference for professional counselors who work in clinical, research, and educational settings.

Keywords: Factor analysis, overview, professional counseling, internal structure, validity

Professional counselors have a duty to ensure the veracity of tests before interpreting the results of clients’ scores because clients rely on their counselors to administer and interpret the results of tests that accurately represent their lived experience (American Educational Research Association [AERA] et al., 2014; National Board for Certified Counselors [NBCC], 2016). Internal structure validity of test scores is a key assessment literacy area and involves the extent to which the test items cluster together and represent the intended construct of measurement.

Factor analysis is a method for testing the internal structure of scores on instruments in professional counseling (Kalkbrenner, 2021b; Mvududu & Sink, 2013). The rigor of quantitative research, including psychometrics, has been identified as a weakness of the discipline, and instrumentation with sound psychometric evidence is underutilized by professional counselors (Castillo, 2020; C.-C. Chen et al., 2020; Mvududu & Sink, 2013; Tate et al., 2014). As a result, there is an imperative need for assessment literacy resources in the professional counseling literature, as assessment literacy is a critical competency for professional counselors who work in clinical, research, and educational settings alike.

Assessment Literacy in Professional Counseling
Assessment literacy is a crucial proficiency area for professional counselors, as counselors in a variety of the specialty areas of the Council for Accreditation of Counseling and Related Educational Programs (2015), such as clinical rehabilitation (5.D.1.g. & 5.D.3.a.), clinical mental health (5.C.1.e. & 5.C.3.a.), and addiction (5.A.1.f. & 5.A.3.a.), select and administer tests to clients and use the results to inform diagnosis and treatment planning, and to evaluate the utility of clinical interventions (Mvududu & Sink, 2013; NBCC, 2016; Neukrug & Fawcett, 2015). The extant literature includes a series of articles on factor analysis, including exploratory factor analysis (EFA; Watson, 2017), confirmatory factor analysis (CFA; Lewis, 2017), higher-order CFA (Credé & Harms, 2015), and multiple-group CFA (Dimitrov, 2010). However, reading several articles on factor analysis is likely to overwhelm professional counselors who are looking for a desk reference and/or comparative guidelines to evaluate the validity evidence of scores on instruments before administering them to clients. To these ends, professional counselors need a single resource (“one-stop shop”) that provides a brief and practical overview of factor analysis. The primary purpose of this manuscript is to provide an overview for the layperson of the major types and extensions of factor analysis that counselors can use as a desk reference.

Construct Validity and Internal Structure

     Construct validity, the degree to which a test measures its intended theoretical trait, is a foundation of assessment literacy for demonstrating validity evidence of test scores (Bandalos & Finney, 2019). Internal structure validity, more specifically, is an essential aspect of construct validity and assessment literacy. Internal structure validity is vital for determining the extent to which items on a test combine to represent the construct of measurement (Bandalos & Finney, 2019). Factor analysis is a key method for testing the internal structure of scores on instruments in professional counseling as well as in social sciences research in general (Bandalos & Finney, 2019; Kalkbrenner, 2021b; Mvududu & Sink, 2013). In the following sections, I will provide a practical overview of the two primary methodologies of factor analysis (EFA and CFA) as well as the two main extensions of CFA (higher-order CFA and multiple-group CFA). These factor analytic techniques are particularly important elements of assessment literacy for professional counselors, as they are among the most common psychometric analyses used to validate scores on psychological screening tools (Kalkbrenner, 2021b). Readers might find it helpful to refer to Figure 1 before reading further to become familiar with some common psychometric terms that are discussed in this article and terms that also tend to appear in the measurement literature.

Figure 1

Technical and Layperson’s Definitions of Common Psychometric Terms
Note. Italicized terms are defined in this figure.

Exploratory Factor Analysis
EFA is “exploratory” in that the analysis reveals how, if at all, test items band together to form factors or subscales (Mvududu & Sink, 2013; Watson, 2017). EFA has utility for testing the factor structure (i.e., how the test items group together to form one or more scales) for newly developed or untested instruments. When evaluating the rigor of EFA in an existing psychometric study or conducting an EFA firsthand, counselors should consider sample size, assumption checking, preliminary testing, factor extraction, factor retention, factor rotation, and naming rotated factors (see Figure 2).

EFA: Sample Size, Assumption Checking, and Preliminary Testing
     Researchers should carefully select the minimum sample size for EFA before initiating data collection (Mvududu & Sink, 2013). My 2021 study (Kalkbrenner, 2021b) recommended that the minimal a priori sample size for EFA include either a subjects-to-variables ratio (STV) of 10:1 (at least 10 participants for each test item) or 200 participants, whichever produces a larger sample. EFA tends to be robust to moderate violations of normality; however, results are enriched if data are normally distributed (Mvududu & Sink, 2013). A review of skewness and kurtosis values is one way to test for univariate normality; according to Dimitrov (2012), extreme deviations from normality include skewness values > ±2 and kurtosis > ±7; however, ideally these values are ≤ ±1 (Mvududu & Sink, 2013). The Shapiro-Wilk and Kolmogorov-Smirnov tests can also be computed to test for normality, with non-significant p-values indicating that the parametric properties of the data are not statistically different from a normal distribution (Field, 2018); however, the Shapiro-Wilk and Kolmogorov-Smirnov tests are sensitive to large sample sizes and should be interpreted cautiously. In addition, the data should be tested for linearity (Mvududu & Sink, 2013). Furthermore, extreme univariate and multivariate outliers must be identified and dealt with (i.e., removed, transformed, or winsorized; see Field, 2018) before a researcher can proceed with factor analysis. Univariate outliers can be identified via z-scores (> 3.29), box plots, or scatter plots, and multivariate outliers can be discovered by computing Mahalanobis distance (see Field, 2018).

Figure 2

Flow Chart for Reviewing Exploratory Factor Analysis

 

Three preliminary tests are necessary to determine if data are factorable, including (a) an inter-item correlation matrix, (b) the Kaiser–Meyer–Olkin (KMO) test for sampling adequacy, and (c) Bartlett’s test of sphericity (Beavers et al., 2013; Mvududu & Sink, 2013; Watson, 2017). The purpose of computing an inter-item correlation matrix is to identify redundant items (highly correlated) and individual items that do not fit with any of the other items (weakly correlated). An inter-item correlation matrix is factorable if a number of correlation coefficients for each item are between approximately r = .20 and r = .80 or .85 (Mvududu & Sink, 2013; Watson, 2017). Generally, a factor or subscale should be composed of at least three items (Mvududu & Sink, 2013); thus, an item should display intercorrelations between r = .20 and r = .80/.85 with at least three other items. However, inter-item correlations in this range with five to 10+ items are desirable (depending on the total number of items in the inter-item correlation matrix).

Bartlett’s test of sphericity is computed to test if the inter-item correlation matrix is an identity matrix, in which the correlations between the items is zero (Mvududu & Sink, 2013). An identity matrix is completely unfactorable (Mvududu & Sink, 2013); thus, desirable findings are a significant p-value, indicating that the correlation matrix is significantly different from an identity matrix. Finally, before proceeding with EFA, researchers should compute the KMO test for sampling adequacy, which is a measure of the shared variance among the items in the correlation matrix (Watson, 2017). Kaiser (1974) suggested the following guidelines for interpreting KMO values: “in the .90s – marvelous, in the .80s – meritorious, in the .70s – middling, in the .60s – mediocre, in the .50s – miserable, below .50 – unacceptable” (p. 35).

Factor Extraction Methods
     Factor extraction produces a factor solution by dividing up shared variance (also known as common variance) between each test item from its unique variance, or variance that is not shared with any other variables, and error variance, or variation in an item that cannot be accounted for by the factor solution (Mvududu & Sink, 2013). Historically, principal component analysis (PCA) was the dominant factor extraction method used in social sciences research. PCA, however, is now considered a method of data reduction rather than an approach to factor analysis because PCA extracts all of the variance (shared, unique, and error) in the model. Thus, although PCA can reduce the number of items in an inter-item correlation matrix, one cannot be sure if the factor solution is held together by shared variance (a potential theoretical model) or just by random error variance.

More contemporary factor extraction methods that only extract shared variance—for example, principal axis factoring (PAF) and maximum likelihood (ML) estimation methods—are generally recommended for EFA (Mvududu & Sink, 2013). PAF has utility if the data violate the assumption of normality, as PAF is robust to modest violations of normality (Mvududu & Sink, 2013). If, however, data are largely consistent with a normal distribution (skewness and kurtosis values ≤ ±1), researchers should consider using the ML extraction method. ML is advantageous, as it computes the likelihood that the inter-item correlation matrix was acquired from a population in which the extracted factor solution is a derivative of the scores on the items (Watson, 2017).

     Factor Retention. Once a factor extraction method is deployed, psychometric researchers are tasked with retaining the most parsimonious (simple) factor solution (Watson, 2017), as the purpose of factor analysis is to account for the maximum proportion of variance (ideally, 50%–75%+) in an inter-item correlation matrix while retaining the fewest possible number of items and factors (Mvududu & Sink, 2013). Four of the most commonly used criteria for determining the appropriate number of factors to retain in social sciences research include the (a) Kaiser criterion, (b) percentage of variance among items explained by each factor, (c) scree plot, and (d) parallel analysis (Mvududu & Sink, 2013; Watson, 2017). Kaiser’s criterion is a standard for retaining factors with Eigenvalues (EV) ≥ 1. An EV represents the proportion of variance that is explained by each factor in relation to the total amount of variance in the factor matrix.

The Kaiser criterion tends to overestimate the number of retainable factors; however, this criterion can be used to extract an initial factor solution (i.e., when computing the EFA for the first time). Interpreting the percentage of variance among items explained by each factor is another factor retention criterion based on the notion that a factor must account for a large enough percentage of variance to be considered meaningful (Mvududu & Sink, 2013). Typically, a factor should account for at least 5% of the variance in the total model. A scree plot is a graphical representation or a line graph that depicts the number of factors on the X-axis and the corresponding EVs on the Y-axis (see Figure 6 in Mvududu & Sink, 2013, p. 87, for a sample scree plot). The cutoff for the number of factors to retain is portrayed by a clear bend in the line graph, indicating the point at which additional factors fail to contribute a substantive amount of variance to the total model. Finally, in a parallel analysis, EVs are generated from a random data set based on the number of items and the sample size of the real (sample) data. The factors from the sample data with EVs larger than the EVs from the randomly generated data are retained based on the notion that these factors explain more variance than would be expected by random chance. In some instances, these four criteria will reveal different factor solutions. In such cases, researchers should retain the simplest factor solution that makes both statistical and substantive sense.

     Factor Rotation. After determining the number of factors to retain, researchers seek to uncover the association between the items and the factors or subscales (i.e., determining which items load on which factors) and strive to find simple structure or items with high factor loadings (close to ±1) on one factor and low factor loadings (near zero) on the other factors (Watson, 2017). The factors are rotated on vectors to enhance the readability or detection of simple structure (Mvududu & Sink, 2013). Orthogonal rotation methods (e.g., varimax, equamax, and quartimax) are appropriate when a researcher is measuring distinct or uncorrelated constructs of measurement. However, orthogonal rotation methods are rarely appropriate for use in counseling research, as counselors almost exclusively appraise variables that display some degree of inter-correlation (Mvududu & Sink, 2013). Oblique rotation methods (e.g., direct oblimin and promax) are generally more appropriate in counseling research, as they allow factors to inter-correlate by rotating the data on vectors at angles less than 90. The nature of oblique rotations allows the total variance accounted for by each factor to overlap; thus, the total variance explained in a post–oblique rotated factor solution can be misleading (Bandalos & Finney, 2019). For example, the total variance accounted for in a post–oblique rotated factor solution might add up to more than 100%. To this end, counselors should report the total variance explained by the factor solution before rotation as well as the sum of each factor’s squared structure coefficient following an oblique factor rotation.

Following factor rotation, researchers examine a number of factor retention criteria to determine the items that load on each factor (Watson, 2017). Commonality values (h2) represent the proportion of variance that the extracted factor solution explains for each item. Items with h2 values that range between .30 and .99 should be retained, as they share an adequate amount of shared variance with the other items and factors (Watson, 2017). Items with small h2 values (< .30) should be considered for removal. However, commonality values should not be too high (≥ 1), as this suggests one’s sample size was insufficient or too many factors were extracted (Watson, 2017). Items with problematic h2 values should be removed one at a time, and the EFA should be re-computed after each removal because these values will fluctuate following each deletion. Oblique factor rotation methods produce two matrices, including the pattern matrix, which displays the relationship between the items and a factor while controlling for the items’ association with the other factors, and the structure matrix, which depicts the correlation between the items and all of the factors (Mvududu & Sink, 2013). Researchers should examine both the pattern and the structure matrices and interpret the one that displays the clearest evidence of simple structure with the least evidence of cross-loadings.

Items should display a factor loading of at least ≥ .40 (≥ .50 is desirable) to mark a factor. Items that fail to meet a minimum factor loading of ≥ .40 should be deleted. Cross-loading is evident when an item displays factor loadings ≥ .30 to .35 on two or more factors (Beavers et al., 2013; Mvududu & Sink, 2013; Watson, 2017). Researchers may elect to assign a variable to one factor if that item’s loading is .10 higher than the next highest loading. Items that cross-load might also be deleted. Once again, items should be deleted one at a time and the EFA should be re-computed after each removal.

Naming the Rotated Factors
     The final step in EFA is naming the rotated factors; factor names should be brief (approximately one to four words) and capture the theoretical meaning of the group of items that comprise the factor (Mvududu & Sink, 2013). This is a subjective process, and the literature is lacking consistent guidelines for the process of naming factors. A research team can be incorporated into the process of naming their factors. Test developers can separately name each factor and then meet with their research team to discuss and eventually come to an agreement about the most appropriate name for each factor.

Confirmatory Factor Analysis
     CFA is an application of structural equation modeling for testing the extent to which a hypothesized factor solution (e.g., the factor solution that emerged in the EFA or another existing factor solution) demonstrates an adequate fit with a different sample (Kahn, 2006; Lewis, 2017). When validating scores on a new test, investigators should compute both EFA and CFA with two different samples from the same population, as the emergent internal structure in EFA can vary substantially. Researchers can collect two sequential samples or they may elect to collect one large sample and divide it into two smaller samples, one for EFA and the second for CFA.

Evaluating model fit in CFA is a complex task that is typically determined by examining the collective implications of multiple goodness-of-fit (GOF) indices, which include absolute, incremental, and parsimonious (Lewis, 2017). Absolute fit indices evaluate the extent to which the hypothesized model or the dimensionality of the existing measure fits with the data collected from a new sample. Incremental fit indices compare the improvement in fit between the hypothesized model and a null model (also referred to as an independence model) in which there is no correlation between observed variables. Parsimonious fit indices take the model’s complexity into account by testing the extent to which model fit is improved by estimating fewer pathways (i.e., creating a more parsimonious or simple model). Psychometric researchers generally report a combination of absolute, incremental, and parsimonious fit indices to demonstrate acceptable model fit (Mvududu & Sink, 2013). Table 1 includes tentative guidelines for interpreting model fit based on the synthesized recommendations of leading psychometric researchers from a comprehensive search of the measurement literature (Byrne, 2016; Dimitrov, 2012; Fabrigar et al., 1999; Hooper et al., 2008; Hu & Bentler, 1999; Kahn, 2006; Lewis, 2017; Mvududu & Sink, 2013; Schreiber et al., 2006; Worthington & Whittaker, 2006).

Table 1

Fit Indices and Tentative Thresholds for Evaluating Model Fit

Note. The fit indices and benchmarks to estimate the degree of model fit in this table are offered as tentative guidelines for scores on attitudinal measures based on the synthesized recommendations of numerous psychometric researchers (see citations in the “Confirmatory Factor Analysis” section of this article). The list of fit indices in this table are not all-inclusive (i.e., not all of them are typically reported). There is no universal approach for determining which fit indices to investigate nor are there any absolute thresholds for determining the degree of model fit. No single fix index is sufficient for determining model fit. Researchers are tasked with selecting and interpreting fit indices holistically (i.e., collectively), in ways that make both statistical and substantive sense based on their construct of measurement and goals of the study.
*.90 to .94 can denote an acceptable model fit for incremental fix indices; however, the majority of values should be ≥ .95.

 

Model Respecification
     The results of a CFA might reveal a poor or unacceptable model fit (see Table 1), indicating that the dimensionality of the hypothesized model that emerged from the EFA was not replicated or confirmed with a second sample (Mvududu & Sink, 2013). CFA is a rigorous model-fitting procedure and poor model fit in a CFA might indicate that the EFA-derived factor solution is insufficient for appraising the construct of measurement. CFA, however, is a more stringent test of structural validity than EFA, and psychometric researchers sometimes refer to the modification indices (also referred to as Lagrange multiplier statistics), which denote the expected decrease in the X2 value (i.e., degree of improvement in model fit) if the parameter is freely estimated (Dimitrov, 2012). In these instances, correlating the error terms between items or removing problematic items will improve model fit; however, when considering model respecification, psychometric researchers should proceed cautiously, if at all, as a strong theoretical justification is necessary to defend model respecification (Byrne, 2016; Lewis, 2017; Schreiber et al., 2006). Researchers should also be clear that model respecification causes the CFA to become an EFA because they are investigating the dimensionality of a different or modified model rather than confirming the structure of an existing, hypothesized model.

Higher-Order CFA
     Higher-order CFA is an extension of CFA that allows researchers to test nested models and determine if a second-order latent variable (factor) explains the associations between the factors in a single-order CFA (Credé & Harms, 2015). Similar to single-order CFA (see Figure 3, Model 1) in which the test items cluster together to form the factors or subscales, higher-order CFA reveals if the factors are related to one another strongly enough to suggest the presence of a global factor (see Figure 3, Model 3). Suppose, for example, the test developer of a scale for measuring dimensions of the therapeutic alliance confirmed the three following subscales via single-order CFA (see Figure 3, Model 1): Empathy, Unconditional Positive Regard, and Congruence. Computing a higher-order CFA would reveal if a higher-order construct, which the research team might name Therapeutic Climate, is present in the data. In other words, higher-order CFA reveals if Empathy, Unconditional Positive Regard, and Congruence, collectively, comprise the second-order factor of Therapeutic Climate.

Determining if a higher-order factor explains the co-variation (association) between single-order factors is a complex undertaking. Thus, researchers should consider a number of criteria when deciding if their data are appropriate for higher-order CFA (Credé & Harms, 2015). First, moderate-to-strong associations (co-variance) should exist between first-order factors. Second, the unidimensional factor solution (see Figure 3, Model 2) should display a poor model fit (see Table 1) with the data. Third, theoretical support should exist for the presence of a higher-order factor. Referring to the example in the previous paragraph, person-centered therapy provides a theory-based explanation for the presence of a second-order or global factor (Therapeutic Climate) based on the integration of the single-order factors (Empathy, Unconditional Positive Regard, and Congruence). In other words, the presence of a second-order factor suggests that Therapeutic Climate explains the strong association between Empathy, Unconditional Positive Regard, and Congruence.

Finally, the single-order factors should display strong factor loadings (approximately ≥ .70) on the higher-order factor. However, there is not an absolute consensus among psychometric researchers regarding the criteria for higher-order CFA and the criteria summarized in this section are not a dualistic decision rule for retaining or rejecting a higher-order model. Thus, researchers are tasked with presenting that their data meet a number of criteria to justify the presence of a higher-order factor. If the results of a higher-order CFA reveal an acceptable model fit (see Table 1), researchers should directly compare (e.g., chi-squared test of difference) the single-order and higher-order models to determine if one model demonstrates a superior fit with the data at a statistically significant level.

Figure 3

Single-Order, Unidimensional, and Higher-Order Factor Solutions

 

Multiple-Group Confirmatory Factor Analysis
     Multiple-group confirmatory factor analysis (MCFA) is an extension of CFA for testing the factorial invariance (psychometric equivalence) of a scale across subgroups of a sample or population (C.-C. Chen et al., 2020; Dimitrov, 2010). In other words, MCFA has utility for testing the extent to which a particular construct has the same meaning across different groups of a larger sample or population. Suppose, for example, the developer of the Therapeutic Climate scale (see example in the previous section) validated scores on their scale with undergraduate college students. Invariance testing has potential to provide further support for the internal structure validity of the scale by testing whether Empathy, Unconditional Positive Regard, and Congruence have the same meaning across different subgroups of undergraduate college students (e.g., between different gender identities, ethnic identities, age groups, and other subgroups of the larger sample).

     Levels of Invariance. Factorial invariance can be tested in a number of different ways and includes the following primary levels or aspects: (a) configural invariance, (b) measurement (metric, scalar, and strict) invariance, and (c) structural invariance (Dimitrov, 2010, 2012). Configural invariance (also referred to as pattern invariance) serves as the baseline mode (typically the best fitting model with the data), which is used as the point of comparison when testing for metric, scalar, and structural invariance. In layperson’s terms, configural invariance is a test of whether the scales are approximately similar across groups.

Measurement invariance includes testing for metric and scalar invariance. Metric invariance is a test of whether each test item makes an approximately equal contribution (i.e., approximately equal factor loadings) to the latent variable (composite scale score). In layperson’s terms, metric invariance evaluates if the scale reasonably captures the same construct. Scalar invariance adds a layer of rigor to metric invariance by testing if the differences between the average scores on the items are attributed to differences in the latent variable means. In layperson’s terms, scalar invariance indicates that if the scores change over time, they change in the same way.

Strict invariance is the most stringent level of measurement invariance testing and tests if the sum total of the items’ unique variance (item variation that is not in common with the factor) is comparable to the error variance across groups. In layperson’s terms, the presence of strict invariance demonstrates that score differences between groups are exclusively due to differences in the common latent variables. Strict invariance, however, is typically not examined in social sciences research because the latent factors are not composed of residuals. Thus, residuals are negligible when evaluating mean differences in latent scores (Putnick & Bornstein, 2016).

Finally, structural invariance is a test of whether the latent factor variances are equivalent to the factor covariances (Dimitrov, 2010, 2012). Structural invariance tests the null hypothesis that there are no statistically significant differences between the unconstrained and constrained models (i.e., determines if the unconstrained model is equivalent to the constrained model). Establishing structural invariance indicates that when the structural pathways are allowed to vary across the two groups, they naturally produce equal results, which supports the notion that the structure of the model is invariant across both groups. In layperson’s terms, the presence of structural invariance indicates that the pathways (directionality) between variables behave in the same way across both groups. It is necessary to establish configural and metric invariance prior to testing for structural invariance.

     Sample Size and Criteria for Evaluating Invariance. Researchers should check their sample size before computing invariance testing, as small samples (approximately < 200) can overestimate model fit (Dimitrov, 2010). Similar to single-order CFA, no absolute sample size guidelines exist in the literature for invariance testing. Generally, a minimum sample of at least 200 participants per group is recommended for invariance testing (although < 200 to 300+ is advantageous). Referring back to the Therapeutic Climate scale example (see the previous section), investigators would need a minimum sample of 400 if they were seeking to test the invariance of the scale by generational status (200 first generation + 200 non-first generation = 400). The minimum sample size would increase as more levels are added. For example, a minimum sample of 600 would be recommended if investigators quantified generational status on three levels (200 first generation + 200 second generation + 200 third generation and beyond = 600).

Factorial invariance is investigated through a computation of the change in model fit at each level of invariance testing (F. F. Chen, 2007). Historically, the Satorra and Bentler chi-square difference test was the sole criteria for testing factorial invariance, with a non-significant p-value indicating factorial invariance (Putnick & Bornstein, 2016). The chi-square difference test is still commonly reported by contemporary psychometric researchers; however, it is rarely used as the sole criteria for determining invariance, as the test is sensitive to large samples. The combined recommendations of F. F. Chen (2007) and Putnick and Bornstein (2016) include the following thresholds for investigating invariance: ≤ ∆ 0.010 in CFI, ≤ ∆ 0.015 in RMSEA, and ≤ ∆ 0.030 in SRMR for metric invariance or ≤ ∆ 0.015 in SRMR for scalar invariance. In a simulation study, Kang et al. (2016) found that McDonald’s NCI (MNCI) outperformed the CFI in terms of stability. Kang et al. (2016) recommend < ∆ 0.007 in MNCI for the 5th percentile and ≤ ∆ 0.007 in MNCI for the 1st percentile as cutoff values for measurement quality. Strong measurement invariance is achieved when both metric and scalar invariance are met, and weak invariance is accomplished when only metric invariance is present (Dimitrov, 2010).

Exemplar Review of a Psychometric Study

     The following section will include a review of an exemplar psychometric study based on the recommendations for EFA (see Figure 2) and CFA (see Table 1) that are provided in this manuscript. In 2020, I collaborated with Ryan Flinn on the development and validation of scores on the Mental Distress Response Scale (MDRS) for appraising how college students are likely to respond when encountering a peer in mental distress (Kalkbrenner & Flinn, 2020). A total of 13 items were entered into an EFA. Following the steps for EFA (see Figure 1), the sample size (N = 569) exceeded the guidelines for sample size that I published in my 2021 article (Kalkbrenner, 2021b), including an STV of 10:1 or 200 participants, whichever produces a larger sample. Flinn and I (2020) ensured that our 2020 study’s data were consistent with a normal distribution (skewness & kurtosis values ≤ ±1) and computed preliminary assumption checking, including inter-item correlation matrix, KMO (.73), and Bartlett’s test of sphericity (p < .001).

An ML factor extraction method was employed, as the data were largely consistent (skewness & kurtosis values ≤ ±1) with a normal distribution. We used the three most rigorous factor retention criteria—percentage of variance accounted for, scree test, and parallel analysis—to extract a two-factor solution. An oblique factor rotation method (direct oblimin) was employed, as the two factors were correlated. We referred to the recommended factor retention criteria, including h2 values .30 to .99, factor loadings ≥ .40, and cross-loading ≥ .30, to eliminate one item with low commonalities and two cross-loading items. Using a research team, we named the first factor Diminish/Avoid, as each item that marked this factor reflected a dismissive or evasive response to encountering a peer in mental distress. The second factor was named Approach/Encourage because each item that marked this factor included a response to a peer in mental distress that was active and likely to help connect their peer to mental health support services.

Our next step was to compute a CFA by administering the MDRS to a second sample of undergraduate college students to confirm the two-dimensional factor solution that emerged in the EFA. The sample size (N = 247) was sufficient for CFA (STV > 10:1 and > 200 participants). The MDRS items were entered into a CFA and the following GOF indices emerged: CMIN = χ2 (34) = 61.34, p = .003, CMIN/DF = 1.80, CFI = .96, IFI = .96, RMSEA = .06, 90% CI [0.03, 0.08], and SRMR = .04. A comparison between our GOF indices from the 2020 study with the thresholds for evaluating model fit in Table 1 reveal an acceptable-to-strong fit between the MDRS model and the data. Collectively, our 2020 procedures for EFA and CFA were consistent with the recommendations in this manuscript.

Implications for the Profession

Implications for Counseling Practitioners
     Assessment literacy is a vital component of professional counseling practice, as counselors who practice in a variety of specialty areas select and administer tests to clients and use the results to inform diagnosis and treatment planning (C.-C. Chen et al., 2020; Mvududu & Sink, 2013; NBCC, 2016; Neukrug & Fawcett, 2015). It is important to note that test results alone should not be used to make diagnoses, as tests are not inherently valid (Kalkbrenner, 2021b). In fact, the authors of the Diagnostic and Statistical Manual of Mental Disorders stated that “scores from standardized measures and interview sources must be interpreted using clinical judgment” (American Psychiatric Association, 2013, p. 37). Professional counselors can use test results to inform their diagnoses; however, diagnostic decision making should ultimately come down to a counselor’s clinical judgment.

Counseling practitioners can refer to this manuscript as a reference for evaluating the internal structure validity of scores on a test to help determine the extent to which, if any at all, the test in question is appropriate for use with clients. When evaluating the rigor of an EFA for example, professional counselors can refer to this manuscript to evaluate the extent to which test developers followed the appropriate procedures (e.g., preliminary assumption checking, factor extraction, retention, and rotation [see Figure 2]). Professional counselors are encouraged to pay particular attention to the factor extraction method that the test developers employed, as PCA is sometimes used in lieu of more appropriate methods (e.g., PAF/ML). Relatedly, professional counselors should be vigilant when evaluating the factor rotation method employed by test developers because oblique rotation methods are typically more appropriate than orthogonal (e.g., varimax) for counseling tests.

CFA is one of the most commonly used tests of the internal structure validity of scores on psychological assessments (Kalkbrenner, 2021b). Professional counselors can compare the CFA fit indices in a test manual or journal article to the benchmarks in Table 1 and come to their own conclusion about the internal structure validity of scores on a test before using it with clients. Relatedly, the layperson’s definitions of common psychometric terms in Figure 1 might have utility for increasing professional counselors’ assessment literacy by helping them decipher some of the psychometric jargon that commonly appears in psychometric studies and test manuals.

Implications for Counselor Education
     Assessment literacy begins in one’s counselor education program and it is imperative that counselor educators teach their students to be proficient in recognizing and evaluating internal structure validity evidence of test scores. Teaching internal structure validity evidence can be an especially challenging pursuit because counseling students tend to fear learning about psychometrics and statistics (Castillo, 2020; Steele & Rawls, 2015), which can contribute to their reticence and uncertainty when encountering psychometric research. This reticence can lead one to read the methodology section of a psychometric study briefly, if at all. Counselor educators might suggest the present article as a resource for students taking classes in research methods and assessment as well as for students who are completing their practicum, internship, or dissertation who are evaluating the rigor of existing measures for use with clients or research participants.

Counselor educators should urge their students not to skip over the methodology section of a psychometric study. When selecting instrumentation for use with clients or research participants, counseling students and professionals should begin by reviewing the methodology sections of journal articles and test manuals to ensure that test developers employed rigorous and empirically supported procedures for test development and score validation. Professional counselors and their students can compare the empirical steps and guidelines for structural validation of scores that are presented in this manuscript with the information in test manuals and journal articles of existing instrumentation to evaluate its internal structure. Counselor educators who teach classes in assessment or psychometrics might integrate an instrument evaluation assignment into the course in which students select a psychological instrument and critique its psychometric properties. Another way that counselor educators who teach classes in current issues, research methods, assessment, or ethics can facilitate their students’ assessment literacy development is by creating an assignment that requires students to interview a psychometric researcher. Students can find psychometric researchers by reviewing the editorial board members and authors of articles published in the two peer-reviewed journals of the Association for Assessment and Research in Counseling, Measurement and Evaluation in Counseling and Development and Counseling Outcome Research and Evaluation. Students might increase their interest and understanding about the necessity of assessment literacy by talking to researchers who are passionate about psychometrics.

Assessment Literacy: Additional Considerations

Internal structure validity of scores is a crucial component of assessment literacy for evaluating the construct validity of test scores (Bandalos & Finney, 2019). Assessment literacy, however, is a vast construct and professional counselors should consider a number of additional aspects of test worthiness when evaluating the potential utility of instrumentation for use with clients. Reviewing these additional considerations is beyond the scope of this manuscript; however, readers can refer to the following features of assessment literacy and corresponding resources: reliability (Kalkbrenner, 2021a), practicality (Neukrug & Fawcett, 2015), steps in the instrument development process (Kalkbrenner, 2021b), and convergent and divergent validity evidence of scores (Swank & Mullen, 2017). Moreover, the discussion of internal structure validity evidence of scores in this manuscript is based on Classical Test Theory (CTT), which tends to be an appropriate platform for attitudinal measures. However, Item Response Theory (see Amarnani, 2009) is an alternative to CTT with particular utility for achievement and aptitude testing.

Cross-Cultural Considerations in Assessment Literacy
     Professional counselors have an ethical obligation to consider the cross-cultural fairness of a test before use with clients, as the validity of test scores are culturally dependent (American Counseling Association [ACA], 2014; Kane, 2010; Neukrug & Fawcett, 2015; Swanepoel & Kruger, 2011). Cross-cultural fairness (also known as test fairness) in testing and assessment “refers to the comparability of score meanings across individuals, groups or settings” (Swanepoel & Kruger, 2011, p. 10). There exists some overlap between internal structure validity and cross-cultural fairness; however, some distinct differences exist as well.

Using CFA to confirm the factor structure of an established test with participants from a different culture is one way to investigate the cross-cultural fairness of scores. Suppose, for example, an investigator found acceptable internal structure validity evidence (see Table 1) for scores on an anxiety inventory that was normed in America with participants in Eastern Europe who identify with a collectivist cultural background. Such findings would suggest that the dimensionality of the anxiety inventory extends to the sample of Eastern European participants. However, internal structure validity testing alone might not be sufficient for testing the cross-cultural fairness of scores, as factor analysis does not test for content validity. In other words, although the CFA confirmed the dimensionality of an American model with a sample of Eastern European participants, the analysis did not take potential qualitative differences about the construct of measurement (anxiety severity) into account. It is possible (and perhaps likely) that the lived experience of anxiety differs between those living in two different cultures. Accordingly, a systems-level approach to test development and score validation can have utility for enhancing the cross-cultural fairness of scores (Swanepoel & Kruger, 2011).

A Systems-Level Approach to Test Development and Score Validation
     Swanepoel and Kruger (2011) outlined a systemic approach to test development that involves circularity, which includes incorporating qualitative inquiry into the test development process, as qualitative inquiry has utility for uncovering the nuances of participants’ lived experiences that quantitative data fail to capture. For example, an exploratory-sequential mixed-methods design in which qualitative findings are used to guide the quantitative analyses is a particularly good fit with systemic approaches to test development and score validation. Referring to the example in the previous section, test developers might conduct qualitative interviews to develop a grounded theory of anxiety severity in the context of the collectivist culture. The grounded theory findings could then be used as the theoretical framework (see Kalkbrenner, 2021b) for a psychometric study aimed at testing the generalizability of the qualitative findings. Thus, in addition to evaluating the rigor of factor analytic results, professional counselors should also review the cultural context in which test items were developed before administering a test to clients.

Language adaptions of instrumentation are another relevant cross-cultural fairness consideration in counseling research and practice. Word-for-word translations alone are insufficient for capturing cross-cultural fairness of instrumentation, as culture extends beyond just language (Lenz et al., 2017; Swanepoel & Kruger, 2011). Pure word-for-word translations can also cause semantic errors. For example, feeling “fed up” might translate to feeling angry in one language and to feeling full after a meal in another language. Accordingly, professional counselors should ensure that a translated instrument was subjected to rigorous procedures for maintaining cross-cultural fairness. Reviewing such procedures is beyond the scope of this manuscript; however, Lenz et al. (2017) outlined a 6-step process for language translation and cross-cultural adaptation of instruments.

Conclusion

Gaining a deeper understanding of the major approaches to factor analysis for demonstrating internal structure validity in counseling research has potential to increase assessment literacy among professional counselors who work in a variety of specialty areas. It should be noted that the thresholds for interpreting the strength of internal structure validity coefficients that are provided throughout this manuscript should be used as tentative guidelines, not unconditional standards. Ultimately, internal structure validity is a function of test scores and the construct of measurement. The stakes or consequences of test results should be considered when making final decisions about the strength of validity coefficients. As professional counselors increase their familiarity with factor analysis, they will most likely become more cognizant of the strengths and limitations of counseling-related tests to determine their utility for use with clients. The practical overview of factor analysis presented in this manuscript can serve as a one-stop shop or resource that professional counselors can refer to as a reference for selecting tests with validated scores for use with clients, a primer for teaching courses, and a resource for conducting their own research.

 

Conflict of Interest and Funding Disclosure
The author reported no conflict of interest
or funding contributions for the development
of this manuscript.


References

Amarnani, R. (2009). Two theories, one theta: A gentle introduction to item response theory as an alternative to classical test theory. The International Journal of Educational and Psychological Assessment, 3, 104–109.

American Counseling Association. (2014). ACA code of ethics. https://www.counseling.org/resources/aca-code-of-ethics.pdf

American Educational Research Association, American Psychological Association, National Council on Measurement in Education. (2014). Standards for educational and psychological testing. https://www.apa.org/science/programs/testing/standards

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).
https://doi.org/10.1176/appi.books.9780890425596

Bandalos, D. L., & Finney, S. J. (2019). Factor analysis: Exploratory and confirmatory. In G. R. Hancock, L. M. Stapleton, & R. O. Mueller (Eds.), The reviewer’s guide to quantitative methods in the social sciences (2nd ed., pp. 98–122). Routledge.

Beavers, A. S., Lounsbury, J. W., Richards, J. K., Huck, S. W., Skolits, G. J., & Esquivel, S. L. (2013). Practical considerations for using exploratory factor analysis in educational research. Practical Assessment, Research and Evaluation, 18(5/6), 1–13. https://doi.org/10.7275/qv2q-rk76

Byrne, B. M. (2016). Structural equation modeling with AMOS: Basic concepts, applications, and programming (3rd ed.). Routledge.

Castillo, J. H. (2020). Teaching counseling students the science of research. In M. O. Adekson (Ed.), Beginning your counseling career: Graduate preparation and beyond (pp. 122–130). Routledge.

Chen, C.-C., Lau, J. M., Richardson, G. B., & Dai, C.-L. (2020). Measurement invariance testing in counseling. Journal of Professional Counseling: Practice, Theory & Research, 47(2), 89–104.
https://doi.org/10.1080/15566382.2020.1795806

Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling, 14(3), 464–504. https://doi.org/10.1080/10705510701301834

Council for Accreditation of Counseling and Related Educational Programs. (2015). 2016 CACREP standards. http://www.cacrep.org/wp-content/uploads/2017/08/2016-Standards-with-citations.pdf

Credé, M., & Harms, P. D. (2015). 25 years of higher-order confirmatory factor analysis in the organizational sciences: A critical review and development of reporting recommendations. Journal of Organizational
Behavior
, 36(6), 845–872. https://doi.org/10.1002/job.2008

Dimitrov, D. M. (2010). Testing for factorial invariance in the context of construct validation. Measurement and Evaluation in Counseling and Development, 43(2), 121–149. https://doi.org/10.1177/0748175610373459

Dimitrov, D. M. (2012). Statistical methods for validation of assessment scale data in counseling and related fields. American Counseling Association.

Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299.
https://doi.org/10.1037/1082-989X.4.3.272

Field, A. (2018). Discovering statistics using IBM SPSS statistics (5th ed.). SAGE.

Hooper, D., Coughlan, J., & Mullen, M. R. (2008). Structural equation modelling: Guidelines for determining model fit. The Electronic Journal of Business Research Methods, 6(1), 53–60.

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

Kahn, J. H. (2006). Factor analysis in counseling psychology research, training, and practice: Principles, advances, and applications. The Counseling Psychologist, 34(5), 684–718. https://doi.org/10.1177/0011000006286347

Kaiser, H. F. (1974). An index of factorial simplicity. Psychometrika, 39(1), 31–36. https://doi.org/10.1007/BF02291575

Kalkbrenner, M. T. (2021a). Alpha, omega, and H internal consistency reliability estimates: Reviewing these options and when to use them. Counseling Outcome Research and Evaluation. Advance online publication. https://doi.org/10.1080/21501378.2021.1940118

Kalkbrenner, M. T. (2021b). A practical guide to instrument development and score validation in the social sciences: The MEASURE Approach. Practical Assessment, Research, and Evaluation, 26, Article 1. https://scholarworks.umass.edu/pare/vol26/iss1/1

Kalkbrenner, M. T., & Flinn, R. E. (2020). The Mental Distress Response Scale and promoting peer-to-peer mental health support: Implications for college counselors and student affairs officials. Journal of College Student Development, 61(2), 246–251. https://doi.org/10.1353/csd.2020.0021

Kane, M. (2010). Validity and fairness. Language Testing, 27(2), 177–182. https://doi.org/10.1177/0265532209349467

Kang, Y., McNeish, D. M., & Hancock, G. R. (2016). The role of measurement quality on practical guidelines for assessing measurement and structural invariance. Educational and Psychological Measurement, 76(4), 533–561. https://doi.org/10.1177/0013164415603764

Lenz, A. S., Gómez Soler, I., Dell’Aquilla, J., & Uribe, P. M. (2017). Translation and cross-cultural adaptation of assessments for use in counseling research. Measurement and Evaluation in Counseling and Development, 50(4), 224–231. https://doi.org/10.1080/07481756.2017.1320947

Lewis, T. F. (2017). Evidence regarding the internal structure: Confirmatory factor analysis. Measurement and Evaluation in Counseling and Development, 50(4), 239–247. https://doi.org/10.1080/07481756.2017.1336929

Mvududu, N. H., & Sink, C. A. (2013). Factor analysis in counseling research and practice. Counseling Outcome Research and Evaluation, 4(2), 75–98. https://doi.org/10.1177/2150137813494766

National Board for Certified Counselors. (2016). NBCC code of ethics. https://www.nbcc.org/Assets/Ethics/NBCCCodeofEthics.pdf

Neukrug, E. S., & Fawcett, R. C. (2015). Essentials of testing and assessment: A practical guide for counselors, social workers, and psychologists (3rd ed.). Cengage.

Putnick, D. L., & Bornstein, M. H. (2016). Measurement invariance conventions and reporting: The state of the art and future directions for psychological research. Developmental Review, 41, 71–90.  https://doi.org/10.1016/j.dr.2016.06.004

Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting structural equation modeling and confirmatory factor analysis results: A review. Journal of Educational Research, 99(6), 323–338.
https://doi:10.3200/JOER.99.6.323-338

Steele, J. M., & Rawls, G. J. (2015). Quantitative research attitudes and research training perceptions among master’s-level students. Counselor Education and Supervision, 54(2), 134–146. https://doi.org/10.1002/ceas.12010

Swanepoel, I., & Kruger, C. (2011). Revisiting validity in cross-cultural psychometric-test development: A systems-informed shift towards qualitative research designs. South African Journal of Psychiatry, 17(1), 10–15. https://doi.org/10.4102/sajpsychiatry.v17i1.250

Swank, J. M., & Mullen, P. R. (2017). Evaluating evidence for conceptually related constructs using bivariate correlations. Measurement and Evaluation in Counseling and Development, 50(4), 270–274.
https://doi.org/10.1080/07481756.2017.1339562

Tate, K. A., Bloom, M. L., Tassara, M. H., & Caperton, W. (2014). Counselor competence, performance assessment, and program evaluation: Using psychometric instruments. Measurement and Evaluation in Counseling and Development, 47(4), 291–306. https://doi.org/10.1177/0748175614538063

Watson, J. C. (2017). Establishing evidence for internal structure using exploratory factor analysis. Measurement and Evaluation in Counseling and Development, 50(4), 232–238. https://doi.org/10.1080/07481756.2017.1336931

Worthington, R. L., & Whittaker, T. A. (2006). Scale development research: A content analysis and recommendations for best practices. The Counseling Psychologist, 34(6), 806–838. https://doi.org/10.1177/0011000006288127

Michael T. Kalkbrenner, PhD, NCC, is an associate professor at New Mexico State University. Correspondence may be addressed to Michael T. Kalkbrenner, Department of Counseling and Educational Psychology, New Mexico State University, Las Cruces, NM 88003, mkalk001@nmsu.edu.

 

Development of the Psychological Maltreatment Inventory

Alison M. Boughn, Daniel A. DeCino

 

This article introduces the development and implementation of the Psychological Maltreatment Inventory (PMI) assessment with child respondents receiving services because of an open child abuse and/or neglect case in the Midwest (N = 166). Sixteen items were selected based on the literature, subject matter expert refinement, and readability assessments. Results indicate the PMI has high reliability (α = .91). There was no evidence the PMI total score was influenced by demographic characteristics. A positive relationship was discovered between PMI scores and general trauma symptom scores on the Trauma Symptom Checklist for Children Screening Form (TSCC-SF; r = .78, p = .01). Evidence from this study demonstrates the need to refine the PMI for continued use with children. Implications for future research include identification of psychological maltreatment in isolation, further testing and refinement of the PMI, and exploring the potential relationship between psychological maltreatment and suicidal ideation. 

Keywords: psychological maltreatment, child abuse, neglect, assessment, trauma

 

In 2012, the Centers for Disease Control (CDC; 2012) reported that the total cost of child maltreatment (CM) in 2008, including psychological maltreatment (PM), was $124 billion. Fang et al. (2012) estimated the lifetime burden of CM in 2008 was as high as $585 billion. The CDC (2012) characterized CM as rivaling “other high profile public health problems” (para. 1). By 2015, the National Institutes of Health reported the total cost of CM, based on substantiated incidents, was reported to be $428 billion, a 345% increase in just 7 years; the true cost was predictably much higher (Peterson et al., 2018). Using the sensitivity analysis done by Fang et al. (2012), the lifetime burden of CM in 2015 may have been as high as $2 trillion. If these trends continue unabated, the United States could expect a total cost for CM, including PM, of $5.1 trillion by 2030, with a total lifetime cost of $24 trillion. More concerning, this increase would not account for any impact from the COVID-19 pandemic.

Mental health first responders and child protection professionals may encounter PM regularly in their careers (Klika & Conte, 2017; U.S. Department of Health and Human Services [DHHS], 2018). PM experiences are defined as inappropriate emotional and psychological acts (e.g., excessive yelling, threatening language or behavior) and/or lack of appropriate acts (e.g., saying I love you) used by perpetrators of abuse and neglect to gain organizational control of their victims (American Professional Society on the Abuse of Children [APSAC], 2019; Klika & Conte, 2017; Slep et al., 2015). Victims may experience negative societal perceptions (i.e., stigma), fear of retribution from caregivers or guardians, or misdiagnosis by professional helpers (Iwaniec, 2006; López et al., 2015). They often face adverse consequences that last their entire lifetime (Spinazzola et al., 2014; Tyrka et al., 2013; Vachon et al., 2015; van der Kolk, 2014; van Harmelen et al., 2010; Zimmerman & Mercy, 2010). PM can be difficult to identify because it leaves no readily visible trace of injury (e.g., bruises, cuts, or broken bones), making it complicated to substantiate that a crime has occurred (Ahern et al., 2014; López et al., 2015). Retrospective data outlines evaluation processes for PM identification in adulthood; however, childhood PM lacks a single definition and remains difficult to assess (Tonmyr et al., 2011). These complexities in identifying PM in children may prevent mental health professionals from intervening early, providing crucial care, and referring victims for psychological health services (Marshall, 2012; Spinazzola et al., 2014). The Psychological Maltreatment Inventory (PMI) is the first instrument of its kind to address these deficits.

Child Psychological Maltreatment
     Although broadly conceptualized, child PM experiences are described as literal acts, events, or experiences that create current or future symptoms that can affect a victim without immediate physical evidence (López et al., 2015). Others have extended child PM to include continued patterns of severe events that impede a child from securing basic psychological needs and convey to the child that they are worthless, flawed, or unwanted (APSAC, 2019). Unfortunately, these broad concepts lack the specificity to guide legal and mental health interventions (Ahern et al., 2014). Furthermore, legal definitions of child PM vary from jurisdiction to jurisdiction and state to state (Spinazzola et al., 2014). The lack of consistent definitions and quantifiable measures of child PM may create barriers for prosecutors and other helping professionals within the legal system as well as a limited understanding of PM in evidence-based research (American Psychiatric Association [APA], 2013; APSAC, 2019; Klika & Conte, 2017). These challenges are exacerbated by comorbidity with other forms of maltreatment.

Co-Occurring Forms of Maltreatment
     According to DHHS (2018), child PM is rarely documented as occurring in isolation compared to other forms of maltreatment (i.e., physical abuse, sexual abuse, or neglect). Rather, researchers have found PM typically coexists with other forms of maltreatment (DHHS, 2018; Iwaniec, 2006; Marshall, 2012). Klika and Conte (2017) reported that perpetrators who use physical abuse, inappropriate language, and isolation facilitate conditions for PM to coexist with other forms of abuse. Van Harmelen et al. (2011) argued that neglectful acts constitute evidence of PM (e.g., seclusion; withholding medical attention; denying or limiting food, water, shelter, and other basic needs).

Consequences of PM Experienced in Childhood
     Mills et al. (2013) and Greenfield and Marks (2010) noted PM experiences in early childhood might manifest in physical growth delays and require access to long-term care throughout a victim’s lifetime. Children who have experienced PM may suffer from behaviors that delay or prevent meeting developmental milestones, achieving academic success in school, engaging in healthy peer relationships, maintaining physical health and well-being, forming appropriate sexual relationships as adults, and enjoying satisfying daily living experiences (Glaser, 2002; Maguire et al., 2015). Neurological and cognitive effects of PM in childhood impact children as they transition into adulthood, including abnormalities in the amygdala and hippocampus (Tyrka at al., 2013). Brown et al. (2019) found that adults who reported experiences of CM had higher rates of negative responses to everyday stress, a larger constellation of unproductive coping skills, and earlier mortality rates (Brown et al., 2019; Felitti et al., 1998). Furthermore, adults with childhood PM experiences reported higher rates of substance abuse than those compared to control groups (Felitti et al., 1998).

     Trauma-Related Symptomology. Researchers speculate that children exposed to maltreatment and crises, especially those that come without warning, are at greater risk for developing a host of trauma-related symptoms (Spinazzola et al., 2014). Developmentally, children lack the ability to process and contextualize their lived experiences. Van Harmelen et al. (2010) discovered that adults who experienced child PM had decreased prefrontal cortex mass compared to those without evidence of PM. Similarly, Field et al. (2017) found those unable to process traumatic events produced higher levels of stress hormones (i.e., cortisol, epinephrine, norepinephrine); these hormones are produced from the hypothalamic-pituitary-adrenal (HPA) and sympathetic-adrenal-medullary (SAM) regions in the brain. Some researchers speculate that elevated levels of certain hormones and hyperactive regions within the brain signal the body’s biological attempt to reduce the negative impact of PM through the fight-flight-freeze response (Porges, 2011; van der Kolk, 2014).

Purpose of Present Study
     At the time of this research, there were few formal measures using child self-report to assess how children experience PM. We developed the PMI as an initial quantifiable measure of child PM for children and adolescents between the ages of 8 and 17, as modeled by Tonmyr and colleagues (2011). The PMI was developed in multiple stages, including 1) a review of the literature, 2) a content validity survey with subject matter experts (SMEs), 3) a pilot study (N = 21), and 4) a large sample study (N = 166). An additional instrument, the Trauma Symptom Checklist for Children Screening Form (TSCC-SF; Briere & Wherry, 2016), was utilized in conjunction with the PMI to explore occurrences of general trauma symptoms among respondents. The following four research questions were investigated:

  1. How do respondent demographics relate to PM?
  2. What is the rate of PM experience with respondents who are presently involved in an open CM case?
  3. What is the co-occurrence of PM among various forms of CM allegations?
  4. What is the relationship between the frequency of reported PM experiences and the frequency of general trauma symptoms?

Method

Study 1: PMI Item Development and Pilot
     Following the steps of scale construction (Heppner et al., 2016), the initial version of the PMI used current literature and definitions from facilities nationwide that provide care for children who have experienced maltreatment and who are engaged with court systems, mental health agencies, or social services. Our lead researcher, Alison M. Boughn, developed a list of 20 items using category identifications from Glaser (2002) and APSAC (2019). Items were also created using Slep et al.’s (2015) proposed inclusion language for the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) diagnostic codes and codes from the International Classification of Diseases, 11th edition (ICD-11) definition criteria (APA, 2013). Both Boughn and Daniel A. DeCino, our other researcher, reviewed items for consistency with the research literature and removed four redundant items. The final 16 items were reevaluated for readability for future child respondents using a web-based, age range–appropriate readability checker (Readable, n.d.) and were then presented to local SMEs in a content validity survey to determine which would be considered essential for children to report as part of a child PM assessment.

Expert Validation
     A multidisciplinary team (MDT) serving as SMEs completed an online content validity survey created by Boughn. The survey was distributed by a Child Advocacy Center (CAC) manager to the MDT. Boughn used the survey results to validate the PMI’s item content relevance. Twenty respondents from the following professions completed the survey: mental health (n = 6), social services (n = 6), law enforcement (n = 3), and legal services (n = 5). The content validity ratio (CVR) was then calculated for the 16 proposed items.

     Results. The content validity survey scale used a 3-point Likert-type scale: 0 = not necessary; 1 = useful, but not essential; and 2 = essential. A minimum of 15 of the 20 SMEs (75% of the sample), or a CVR ≥ .5, was required to deem an item essential (Lawshe, 1975). The significance level for each item’s content validity was set at α = .05 (Ayre & Scally, 2014). After conducting Lawshe’s (1975) CVR and applying the ratio correction developed by Ayre and Scally (2014), it was determined that eight items were essential: Item 2 (CVR = .7), Item 3 (CVR = .9), Item 4 (CVR = .6), Item 6 (CVR = .6), Item 7 (CVR = .8), Item 10 (CVR = .6), Item 15 (CVR = .5), and Item 16 (CVR = .6).

Upon further evaluation, and in an effort to ensure that the PMI items served the needs of interdisciplinary professionals, some items were rated essential for specific professions; these items still met the CVR requirements (CVR = 1) for the smaller within-group sample. These four items were unanimously endorsed by SMEs for a particular profession as essential: Item 5 (CVR Social Services = 1; CVR Law Enforcement = 1), Item 11 (CVR Law Enforcement = 1), Item 13 (CVR Law Enforcement = 1), and Item 14 (CVR Law Enforcement = 1).

Finally, an evaluation of the remaining four items was completed to explore if items were useful, but not essential. Using the minimum CVR ≥ .5, it was determined that these items should remain on the PMI: Item 1 (CVR = .9), Item 8 (CVR = .8), Item 9 (CVR = .9), and Item 12 (CVR = .9). The use of Siegle’s (2017) Reliability Calculator determined the Cronbach’s α level for the PMI to be 0.83, indicating adequate internal consistency. Additionally, a split-half (odd-even) correlation was completed with the Spearman-Brown adjustment of 0.88, indicating high reliability (Siegle, 2017).

Pilot Summary
     The focus of the pilot study was to ensure effective implementation of the proposed research protocol following each respondent’s appointment at the CAC research site. The pilot was implemented to ensure research procedures did not interfere with typical appointments and standard procedures at the CAC. Participation in the PMI pilot was voluntary and no compensation was provided for respondents.

     Sample. The study used a purposeful sample of children at a local, nationally accredited CAC in the Midwest; both the child and the child’s legal guardian agreed to participate. Because of the expected integration of PM with other forms of abuse, this population was selected to help create an understanding of how PM is experienced specifically with co-occurring cases of maltreatment. Respondents were children who (a) had an open CM case with social services and/or law enforcement, (b) were scheduled for an appointment at the CAC, and (c) were between the ages of 8 and 17.

     Measures. The two measures implemented in this study were the developing PMI and the TSCC-SF. At the time of data collection, CAC staff implemented the TSCC-SF as a screening tool for referral services during CAC victim appointments. To ensure the research process did not interfere with chain-of-custody procedures, collected investigative testimony, or physical evidence that was obtained, the PMI was administered only after all normally scheduled CAC procedures were followed during appointments.

     PMI. The current version of the PMI is a self-report measure that consists of 16 items on a 4-point Likert-type scale that mirrors the language of the TSCC-SF (0 = never to 3 = almost all the time). Respondents typically needed 5 minutes complete the PMI. Sample items from the PMI included questions like: “How often have you been told or made to feel like you are not important or unlovable?” The full instrument is not provided for use in this publication to ensure the PMI is not misused, as refinement of the PMI is still in progress.

     TSCC-SF. In addition to the PMI, Boughn gathered data from the TSCC-SF (Briere & Wherry, 2016) because of its widespread use among clinicians to efficiently assess for sexual concerns, suicidal ideation frequency, and general trauma symptoms such as post-traumatic stress, depression, anger, disassociation, and anxiety (Wherry et al., 2013). The TSCC-SF measures a respondent’s frequency of perceived experiences and has been successfully implemented with children as young as 8 years old (Briere, 1996). The 20-item form uses a 4-point Likert-type scale (0 = never to 3 = almost all the time) composed of general trauma and sexual concerns subscales. The TSCC-SF has demonstrated high internal consistency and alpha values in the good to excellent ranges; it also has high intercorrelations between sexual concerns and other general trauma scales (Wherry & Dunlop, 2018).

     Procedures. Respondents were recruited during their scheduled CAC appointment time. Each investigating agency (law enforcement or social services) scheduled a CAC appointment in accordance with an open maltreatment case. At the beginning of each respondent’s appointment, Boughn provided them with an introduction and description of the study. This included the IRB approvals from the hospital and university, an explanation of the informed consent and protected health information (PHI) authorization, and assent forms. Respondents aged 12 and older were asked to read and review the informed consent document with their legal guardian; respondents aged from 8 to 11 were provided an additional assent document to read. Respondents were informed they could stop the study at any time. After each respondent and legal guardian consented, respondents proceeded with their CAC appointment.

Typical CAC appointments consisted of a forensic interview, at times a medical exam, and administration of the TSCC-SF to determine referral needs. After these steps were completed, Boughn administered the PMI to those who agreed to participate in this research study. Following the completion of the TSCC-SF, respondents were verbally reminded of the study and asked if they were still willing to participate by completing the PMI. Willing respondents completed the PMI; afterward, Boughn asked respondents if they were comfortable leaving the assessment room. In the event the respondent voiced additional concerns of maltreatment during the PMI administration, Boughn made a direct report to the respondent’s investigator (i.e., law enforcement officer or social worker assigned to the respondent’s case).

Boughn accessed each respondent’s completed TSCC-SF from their electronic health record in accordance with the PHI authorization and consent after the respondent’s appointment. Data completed on the TSCC-SF allowed Boughn to gather information related to sexual concerns, suicidal ideation, and trauma symptomology. Data gathered from the TSCC-SF were examined with each respondent’s PMI responses.

     Results. Respondents were 21 children (15 female, six male) with age ranges from 8 to 17 years with a median age of 12 years. Respondents described themselves as White (47.6%), Biracial (14.2%), Multiracial (14.2%), American Indian/Alaskan Native (10.0%), Black (10.0%), and Hispanic/Latino (5.0%). CM allegations for the respondents consisted of allegations of sexual abuse (86.0%), physical abuse (10.0%), and neglect (5.0%).

Every respondent’s responses were included in the analyses to ensure all maltreatment situations were considered. The reliability of the PMI observed in the pilot sample (N = 21) demonstrated high internal consistency with all 16 initial items (α = .88). The average total score on the PMI in the pilot was 13.29, with respondents’ scores ranging from 1 to 30. A Pearson correlation indicated total scores for the PMI and General Trauma Scale scores (reported on the TSCC-SF) were significantly correlated (r = .517, p < .05).

Study 2: Full Testing of the PMI
     The next phase of research proceeded with the collection of a larger data sample (N = 166) to explore the item construct validity and internal reliability (Siyez et al., 2020). Study procedures, data collection, and data storage followed in the pilot study were also implemented with the larger sample. Boughn maintained tracking of respondents who did not want to participate in the study or were unable to because of cognitive functioning level, emergency situations, and emotional dysregulation concerns.

Sample
     Based on a power analysis performed using the Raosoft (2004) sample size calculator, the large sample study required a minimum of 166 respondents for statistical significance (Ali, 2012; Heppner et al., 2016). The sample size was expected to account for a 10% margin of error and a 99% confidence level. The calculation of a 99% confidence interval was used to ensure the number of respondents could effectively represent the population accessed within the CAC based on the data from the CM Report (DHHS, 2018). Large sample population data was gathered between September 2018 and May 2019.

Measures
     The PMI and TSCC-SF were also employed in Study 2 because of their successful implementation in the pilot. Administration of the TSCC-SF ensured a normed and standardized measure could aid in providing context to the information gathered on the PMI. No changes were made to the PMI or TSCC-SF measures following the review of procedures and analyses in the pilot.

Procedures
     Recruitment and data collection/analyses processes mirrored that of the pilot study. Voluntary respondents were recruited at the CAC during their scheduled appointments. Respondents completed an informed consent, child assent, PHI authorization form, TSCC-SF, and PMI. Following the completion of data collection, Boughn completed data entry in the electronic health record to de-identify and analyze the results.

Results

Demographics
     All data were analyzed using Statistical Package for the Social Sciences version 24 (SPSS-24). Initial data evaluation consisted of exploration of descriptive statistics, including demographic and criteria-based information related to respondents’ identities and case details. Respondents were between 8 to 17 years of age (M = 12.39) and primarily female (73.5%, n = 122), followed by male (25.3%, n = 42). Additionally, two respondents (n = 2) reported both male and female gender identities. Racial identities were marked by two categories: White (59.6%, n = 99) and Racially Diverse (40.4%, n = 67) respondents. The presenting maltreatment concerns and the child’s relationship to the offender are outlined in Table 1 and Table 2, respectively.

Reliability and Validity of the PMI
     The reliability of the PMI observed in its implementation in Study 2 (N = 166) showed even better internal consistency with all 16 initial items (α =.91) than observed in the pilot. Using the Spearman-Brown adjustment (Warner, 2013), split-half reliability was calculated, indicating high internal reliability (.92). Internal consistencies were calculated using gender identity and age demographic variables (see Table 3).

 

Table 1

Child Maltreatment Allegation by Type (N = 166)

Allegation f Rel f cf %
Sexual Abuse 113 0.68 166 68.07
Physical Abuse  29 0.17 53 17.47
Neglect  14 0.08 24   8.43
Multiple Allegations    6 0.04 10   3.61
Witness to Violence    3 0.02   4   1.81
Kidnapping    1 0.01   1   0.60

Note. Allegation type reported at initial appointment scheduling

 

Table 2

Identified Offender by Relationship to Victim (N = 166)

Offender Relationship f Rel f cf %
Other Known Adult 60 0.36 166 36.14
Parent 48 0.29 106 28.92
Other Known Child (≤ age 15 years) 15 0.09  58   9.04
Sibling-Child (≤ age 15 years) 10 0.06  43   6.02
Unknown Adult   9 0.05  33     5.42
Step-Parent   8 0.05  24   4.82
Multiple Offenders   6 0.04  16   3.61
Grandparent   6 0.04  10   3.61
Sibling-Adult (≥ age 16 years)   3 0.02   4   1.81
Unknown Child (≤ age 15 years)   1 0.01   1   0.60

Note. Respondent knew the offender (n =156); Respondent did not know offender (n =10)

 

Table 3

Internal Consistency Coefficients (α) by Gender Identity and Age (N = 166)

Gender n α M SD
 Female 122 0.90 13.2   9.1
 Male   42 0.94 13.5 11.0
 Male–Female    2 0.26   8.5  2.5
Age
 8–12 83 0.92 12.75 10.06
 13–17 83 0.90 13.69   9.01

Note. SD = Standard Deviation; M = Mean

 

Respondents Demographic Characteristics and PM Experiences
For Research Question (RQ) 1 and RQ2, descriptive data were used to generate frequencies and determine the impact of demographic characteristics on average PMI score. To explore this further in RQ1, one-way ANOVAs were completed for the variables of age, gender, racial identity, allegation type, and offender relationships. No significant correlations were found between demographic variables and the PMI items. On average, respondents reported a frequency score of 13.5 (M = 13.5, SD = 9.5) on the PMI. Eight respondents (5%) endorsed no frequency of PM while 95% (N = 158) experienced PM.

Co-Occurrence of PM With Other Forms of Maltreatment
     For RQ3, frequency and descriptive data were generated, revealing average age rates of PM reported by maltreatment type. Varying sample representations were discovered in each form of maltreatment (see Table 4). Clear evidence was found that PM co-occurs with each form of maltreatment type; however, how each form of maltreatment interacts with PM is currently unclear given the multiple dimensions of each maltreatment case including, but not limited to, severity, frequency, offender, and victim characteristics.

 

Table 4

Descriptive and Frequency Data for Co-Occurrence of PM (N = 166)

Allegation n M SD 95% CI
Sexual Abuse 113 13.04   9.01 [11.37, 14.72]
Physical Abuse   29 12.45 10.53   [8.44, 16.45]
Neglect   14 14.57 12.16   [7.55, 21.60]
Multiple Allegations    5 17.40   8.88   [6.38, 28.42]
Witness to Violence    3   7.67   5.03  [–4.84, 20.17]
Kidnapping    1 n/a n/a Missing

Note. CI = Confidence Interval; SD = Standard Deviation; M = Mean; n/a = not applicable

 

PM Frequency and General Trauma Symptoms
     For RQ4, Pearson’s correlation was used to calculate frequency score relationships between the PMI and TSCC-SF. There was a statistically significant relationship between the PMI and total frequency of general trauma symptoms on the TSCC-SF [r(164) = .78, p < .01, r² = .61] (Sullivan & Feinn, 2012). Cohen’s d, calculated from the means for each item as well as the pooled standard deviation, indicated a small effect relationship (d = .15) between general trauma and PMI frequencies (see Figure 1).

 

Figure 1

Correlation Between PMI and TSCC-SF General Trauma Subscale

Note. Scores were endorsed by respondents’ self-reports.

 

Child Suicidal Ideation Reports and the PMI
     Following a review of the findings of Thompson et al. (2005) and Wherry et al. (2013) that children who reported experiencing CM also experienced suicidal ideation, Boughn performed an additional two-way ANOVA that examined the effect of suicidal ideation on the PMI total score. A significant relationship—F(1, 164) = 49.52, p < .01, η2 = .23—between respondents’ PMI scores and thoughts of suicide was found. Respondents who did not report thoughts of suicide (59.0%, n = 98) indicated lower rates of PM (M = 9.37, SD = 7.97) compared to children who did report thoughts of suicide (41.0%, n = 68, M = 18.77, SD = 9.12). A preliminary review of this finding demonstrates the severity of PM’s impact on child victims.

Discussion

This study was designed with the aim of developing a tool to support accurate identification of PM among children and adolescents. Findings from its first large-scale implementation provide a foundational view to the occurrence of PM in terms of demographic characteristics, comorbidity of PM with other forms of abuse, and the relationship between PM and trauma. The analyses yielded both expected and unexpected results based on the extant research.

PM and Demographic Characteristics
Race
     There was no significant effect when exploring the data related to racial demographics and PM. The respondent sample closely reflected the geographical area’s known racial demographics at the time of the study, reflecting a population approaching 80% White with residents of all other known races below 5% for each racial group (U.S. Census Bureau, 2020). Although researchers (Dakil et al., 2011) anticipated children identifying as racial minorities would be included in the representation of CM reports, evidence from this study potentially reveals a greater than expected gap in reporting for minority-race populations (Bernard & Harris, 2018; Font & Maguire-Jack, 2015). This suggests that there may be additional, unidentified barriers influencing the reporting of maltreatment among minority-race populations.

Gender
     A lack of gender identity representation was evident in the data, consistent with prior research (Sivagurunathan et al., 2019). Respondents who self-identified with both male and female gender identities (1.2%) and as male (25.3%) were represented less frequently compared to female respondents (73.5%). This is not inherently a limitation of this study, as research shows that just 10% of males in the United States report their sexual abuse (Sivagurunathan et al., 2019). People who identify as male may face harmful cultural messages that enhance negative stigma for victims of abuse, causing increased feelings of weakness or vulnerability (Alaggia & Mishna, 2014). This finding may support claims that male trauma survivors feel stigmatized and report their experiences less frequently (Easton, 2012).

Additionally, children who identify outside traditional gender binary norms and definitions need more access to inclusive representation on screening assessments. Assessments like the TSCC-SF may be using antiquated gender- or biological sex–normed checkboxes, which leave certain groups underrepresented in research studies (Neukrug & Fawcett, 2015). These practices may present inaccurate findings, inadvertently reinforce discriminatory expectations, and generate inaccurate referrals. Non-binary youth encounter barriers that may compound their ability to effectively access supports in their daily life related to coming out, social violence, lack of peer and/or adult acceptance, discrimination, isolation, higher rates of suicide, and lack of representation in mainstream society (Bialer & McIntosh, 2016; Zimman, 2009). In this study, representation of non-binary respondents, specifically those who reported both male and female gender identities, was reported; this warrants further exploration to assess barriers among non-binary gender youth and their experiences with child PM (Bos et al., 2019).

Offender Relationships
     Frequency data for a child’s relationship with the offender were not found to be significant either for known offenders (M = 13.35) or unknown offenders (M = 11.2). In this study, 94% of the respondents already knew their offender (n = 156). This finding is consistent with previous research that has found that although child abduction and stranger danger are real phenomena, children are more likely to experience CM as a result of relationships with familiar individuals (Walsh & Brandon, 2011).

Co-Occurrence of PM With Other Abuse
     Only eight respondents (5%) endorsed no frequency of PM; the average total PM frequency rate for respondents in this study was 13.5 out of a possible 48, indicating extreme severity. In this study, we found evidence that PM is a co-occurring experience for children with open maltreatment cases, yet clinicians still lack formal, valid assessments to determine PM alone. Our findings support the National Children’s Alliance’s (NCA; 2016) call for clinicians to follow practice guidelines in accordance with state and national guidelines as they relate to mandatory reporting of CM concerns and determination of whether PM plus other forms of maltreatment may be present for child victims seeking services.

Comorbidity of PM and Trauma
     PM-related experiences on the PMI and general trauma symptoms from the TSCC-SF warrant discussion. The PMI illustrated a significant relationship with the TSCC-SF general trauma subscale (Briere & Wherry, 2016). More than half (61%) of the variance on the PMI was connected to general trauma symptoms, suggesting that higher rates of PM experiences may increase trauma-related symptoms. For example, previous researchers have found adverse childhood experiences and signs of trauma-related symptoms lead to serious mental health diagnoses, early mortality, and/or significant biological health risks in children (Tyrka et al., 2013; Vachon et al., 2015; Zimmerman & Mercy, 2010). Further exploration to determine if and how PM influences other trauma-related symptoms in children throughout their life span would expand upon the results of this study.

Suicidal Ideation
     Finally, our data revealed a significant effect between respondent endorsement of suicidal ideation and PMI total scores. PM experiences accounted for 23% of the variance for children who reported thoughts of suicide (41%, n = 68) compared to those who did not report thoughts of suicide (59%, n = 98). This finding is consistent with prior research exploring children’s experiences with maltreatment and suicidal thoughts (Thompson et al., 2005; Wherry et al., 2013).

Limitations
     This study has several limitations. First, by developing the PMI using national definitions, some regional and localized nuances were not considered. Second, data collected for this study were from a single Midwest CAC; thus, the data are limited in geographic generalizability. Third, the majority of respondents were White, and a more diverse sample would have been more representative of the region in which data were collected. Fourth, 99% of respondents identified as either male or female and may reflect an underrepresentation of non-binary or gender fluid youth in the results of this study. Fifth, this study relied heavily on quantitative data, which limited the ability to analyze each individual’s experiences with PM as they might describe from their unique perspectives.

Implications for Research and Practice
     The results of this study provide several areas for future research. While the PMI demonstrated good internal consistency across all items (α =.91), more research with diverse populations across the United States is needed. Research from other geographical locations may demonstrate how reporting patterns for PM interact with ethnicity, culture, and elements of social expectations (Spinazzola et al., 2014).

The initial results of this study indicate the PMI may be a useful tool for children to report PM experiences in CAC settings; however, future research at other CACs and similar treatment facilities is needed to determine the PMI’s true utility and scalability. Future analysis (i.e., exploratory factor analysis and confirmatory factor analysis) of the PMI may also identify factors and help refine the instrument.

More research with the PMI can expand researchers’ knowledge of PM and services needed to help children. Working with other CACs, child protection professionals, and the NCA may help bridge current gaps in interdisciplinary assessment and care and establish a stable and comprehensive understanding of PM (López et al., 2015). Furthermore, understanding how CACs are equipped to identify and handle PM cases may provide useful insights to help improve services for children in need. Although some CACs may have a variety of professionals working in specific roles, some CACs may be understaffed, causing staff to take on multiple and overlapping roles. It is important to understand if and how different combinations of trained professionals influence children reporting PM (Hart & Glaser, 2011; NCA, 2016).

More research with the PMI is needed for refinement and to ensure the instrument is not misused. Releasing the PMI at this stage to clinicians and researchers without a fully developed assessment manual may lead to inappropriate or ineffective administration of the PMI and potentially unethical practice that could place children at risk. Future research and refinement of the PMI may provide clinicians and researchers a reliable and valid tool that is grounded in consistent theory and practice.

Conclusion

The PMI was developed to assess child PM and offers researchers and clinicians useful findings. In supporting research (Arslan, 2017; Bernstein et al., 2013; Raparia et al., 2016), child PM is a serious and often harmful combination of experiences that requires professional intervention (APSAC, 2019). For children reporting PM experiences, the PMI may help mental health and other care providers determine which services are needed. Findings from this study suggest differences in demographic variables are minimal for PM. Overall PMI scores were correlated to the general trauma subscale on the TSCC-SF, and the PMI revealed higher rates of PM for children experiencing suicidal ideation. The findings are the beginning of a measure designed to illustrate the depth and frequency of PM for children. With the PMI, early PM intervention becomes possible for a once invisible form of maltreatment.

Conflict of Interest and Funding Disclosure
Data collected and content shared in this study
were part of a dissertation study, which was
awarded the 2020 Dissertation Excellence Award
by the National Board for Certified Counselors.
The Psychological Maltreatment Inventory (PMI)
items were not released in this publication to protect
victims of child maltreatment and to ensure future
publications can address comprehensive revisions
made to the PMI.

 

References

Ahern, E. C., Hershkowitz, I., Lamb, M. E., Blasbalg, U., & Winstanley, A. (2014). Support and reluctance in the pre-substantive phase of alleged child abuse victim investigative interviews: Revised versus standard NICHD protocols. Behavioral Sciences & the Law, 32(6), 762–774. https://doi.org/10.1002/bsl.2149

Alaggia, R., & Mishna, F. (2014). Self psychology and male child sexual abuse: Healing relational betrayal. Clinical Social Work Journal, 42(1), 41–48. https://doi.org/10.1007/s10615-013-0453-2

Ali, S. A. (2012). Sample size calculation and sampling techniques. Journal of the Pakistan Medical Association, 62(6), 624–626. https://jpma.org.pk/PdfDownload/3482

American Professional Society on the Abuse of Children. (2019). APSAC practice guidelines: The investigation and determination of suspected psychological maltreatment of children and adolescents. https://bit.ly/3jI7AhJ

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).

Arslan, G. (2017). Psychological maltreatment, coping strategies, and mental health problems: A brief and effective measure of psychological maltreatment in adolescents. Child Abuse & Neglect, 68, 96–106. https://doi.org/10.1016/j.chiabu.2017.03.023

Ayre, C., & Scally, A. J. (2014). Critical values for Lawshe’s content validity ratio: Revisiting the original methods of calculation. Measurement and Evaluation in Counseling and Development, 47(1), 79–86. https://doi.org/10.1177%2F0748175613513808

Bernard, C., & Harris, P. (2018). Serious case reviews: The lived experience of Black children. Child & Family Social Work, 24(2), 256–263. https://doi.org/10.1111/cfs.12610

Bernstein, R. E., Measelle, J. R., Laurent, H. K., Musser, E. D., & Ablow, J. C. (2013). Sticks and stones may break my bones but words relate to adult physiology? Child abuse experience and women’s sympathetic nervous system response while self-reporting trauma. Journal of Aggression, Maltreatment & Trauma, 22(10), 1117–1136. https://doi.org/10.1080/10926771.2013.850138

Bialer, P. A., & McIntosh, C. A. (2016). Discrimination, stigma, and hate: The impact on the mental health and well-being of LGBT people. Journal of Gay & Lesbian Mental Health, 20(4), 297–298. https://doi.org/10.1080/19359705.2016.1211887

Bos, H., de Haas, S., & Kuyper, L. (2019). Lesbian, gay, and bisexual adults: Childhood gender nonconformity, childhood trauma, and sexual victimization. Journal of Interpersonal Violence, 34(3), 496–515. https://doi.org/10.1177%2F0886260516641285

Briere, J. (1996). Trauma Symptom Checklist for Children (TSCC), professional manual. Psychological Assessment Resources.

Briere, J., & Wherry, J. (2016). Development and validation of the TSCC Screening Form (TSCC-SF) and TSCYC Screening Form (TSCYC-SF). Psychological Assessment Resources.

Brown, S. M., Bender, K., Orsi, R., McCrae, J. S., Phillips, J. D., & Rienks, S. (2019). Adverse childhood experiences and their relationship to complex health profiles among child welfare–involved children: A classification and regression tree analysis. Health Services Research, 54(4), 902–911. https://doi.org/10.1111/1475-6773.13166

Centers for Disease Control. (2012). Child abuse and neglect cost the United States $124 billion [Press release]. https://bit.ly/3jYbpAF

Dakil, S. R., Cox, M., Lin, H., & Flores, G. (2011). Racial and ethnic disparities in physical abuse reporting and Child Protective Services interventions in the United States. Journal of the National Medical Association, 103(9–10), 926–931. https://doi.org/10.1016/S0027-9684(15)30449-1

Easton, S. D. (2012). Disclosure of child sexual abuse among adult male survivors. Clinical Social Work Journal, 41, 344–355. https://doi.org/10.1007/s10615-012-0420-3

Fang, X., Brown, D. S., Florence, C. S., & Mercy, J. A. (2012). The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse & Neglect, 36(2), 156–165. https://doi.org/10.1016/j.chiabu.2011.10.006

Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., Koss, M. P., & Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) study. American Journal of Preventive Medicine, 14(4), 245–258. https://doi.org/10.1016/S0749-3797(98)00017-8

Field, T. A., Jones, L. K., & Russell-Chapin, L. A. (Eds.). (2017). Neurocounseling: Brain-based clinical approaches. American Counseling Association.

Font, S. A., & Maguire-Jack, K. (2015). Decision-making in Child Protective Services: Influences at multiple levels of the social ecology. Child Abuse & Neglect, 47, 70–82. https://doi.org/10.1016/j.chiabu.2015.02.005

Glaser, D. (2002). Emotional abuse and neglect (psychological maltreatment): A conceptual framework. Child Abuse & Neglect, 26(6–7), 697–714. https://doi.org/10.1016/S0145-2134(02)00342-3

Greenfield, E. A., & Marks, N. F. (2010). Identifying experiences of physical and psychological violence in childhood that jeopardize mental health in adulthood. Child Abuse & Neglect, 34(3), 161–171. https://doi.org/10.1016/j.chiabu.2009.08.012

Hart, S. N., & Glaser, D. (2011). Psychological maltreatment – Maltreatment of the mind: A catalyst for advancing child protection toward proactive primary prevention and promotion of personal well-being. Child Abuse & Neglect, 35(10), 758–766. https://doi.org/10.1016/j.chiabu.2011.06.002

Heppner, P. P., Wampold, B. E., Owen, J., Thompson, M. N., & Wang, K. T. (2016). Research design in counseling (4th ed.). Cengage.

Iwaniec, D. (2006). The emotionally abused and neglected child: Identification, assessment and intervention: A practice handbook (2nd ed.). Wiley.

Klika, J. B., & Conte, J. R. (Eds.). (2017). The APSAC handbook on child maltreatment (4th ed.). SAGE.

Lawshe, C. H. (1975). A quantitative approach to content validity. Personnel Psychology, 28(4), 563–575. https://doi.org/10.1111/j.1744-6570.1975.tb01393.x

López, M., Fluke, J. D., Benbenishty, R., & Knorth, E. J. (2015). Commentary on decision-making and judgments in child maltreatment prevention and response: An overview. Child Abuse & Neglect, 49, 1–11. https://doi.org/10.1016/j.chiabu.2015.08.013

Maguire, S. A., Williams, B., Naughton, A. M., Cowley, L. E., Tempest, V., Mann, M. K., Teague, M., & Kemp, A. M. (2015). A systematic review of the emotional, behavioural and cognitive features exhibited by school-aged children experiencing neglect or emotional abuse. Child: Care, Health and Development, 41(5), 641–653. https://doi.org/10.1111/cch.12227

Marshall, N. A. (2012). A clinician’s guide to recognizing and reporting parental psychological maltreatment of children. Professional Psychology: Research and Practice, 43(2), 73–79. https://doi.org/10.1037/a0026677

Mills, R., Scott, J., Alati, R., O’Callaghan, M., Najman, J. M., & Strathearn, L. (2013). Child maltreatment and adolescent mental health problems in a large birth cohort. Child Abuse & Neglect, 37(5), 292–302. https://doi.org/10.1016/j.chiabu.2012.11.008

National Children’s Alliance. (2016). Putting standards into practice: A guide to implementing the 2017 standards for accredited members (revised 2016). http://www.nationalchildrensalliance.org/wp-content/uploads/2015/06/NCA2017-StandardsIntoPractice-web.pdf

Neukrug, E. S., & Fawcett, R. C. (2015). The essentials of testing and assessment: A practical guide for counselors, social workers, and psychologies, enhanced (3rd ed.). Cengage.

Peterson, C., Florence, C., & Klevens, J. (2018). The economic burden of child maltreatment in the United States, 2015. Child Abuse & Neglect, 86, 178–183.
https://doi.org/10.1016/j.chiabu.2018.09.018

Porges, S. W. (2011). The polyvagal theory: Neurophysiological foundations of emotions, attachment, communication, and self-regulation. W. W. Norton.

Raosoft. (2004). Sample size calculator. http://www.raosoft.com/samplesize.html

Raparia, E., Coplan, J. D., Abdallah, C. G., Hof, P. R., Mao, X., Mathew, S. J., & Shungu, D. C. (2016). Impact of childhood emotional abuse on neocortical neurometabolites and complex emotional processing in patients with generalized anxiety disorder. Journal of Affective Disorders, 190, 414–423. https://doi.org/10.1016/j.jad.2015.09.019

Readable. (n.d.). https://readable.com

Siegle, R. (2017). Educational research basics: Excel spreadsheet to calculate instrument reliability estimates. https://researchbasics.education.uconn.edu/excel-spreadsheet-to-calculate-instrument-reliability-estimates

Sivagurunathan, M., Orchard, T., & Evans, M. (2019). Barriers to utilization of mental health services amongst male child sexual abuse survivors: Service providers’ perspective. Journal of Child Sexual Abuse, 28(7), 819–839. https://doi.org/10.1080/10538712.2019.1610823

Siyez, D. M., Esen, E., Seymenler, S., & Öztürk, B. (2020). Development of wellness scale for emerging adults: Validity and reliability study. Current Psychology.
https://doi.org/10.1007/s12144-020-00672-w

Slep, A. M. S., Heyman, R. E., & Foran, H. M. (2015). Child maltreatment in DSM-5 and ICD-11. Family Process, 54(1), 17–32. https://doi.org/10.1111/famp.12131

Spinazzola, J., Hodgdon, H., Liang, L.-J., Ford, J. D., Layne, C. M., Pynoos, R., Briggs, E. C., Stolbach, B., & Kisiel, C. (2014). Unseen wounds: The contribution of psychological maltreatment to child and adolescent mental health and risk outcomes. Psychological Trauma: Theory, Research, Practice, and Policy, 6(Suppl 1), S18–S28. https://doi.org/10.1037/a0037766

Sullivan, G. M., & Feinn, R. (2012). Using effect size—or why the p value is not enough. Journal of Graduate Medical Education, 4(3), 279–282. https://doi.org/10.4300/JGME-D-12-00156.1

Thompson, R., Briggs, E., English, D. J., Dubowitz, H., Lee, L.-C., Brody, K., Everson, M. D., & Hunter, W. M. (2005). Suicidal ideation among 8-year-olds who are maltreated and at risk: Findings from the LONGSCAN studies. Child Maltreatment, 10(1), 26–36.  https://doi.org/10.1177%2F1077559504271271

Tonmyr, L., Draca, J., Crain, J., & MacMillian, H. L. (2011). Measurement of emotional/psychological child maltreatment: A review. Child Abuse & Neglect, 35(10), 767–782.
https://doi.org/10.1016/j.chiabu.2011.04.011

Tyrka, A. R., Burgers, D. E., Philip, N. S., Price, L. H., & Carpenter, L. L. (2013). The neurobiological correlates of childhood adversity and implications for treatment. Acta Psychiatrica Scandinavica, 128(6), 434–447. https://doi.org/10.1111/acps.12143

U.S. Census Bureau. (2020). Quick facts. https://www.census.gov

U.S. Department of Health & Human Services. (2018). Child maltreatment 2016 (27th ed.). https://www.acf.hhs.gov/sites/default/files/documents/cb/cm2016.pdf

Vachon, D. D., Krueger, R. F., Rogosch, F. A., & Cicchetti, D. (2015). Assessment of the harmful psychiatric and behavioral effects of different forms of child maltreatment. JAMA Psychiatry, 72(11), 1135–1142. https://doi.org/10.1001/jamapsychiatry.2015.1792

van der Kolk, B. (2014). The body keeps the score: Brain, mind, and body in the healing of trauma. Penguin Books.

van Harmelen, A.-L., Elzinga, B. M., Kievit, R. A., & Spinhoven, P. (2011). Intrusions of autobiographical memories in individuals reporting childhood emotional maltreatment. European Journal of Psychotraumatology, 2(1), 7336. https://doi.org/10.3402/ejpt.v2i0.7336

van Harmelen, A.-L., van Tol, M.-J., van der Wee, N. J. A., Veltman, D. J., Aleman, A., Spinhoven, P., van Buchem, M. A., Zitman, F. G., Penninx, B. W. J. H., & Elzinga, B. M. (2010). Reduced medial prefrontal cortex volume in adults reporting childhood emotional maltreatment. Biological Psychiatry, 68(9), 832–838. https://doi.org/10.1016/j.biopsych.2010.06.011

Walsh, K., & Brandon, L. (2011). Their children’s first educators: Parents’ views about child sexual abuse prevention education. Journal of Child and Family Studies, 21, 734–746.
https://doi.org/10.1007/s10826-011-9526-4

Warner, R. M. (2013). Applied statistics: From bivariate through multivariate techniques (2nd ed.). SAGE.

Wherry, J. N., Baldwin, S., Junco, K., & Floyd, B. (2013). Suicidal thoughts/behaviors in sexually abused children. Journal of Child Sexual Abuse, 22(5), 534–551. https://doi.org/10.1080/10538712.2013.800938

Wherry, J. N., & Dunlop, C. E. (2018). TSCC and TSCYC screening forms in a clinical sample: Reliability, validity, and creating local clinical norms. Child Maltreatment, 23(1), 74–84.
https://doi.org/10.1177%2F1077559517725207

Zimman, L. (2009). ‘The other kind of coming out’: Transgender people and the coming out narrative genre. Gender and Language, 3(1), 53–80. https://doi.org/10.1558/genl.v3i1.53

Zimmerman, F., & Mercy, J. (2010). A better start: Child maltreatment prevention as a public
health priority. Zero to Three, 30(5), 4–10.

 

Alison M. Boughn, PhD, NCC, LIMHP (NE), LMHC (IA), LPC-MH (SD), ATR-BC, QMHP, TF-CBT, is an assistant professor and counseling department chair at Wayne State College. Daniel A. DeCino, PhD, NCC, LPC, is an assistant professor and Interim Program Coordinator at the University of South Dakota. Correspondence may be addressed to Alison M. Boughn, Wayne State College, 1111 Main Street, Wayne, NE 68787, albough1@wsc.edu.

Validation of the Adapted Response to Stressful Experiences Scale (RSES-4) Among First Responders

Warren N. Ponder, Elizabeth A. Prosek, Tempa Sherrill

 

First responders are continually exposed to trauma-related events. Resilience is evidenced as a protective factor for mental health among first responders. However, there is a lack of assessments that measure the construct of resilience from a strength-based perspective. The present study used archival data from a treatment-seeking sample of 238 first responders to validate the 22-item Response to Stressful Experiences Scale (RSES-22) and its abbreviated version, the RSES-4, with two confirmatory factor analyses. Using a subsample of 190 first responders, correlational analyses were conducted of the RSES-22 and RSES-4 with measures of depressive symptoms, post-traumatic stress, anxiety, and suicidality confirming convergent and criterion validity. The two confirmatory analyses revealed a poor model fit for the RSES-22; however, the RSES-4 demonstrated an acceptable model fit. Overall, the RSES-4 may be a reliable and valid measure of resilience for treatment-seeking first responder populations.

Keywords: first responders, resilience, assessment, mental health, confirmatory factor analysis

 

     First responder populations (i.e., law enforcement, emergency medical technicians, and fire rescue) are often repeatedly exposed to traumatic and life-threatening conditions (Greinacher et al., 2019). Researchers have concluded that such critical incidents could have a deleterious impact on first responders’ mental health, including the development of symptoms associated with post-traumatic stress, anxiety, depression, or other diagnosable mental health disorders (Donnelly & Bennett, 2014; Jetelina et al., 2020; Klimley et al., 2018; Weiss et al., 2010). In a systematic review, Wild et al. (2020) suggested the promise of resilience-based interventions to relieve trauma-related psychological disorders among first responders. However, they noted the operationalization and measure of resilience as limitations to their intervention research. Indeed, researchers have conflicting viewpoints on how to define and assess resilience. For example, White et al. (2010) purported popular measures of resilience rely on a deficit-based approach. Counselors operate from a strength-based lens (American Counseling Association [ACA], 2014) and may prefer measures with a similar perspective. Additionally, counselors are mandated to administer assessments with acceptable psychometric properties that are normed on populations representative of the client (ACA, 2014, E.6.a., E.7.d.). For counselors working with first responder populations, resilience may be a factor of importance; however, appropriately measuring the construct warrants exploration. Therefore, the focus of this study was to validate a measure of resilience with strength-based principles among a sample of first responders.

Risk and Resilience Among First Responders

In a systematic review of the literature, Greinacher et al. (2019) described the incidents that first responders may experience as traumatic, including first-hand life-threatening events; secondary exposure and interaction with survivors of trauma; and frequent exposure to death, dead bodies, and injury. Law enforcement officers (LEOs) reported that the most severe critical incidents they encounter are making a mistake that injures or kills a colleague; having a colleague intentionally killed; and making a mistake that injures or kills a bystander (Weiss et al., 2010). Among emergency medical technicians (EMTs), critical incidents that evoked the most self-reported stress included responding to a scene involving family, friends, or others to the crew and seeing someone dying (Donnelly & Bennett, 2014). Exposure to these critical incidents may have consequences for first responders. For example, researchers concluded first responders may experience mental health symptoms as a result of the stress-related, repeated exposure (Jetelina et al., 2020; Klimley et al., 2018; Weiss et al., 2010). Moreover, considering the cumulative nature of exposure (Donnelly & Bennett, 2014), researchers concluded first responders are at increased risk for post-traumatic stress disorder (PTSD), depression, and generalized anxiety symptoms (Jetelina et al., 2020; Klimley et al., 2018; Weiss et al., 2010). Symptoms commonly experienced among first responders include those associated with post-traumatic stress, anxiety, and depression.

In a collective review of first responders, Kleim and Westphal (2011) determined a prevalence rate for PTSD of 8%–32%, which is higher than the general population lifetime rate of 6.8–7.8 % (American Psychiatric Association [APA], 2013; National Institute of Mental Health [NIMH], 2017). Some researchers have explored rates of PTSD by specific first responder population. For example, Klimley et al. (2018) concluded that 7%–19% of LEOs and 17%–22% of firefighters experience PTSD. Similarly, in a sample of LEOs, Jetelina and colleagues (2020) reported 20% of their participants met criteria for PTSD.

Generalized anxiety and depression are also prevalent mental health symptoms for first responders. Among a sample of firefighters and EMTs, 28% disclosed anxiety at moderate–severe and several levels (Jones et al., 2018). Furthermore, 17% of patrol LEOs reported an overall prevalence of generalized anxiety disorder (Jetelina et al., 2020). Additionally, first responders may be at higher risk for depression (Klimley et al., 2018), with estimated prevalence rates of 16%–26% (Kleim & Westphal, 2011). Comparatively, the past 12-month rate of major depressive disorder among the general population is 7% (APA, 2013). In a recent study, 16% of LEOs met criteria for major depressive disorder (Jetelina et al., 2020). Moreover, in a sample of firefighters and EMTs, 14% reported moderate–severe and severe depressive symptoms (Jones et al., 2018). Given these higher rates of distressful mental health symptoms, including post-traumatic stress, generalized anxiety, and depression, protective factors to reduce negative impacts are warranted.

Resilience
     Broadly defined, resilience is “the ability to adopt to and rebound from change (whether it is from stress or adversity) in a healthy, positive and growth-oriented manner” (Burnett, 2017, p. 2). White and colleagues (2010) promoted a positive psychology approach to researching resilience, relying on strength-based characteristics of individuals who adapt after a stressor event. Similarly, other researchers explored how individuals’ cognitive flexibility, meaning-making, and restoration offer protection that may be collectively defined as resilience (Johnson et al., 2011).

A key element among definitions of resilience is one’s exposure to stress. Given their exposure to trauma-related incidents, first responders require the ability to cope or adapt in stressful situations (Greinacher et al., 2019). Some researchers have defined resilience as a strength-based response to stressful events (Burnett, 2017), in which healthy coping behaviors and cognitions allow individuals to overcome adverse experiences (Johnson et al., 2011; White et al., 2010). When surveyed about positive coping strategies, first responders most frequently reported resilience as important to their well-being (Crowe et al., 2017).

Researchers corroborated the potential impact of resilience for the population. For example, in samples of LEOs, researchers confirmed resilience served as a protective factor for PTSD (Klimley et al., 2018) and as a mediator between social support and PTSD symptoms (McCanlies et al., 2017). In a sample of firefighters, individual resilience mediated the indirect path between traumatic events and global perceived stress of PTSD, along with the direct path between traumatic events and PTSD symptoms (Lee et al., 2014). Their model demonstrated that those with higher levels of resilience were more protected from traumatic stress. Similarly, among emergency dispatchers, resilience was positively correlated with positive affect and post-traumatic growth, and negatively correlated with job stress (Steinkopf et al., 2018). The replete associations of resilience as a protective factor led researchers to develop resilience-based interventions. For example, researchers surmised promising results from mindfulness-based resilience interventions for firefighters (Joyce et al., 2019) and LEOs (Christopher et al., 2018). Moreover, Antony and colleagues (2020) concluded that resilience training programs demonstrated potential to reduce occupational stress among first responders.

Assessment of Resilience
     Recognizing the significance of resilience as a mediating factor in PTSD among first responders and as a promising basis for interventions when working with LEOs, a reliable means to measure it among first responder clients is warranted. In a methodological review of resilience assessments, Windle and colleagues (2011) identified 19 different measures of resilience. They found 15 assessments were from original development and validation studies with four subsequent validation manuscripts from their original assessment, of which none were developed with military or first responder samples.

Subsequently, Johnson et al. (2011) developed the Response to Stressful Experiences Scale (RSES-22) to assess resilience among military populations. Unlike deficit-based assessments of resilience, they proposed a multidimensional construct representing how individuals respond to stressful experiences in adaptive or healthy ways. Cognitive flexibility, meaning-making, and restoration were identified as key elements when assessing for individuals’ characteristics connected to resilience when overcoming hardships. Initially they validated a five-factor structure for the RSES-22 with military active-duty and reserve components. Later, De La Rosa et al. (2016) re-examined the RSES-22. De La Rosa and colleagues discovered a unidimensional factor structure of the RSES-22 and validated a shorter 4-item subset of the instrument, the RSES-4, again among military populations.

It is currently unknown if the performance of the RSES-4 can be generalized to first responder populations. While there are some overlapping experiences between military populations and first responders in terms of exposure to trauma and high-risk occupations, the Substance Abuse and Mental Health Services Administration (SAMHSA; 2018) suggested differences in training and types of risk. In the counseling profession, these populations are categorized together, as evidenced by the Military and Government Counseling Association ACA division. Additionally, there may also be dual identities within the populations. For example, Lewis and Pathak (2014) found that 22% of LEOs and 15% of firefighters identified as veterans. Although the similarities of the populations may be enough to theorize the use of the same resilience measure, validation of the RSES-22 and RSES-4 among first responders remains unexamined.

Purpose of the Study
     First responders are repeatedly exposed to traumatic and stressful events (Greinacher et al., 2019) and this exposure may impact their mental health, including symptoms of post-traumatic stress, anxiety, depression, and suicidality (Jetelina et al., 2020; Klimley et al., 2018). Though most measures of resilience are grounded in a deficit-based approach, researchers using a strength-based approach proposed resilience may be a protective factor for this population (Crowe et al., 2017; Wild et al., 2020). Consequently, counselors need a means to assess resilience in their clinical practice from a strength-based conceptualization of clients.

Johnson et al. (2011) offered a non-deficit approach to measuring resilience in response to stressful events associated with military service. Thus far, researchers have conducted analyses of the RSES-22 and RSES-4 with military populations (De La Rosa et al., 2016; Johnson et al., 2011; Prosek & Ponder, 2021), but not yet with first responders. While there are some overlapping characteristics between the populations, there are also unique differences that warrant research with discrete sampling (SAMHSA, 2018). In light of the importance of resilience as a protective factor for mental health among first responders, the purpose of the current study was to confirm the reliability and validity of the RSES-22 and RSES-4 when utilized with this population. In the current study, we hypothesized the measures would perform similarly among first responders and if so, the RSES-4 would offer counselors a brief assessment option in clinical practice that is both reliable and valid.

Method

Participants
     Participants in the current non-probability, purposive sample study were first responders (N = 238) seeking clinical treatment at an outpatient, mental health nonprofit organization in the Southwestern United States. Participants’ mean age was 37.53 years (SD = 10.66). The majority of participants identified as men (75.2%; n = 179), with women representing 24.8% (n = 59) of the sample. In terms of race and ethnicity, participants identified as White (78.6%; n = 187), Latino/a (11.8%; n = 28), African American or Black (5.5%; n = 13), Native American (1.7%; n = 4), Asian American (1.3%; n = 3), and multiple ethnicities (1.3%; n = 3). The participants identified as first responders in three main categories: LEO (34.9%; n = 83), EMT (28.2%; n = 67), and fire rescue (25.2%; n = 60). Among the first responders, 26.9% reported previous military affiliation. As part of the secondary analysis, we utilized a subsample (n = 190) that was reflective of the larger sample (see Table 1).

Procedure
     The data for this study were collected between 2015–2020 as part of the routine clinical assessment procedures at a nonprofit organization serving military service members, first responders, frontline health care workers, and their families. The agency representatives conduct clinical assessments with clients at intake, Session 6, Session 12, and Session 18 or when clinical services are concluded. We consulted with the second author’s Institutional Review Board, which determined the research as exempt, given the de-identified, archival nature of the data. For inclusion in this analysis, data needed to represent first responders, ages 18 or older, with a completed RSES-22 at intake. The RSES-4 are four questions within the RSES-22 measure; therefore, the participants did not have to complete an additional measure. For the secondary analysis, data from participants who also completed other mental health measures at intake were also included (see Measures).

 

Table 1

Demographics of Sample

Characteristic Sample 1

(N = 238)

Sample 2

(n = 190)

Age (Years)
    Mean 37.53 37.12
    Median 35.50 35.00
    SD 10.66 10.30
    Range 46 45
Time in Service (Years)
    Mean 11.62 11.65
    Median 10.00 10.00
    SD   9.33   9.37
    Range   41 39
n (%)
First Responder Type
    Emergency Medical
Technicians
67 (28.2%) 54 (28.4%)
    Fire Rescue 60 (25.2%) 45 (23.7%)
    Law Enforcement 83 (34.9%) 72 (37.9%)
    Other  9 (3.8%) 5 (2.6%)
    Two or more 10 (4.2%) 6 (3.2%)
    Not reported  9 (3.8%) 8 (4.2%)
Gender
    Women   59 (24.8%)   47 (24.7%)
    Men 179 (75.2%) 143 (75.3%)
Ethnicity
    African American/Black 13 (5.5%) 8 (4.2%)
    Asian American   3 (1.3%) 3 (1.6%)
    Latino(a)/Hispanic  28 (11.8%) 24 (12.6%)
    Multiple Ethnicities  3 (1.3%) 3 (1.6%)
    Native American  4 (1.7%) 3 (1.6%)
    White 187 (78.6%) 149 (78.4%)

Note. Sample 2 is a subset of Sample 1. Time in service for Sample 1, n = 225;
time in service for Sample 2, n = 190.

 

Measures
Response to Stressful Experiences Scale
     The Response to Stressful Experiences Scale (RSES-22) is a 22-item measure to assess dimensions of resilience, including meaning-making, active coping, cognitive flexibility, spirituality, and self-efficacy (Johnson et al., 2011). Participants respond to the prompt “During and after life’s most stressful events, I tend to” on a 5-point Likert scale from 0 (not at all like me) to 4 (exactly like me). Total scores range from 0 to 88 in which higher scores represent greater resilience. Example items include see it as a challenge that will make me better, pray or meditate, and find strength in the meaning, purpose, or mission of my life. Johnson et al. (2011) reported the RSES-22 demonstrates good internal consistency (α = .92) and test-retest reliability (α = .87) among samples from military populations. Further, the developers confirmed convergent, discriminant, concurrent, and incremental criterion validity (see Johnson et al., 2011). In the current study, Cronbach’s alpha of the total score was .93. 

Adapted Response to Stressful Experiences Scale
     The adapted Response to Stressful Experiences Scale (RSES-4) is a 4-item measure to assess resilience as a unidimensional construct (De La Rosa et al., 2016). The prompt and Likert scale are consistent with the original RSES-22; however, it only includes four items: find a way to do what’s necessary to carry on, know I will bounce back, learn important and useful life lessons, and practice ways to handle it better next time. Total scores range from 0 to 16, with higher scores indicating greater resilience. De La Rosa et al. (2016) reported acceptable internal consistency (α = .76–.78), test-retest reliability, and demonstrated criterion validity among multiple military samples. In the current study, the Cronbach’s alpha of the total score was .74.

Patient Health Questionnaire-9
     The Patient Health Questionnaire-9 (PHQ-9) is a 9-item measure to assess depressive symptoms in the past 2 weeks (Kroenke et al., 2001). Respondents rate the frequency of their symptoms on a 4-point Likert scale ranging from 0 (not at all) to 3 (nearly every day). Total scores range from 0 to 27, in which higher scores indicate increased severity of depressive symptoms. Example items include little interest or pleasure in doing things and feeling tired or having little energy. Kroenke et al. (2001) reported good internal consistency (α = .89) and established criterion and construct validity. In this sample, Cronbach’s alpha of the total score was .88.

PTSD Checklist-5
     The PTSD Checklist-5 (PCL-5) is a 20-item measure for the presence of PTSD symptoms in the past month (Blevins et al., 2015). Participants respond on a 5-point Likert scale indicating frequency of PTSD-related symptoms from 0 (not at all) to 4 (extremely). Total scores range from 0 to 80, in which higher scores indicate more severity of PTSD-related symptoms. Example items include repeated, disturbing dreams of the stressful experience and trouble remembering important parts of the stressful experience. Blevins et al. (2015) reported good internal consistency (α = .94) and determined convergent and discriminant validity. In this sample, Cronbach’s alpha of the total score was .93.

Generalized Anxiety Disorder-7
     The Generalized Anxiety Disorder-7 (GAD-7) is a 7-item measure to assess for anxiety symptoms over the past 2 weeks (Spitzer et al., 2006). Participants rate the frequency of the symptoms on a 4-point Likert scale ranging from 0 (not at all) to 3 (nearly every day). Total scores range from 0 to 21 with higher scores indicating greater severity of anxiety symptoms. Example items include not being able to stop or control worrying and becoming easily annoyed or irritable. Among patients from primary care settings, Spitzer et al. (2006) determined good internal consistency (α = .92) and established criterion, construct, and factorial validity. In this sample, Cronbach’s alpha of the total score was .91.

Suicidal Behaviors Questionnaire-Revised
     The Suicidal Behaviors Questionnaire-Revised (SBQ-R) is a 4-item measure to assess suicidality (Osman et al., 2001). Each item assesses a different dimension of suicidality: lifetime ideation and attempts, frequency of ideation in the past 12 months, threat of suicidal behaviors, and likelihood of suicidal behaviors (Gutierrez et al., 2001). Total scores range from 3 to 18, with higher scores indicating more risk of suicide. Example items include How often have you thought about killing yourself in the past year? and How likely is it that you will attempt suicide someday? In a clinical sample, Osman et al. (2001) reported good internal consistency (α = .87) and established criterion validity. In this sample, Cronbach’s alpha of the total score was .85.

Data Analysis
     Statistical analyses were conducted using SPSS version 26.0 and SPSS Analysis of Moment Structures (AMOS) version 26.0. We examined the dataset for missing values, replacing 0.25% (32 of 12,836 values) of data with series means. We reviewed descriptive statistics of the RSES-22 and RSES-4 scales. We determined multivariate normality as evidenced by skewness less than 2.0 and kurtosis less than 7.0 (Dimitrov, 2012). We assessed reliability for the scales by interpreting Cronbach’s alphas and inter-item correlations to confirm internal consistency.

We conducted two separate confirmatory factor analyses to determine the model fit and factorial validity of the 22-item measure and adapted 4-item measure. We used several indices to conclude model fit: minimum discrepancy per degree of freedom (CMIN/DF) and p-values, root mean residual (RMR), goodness-of-fit index (GFI), comparative fit index (CFI), Tucker-Lewis index (TLI), and the root mean square error of approximation (RMSEA). According to Dimitrov (2012), values for the CMIN/DF < 2.0,p > .05, RMR < .08, GFI > .90, CFI > .90, TLI > .90, and RMSEA < .10 provide evidence of a strong model fit. To determine criterion validity, we assessed a subsample of participants (n = 190) who had completed the RSES-22, RSES-4, and four other psychological measures (i.e., PHQ-9, PCL-5, GAD-7, and SBQ-R). We determined convergent validity by conducting bivariate correlations between the RSES-22 and RSES-4.

Results

Descriptive Analyses
     We computed means, standard deviations, 95% confidence interval (CI), and score ranges for the RSES-22 and RSES-4 (Table 2). Scores on the RSES-22 ranged from 19–88. Scores on the RSES-4 ranged from 3–16. Previous researchers using the RSES-22 on military samples reported mean scores of 57.64–70.74 with standard deviations between 8.15–15.42 (Johnson et al., 2011; Prosek & Ponder, 2021). In previous research of the RSES-4 with military samples, mean scores were 9.95–11.20 with standard deviations between 3.02–3.53(De La Rosa et al., 2016; Prosek & Ponder, 2021).

 

Table 2

Descriptive Statistics for RSES-22 and RSES-4

Variable M SD 95% CI Score Range
RSES-22 scores 60.12 13.76 58.52, 61.86 19–88
RSES-4 scores 11.66 2.62 11.33, 11.99 3–16

Note. N = 238. RSES-22 = Response to Stressful Experiences Scale 22-item; RSES-4 = Response
to Stressful Experiences Scale 4-item adaptation.


Reliability Analyses
     To determine the internal consistency of the resiliency measures, we computed Cronbach’s alphas. For the RSES-22, we found strong evidence of inter-item reliability (α = .93), which was consistent with the developers’ estimates (α = .93; Johnson et al., 2011). For the RSES-4, we assessed acceptable inter-item reliability (α = .74), which was slightly lower than previous estimates (α = .76–.78; De La Rosa et al., 2016). We calculated the correlation between items and computed the average of all the coefficients. The average inter-item correlation for the RSES-22 was .38, which falls within the acceptable range (.15–.50). The average inter-item correlation for the RSES-4 was .51, slightly above the acceptable range. Overall, evidence of internal consistency was confirmed for each scale. 

Factorial Validity Analyses
     We conducted two confirmatory factor analyses to assess the factor structure of the RSES-22 and RSES-4 for our sample of first responders receiving mental health services at a community clinic (Table 3). For the RSES-22, a proper solution converged in 10 iterations. Item loadings ranged between .31–.79, with 15 of 22 items loading significantly ( > .6) on the latent variable. It did not meet statistical criteria for good model fit: χ2 (209) = 825.17, p = .000, 90% CI [0.104, 0.120]. For the RSES-4, a proper solution converged in eight iterations. Item loadings ranged between .47–.80, with three of four items loading significantly ( > .6) on the latent variable. It met statistical criteria for good model fit: χ2 (2) = 5.89, p = .053, 90% CI [0.000, 0.179]. The CMIN/DF was above the suggested < 2.0 benchmark; however, the other fit indices indicated a model fit.

 

Table 3

Confirmatory Factor Analysis Fit Indices for RSES-22 and RSES-4

Variable df χ2 CMIN/DF RMR GFI CFI TLI RMSEA 90% CI
RSES-22 209 825.17/.000 3.95 .093 .749 .771 0.747 .112 0.104, 0.120
RSES-4    2    5.89/.053 2.94 .020 .988 .981 0.944 .091 0.000, 0.179

Note. N = 238. RSES-22 = Response to Stressful Experiences Scale 22-item; RSES-4 = Response to Stressful Experiences Scale 4-item adaptation; CMIN/DF = Minimum Discrepancy per Degree of Freedom; RMR = Root Mean Square Residual;
GFI = Goodness-of-Fit Index; CFI = Comparative Fit Index; TLI = Tucker-Lewis Index; RMSEA = Root Mean Squared Error of Approximation.

 

Criterion and Convergent Validity Analyses
     To assess for criterion validity of the RSES-22 and RSES-4, we conducted correlational analyses with four established psychological measures (Table 4). We utilized a subsample of participants (n = 190) who completed the PHQ-9, PCL-5, GAD-7, and SBQ-R at intake. Normality of the data was not a concern because analyses established appropriate ranges for skewness and kurtosis (± 1.0). The internal consistency of the RSES-22 (α = .93) and RSES-4 (α = .77) of the subsample was comparable to the larger sample and previous studies. The RSES-22 and RSES-4 related to the psychological measures of distress in the expected direction, meaning measures were significantly and negatively related, indicating that higher resiliency scores were associated with lower scores of symptoms associated with diagnosable mental health disorders (i.e., post-traumatic stress, anxiety, depression, and suicidal behavior). We verified convergent validity with a correlational analysis of the RSES-22 and RSES-4, which demonstrated a significant and positive relationship.

 

Table 4

Criterion and Convergent Validity of RSES-22 and RSES-4

M (SD) Cronbach’s α RSES-22 PHQ-9 PCL-5 GAD-7 SBQ-R
RSES-22 60.16 (14.17) .93 −.287* −.331* −.215* −.346*
RSES-4 11.65 (2.68) .77 .918 −.290* −.345* −.220* −.327*

Note. n = 190. RSES-22 = Response to Stressful Experiences Scale 22-item; RSES-4 = Response to Stressful Experiences Scale 4-item adaptation; PHQ-9 = Patient Health Questionnaire-9;
PCL-5 = PTSD Checklist-5; GAD-7 = Generalized Anxiety Disorder-7; SBQ-R = Suicidal Behaviors Questionnaire-Revised.
*p < .01.

 

Discussion

The purpose of this study was to validate the factor structure of the RSES-22 and the abbreviated RSES-4 with a first responder sample. Aggregated means were similar to those in the articles that validated and normed the measures in military samples (De La Rosa et al., 2016; Johnson et al., 2011; Prosek & Ponder, 2021). Additionally, the internal consistency was similar to previous studies. In the original article, Johnson et al. (2011) proposed a five-factor structure for the RSES-22, which was later established as a unidimensional assessment after further exploratory factor analysis (De La Rosa et al., 2016). Subsequently, confirmatory factor analyses with a treatment-seeking veteran population revealed that the RSES-22 demonstrated unacceptable model fit, whereas the RSES-4 demonstrated a good model fit (Prosek & Ponder, 2021). In both samples, the RSES-4 GFI, CFI, and TLI were all .944 or higher, whereas the RSES-22 GFI, CFI, and TLI were all .771 or lower. Additionally, criterion and convergent validity as measured by the PHQ-9, PCL-5, and GAD-7 in both samples were extremely close. Similarly, in this sample of treatment-seeking first responders, confirmatory factor analyses indicated an inadequate model fit for the RSES-22 and a good model fit for the RSES-4. Lastly, convergent and criterion validity were established with correlation analyses of the RSES-22 and RSES-4 with four other standardized assessment instruments (i.e., PHQ-9, PCL-5, GAD-7, SBQ-R). We concluded that among the first responder sample, the RSES-4 demonstrated acceptable psychometric properties, as well as criterion and convergent validity with other mental health variables (i.e., post-traumatic stress, anxiety, depression, and suicidal behavior).

Implications for Clinical Practice
     First responders are a unique population and are regularly exposed to trauma (Donnelly & Bennett, 2014; Jetelina et al., 2020; Klimley et al., 2018; Weiss et al., 2010). Although first responders could potentially benefit from espousing resilience, they are often hesitant to seek mental health services (Crowe et al., 2017; Jones, 2017). The RSES-22 and RSES-4 were originally normed with military populations. The results of the current study indicated initial validity and reliability among a first responder population, revealing that the RSES-4 could be useful for counselors in assessing resilience.

It is important to recognize that first responders have perceived coping with traumatic stress as an individual process (Crowe et al., 2017) and may believe that seeking mental health services is counter to the emotional and physical training expectations of the profession (Crowe et al., 2015). Therefore, when first responders seek mental health care, counselors need to be prepared to provide culturally responsive services, including population-specific assessment practices and resilience-oriented care.

Jones (2017) encouraged a comprehensive intake interview and battery of appropriate assessments be conducted with first responder clients. Counselors need to balance the number of intake questions while responsibly assessing for mental health comorbidities such as post-traumatic stress, anxiety, depression, and suicidality. The RSES-4 provides counselors a brief, yet targeted assessment of resilience.

Part of what cultural competency entails is assessing constructs (e.g., resilience) that have been shown to be a protective factor against PTSD among first responders (Klimley et al., 2018). Since the items forming the RSES-4 were developed to highlight the positive characteristics of coping (Johnson et al., 2011), rather than a deficit approach, this aligns with the grounding of the counseling profession. It is also congruent with first responders’ perceptions of resilience. Indeed, in a content analysis of focus group interviews with first responders, participants defined resilience as a positive coping strategy that involves emotional regulation, perseverance, personal competence, and physical fitness (Crowe et al., 2017).

The RSES-4 is a brief, reliable, and valid measure of resilience with initial empirical support among a treatment-seeking first responder sample. In accordance with the ACA (2014) Code of Ethics, counselors are to administer assessments normed with the client population (E.8.). Thus, the results of the current study support counselors’ use of the measure in practice. First responder communities are facing unprecedented work tasks in response to COVID-19. Subsequently, their mental health might suffer (Centers for Disease Control and Prevention, 2020) and experts have recommended promoting resilience as a protective factor for combating the negative mental health consequences of COVID-19 (Chen & Bonanno, 2020). Therefore, the relevance of assessing resilience among first responder clients in the current context is evident.

Limitations and Future Research
     This study is not without limitations. The sample of first responders was homogeneous in terms of race, ethnicity, and gender. Subsamples of first responders (i.e., LEO, EMT, fire rescue) were too small to conduct within-group analyses to determine if the factor structure of the RSES-22 and RSES-4 would perform similarly. Also, our sample of first responders included two emergency dispatchers. Researchers reported that emergency dispatchers should not be overlooked, given an estimated 13% to 15% of emergency dispatchers experience post-traumatic symptomatology (Steinkopf et al., 2018). Future researchers may develop studies that further explore how, if at all, emergency dispatchers are represented in first responder research.

Furthermore, future researchers could account for first responders who have prior military service. In a study of LEOs, Jetelina et al. (2020) found that participants with military experience were 3.76 times more likely to report mental health concerns compared to LEOs without prior military affiliation. Although we reported the prevalence rate of prior military experience in our sample, the within-group sample size was not sufficient for additional analyses. Finally, our sample represented treatment-seeking first responders. Future researchers may replicate this study with non–treatment-seeking first responder populations.

Conclusion
     First responders are at risk for sustaining injuries, experiencing life-threatening events, and witnessing harm to others (Lanza et al., 2018). The nature of their exposure can be repeated and cumulative over time (Donnelly & Bennett, 2014), indicating an increased risk for post-traumatic stress, anxiety, and depressive symptoms, as well as suicidal behavior (Jones et al., 2018). Resilience is a promising protective factor that promotes wellness and healthy coping among first responders (Wild et al., 2020), and counselors may choose to routinely measure for resilience among first responder clients. The current investigation concluded that among a sample of treatment-seeking first responders, the original factor structure of the RSES-22 was unstable, although it demonstrated good reliability and validity. The adapted version, RSES-4, demonstrated good factor structure while also maintaining acceptable reliability and validity, consistent with studies of military populations (De La Rosa et al., 2016; Johnson et al., 2011; Prosek & Ponder, 2021). The RSES-4 provides counselors with a brief and strength-oriented option for measuring resilience with first responder clients.

 

Conflict of Interest and Funding Disclosure
The authors reported no conflict of interest
or funding contributions for the development
of this manuscript.

 

References

American Counseling Association. (2014). ACA code of ethics.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).

Antony, J., Brar, R., Khan, P. A., Ghassemi, M., Nincic, V., Sharpe, J. P., Straus, S. E., & Tricco, A. C. (2020). Interventions for the prevention and management of occupational stress injury in first responders: A rapid overview of reviews. Systematic Reviews, 9(121), 1–20. https://doi.org/10.1186/s13643-020-01367-w

Blevins, C. A., Weathers, F. W., Davis, M. T., Witte, T. K., & Domino, J. L. (2015). The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation. Journal of Traumatic Stress, 28(6), 489–498. https://doi.org/10.1002/jts.22059

Burnett, H. J., Jr. (2017). Revisiting the compassion fatigue, burnout, compassion satisfaction, and resilience connection among CISM responders. Journal of Police Emergency Response, 7(3), 1–10. https://doi.org/10.1177/2158244017730857

Centers for Disease Control and Prevention. (2020, June 30). Coping with stress. https://www.cdc.gov/coronavirus/2019-ncov/daily-life-coping/managing-stress-anxiety.html

Chen, S., & Bonanno, G. A. (2020). Psychological adjustment during the global outbreak of COVID-19: A resilience perspective. Psychological Trauma: Theory, Research, Practice, and Policy, 12(S1), S51–S54. https://doi.org/10.1037/tra0000685

Christopher, M. S., Hunsinger, M., Goerling, R. J., Bowen, S., Rogers, B. S., Gross, C. R., Dapolonia, E., & Pruessner, J. C. (2018). Mindfulness-based resilience training to reduce health risk, stress reactivity, and aggression among law enforcement officers: A feasibility and preliminary efficacy trial. Psychiatry Research, 264, 104–115. https://doi.org/10.1016/j.psychres.2018.03.059

Crowe, A., Glass, J. S., Lancaster, M. F., Raines, J. M., & Waggy, M. R. (2015). Mental illness stigma among first responders and the general population. Journal of Military and Government Counseling, 3(3), 132–149. http://mgcaonline.org/wp-content/uploads/2013/02/JMGC-Vol-3-Is-3.pdf

Crowe, A., Glass, J. S., Lancaster, M. F., Raines, J. M., & Waggy, M. R. (2017). A content analysis of psychological resilience among first responders. SAGE Open, 7(1), 1–9. https://doi.org/10.1177/2158244017698530

De La Rosa, G. M., Webb-Murphy, J. A., & Johnston, S. L. (2016). Development and validation of a brief measure of psychological resilience: An adaptation of the Response to Stressful Experiences Scale. Military Medicine, 181(3), 202–208. https://doi.org/10.7205/MILMED-D-15-00037

Dimitrov, D. M. (2012). Statistical methods for validation of assessment scale data in counseling and related fields. American Counseling Association.

Donnelly, E. A., & Bennett, M. (2014). Development of a critical incident stress inventory for the emergency medical services. Traumatology, 20(1), 1–8. https://doi.org/10.1177/1534765613496646

Greinacher, A., Derezza-Greeven, C., Herzog, W., & Nikendei, C. (2019). Secondary traumatization in first responders: A systematic review. European Journal of Psychotraumatology, 10(1), 1562840. https://doi.org/10.1080/20008198.2018.1562840

Gutierrez, P. M., Osman, A., Barrios, F. X., & Kopper, B. A. (2001). Development and initial validation of the Self-Harm Behavior Questionnaire. Journal of Personality Assessment, 77(3), 475–490. https://doi.org/10.1207/S15327752JPA7703_08

Jetelina, K. K., Mosberry, R. J., Gonzalez, J. R., Beauchamp, A. M., & Hall, T. (2020). Prevalence of mental illnesses and mental health care use among  police officers. JAMA Network Open, 3(10), 1–12. https://doi.org/10.1001/jamanetworkopen.2020.19658

Johnson, D. C., Polusny, M. A., Erbes, C. R., King, D., King, L., Litz, B. T., Schnurr, P. P., Friedman, M., Pietrzak, R. H., & Southwick, S. M. (2011). Development and initial validation of the Response to Stressful Experiences Scale. Military Medicine, 176(2), 161–169. https://doi.org/10.7205/milmed-d-10-00258

Jones, S. (2017). Describing the mental health profile of first responders: A systematic review. Journal of the American Psychiatric Nurses Association, 23(3), 200–214. https://doi.org/10.1177/1078390317695266

Jones, S., Nagel, C., McSweeney, J., & Curran, G. (2018). Prevalence and correlates of psychiatric symptoms among first responders in a Southern state. Archives of Psychiatric Nursing, 32(6), 828–835. https://doi.org/10.1016/j.apnu.2018.06.007

Joyce, S., Tan, L., Shand, F., Bryant, R. A., & Harvey, S. B. (2019). Can resilience be measured and used to predict mental health symptomology among first responders exposed to repeated trauma? Journal of Occupational and Environmental Medicine, 61(4), 285–292. https://doi.org/10.1097/JOM.0000000000001526

Kleim, B., & Westphal, M. (2011). Mental health in first responders: A review and recommendation for prevention and intervention strategies. Traumatology, 17(4), 17–24. https://doi.org/10.1177/1534765611429079

Klimley, K. E., Van Hasselt, V. B., & Stripling, A. M. (2018). Posttraumatic stress disorder in police, firefighters, and emergency dispatchers. Aggression and Violent Behavior, 43, 33–44.
https://doi.org/10.1016/j.avb.2018.08.005

Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x

Lanza, A., Roysircar, G., & Rodgers, S. (2018). First responder mental healthcare: Evidence-based prevention, postvention, and treatment. Professional Psychology: Research and Practice, 49(3), 193–204. https://doi.org/10.1037/pro0000192

Lee, J.-S., Ahn, Y.-S., Jeong, K.-S. Chae, J.-H., & Choi, K.-S. (2014). Resilience buffers the impact of traumatic events on the development of PTSD symptoms in firefighters. Journal of Affective Disorders, 162, 128–133. https://doi.org/10.1016/j.jad.2014.02.031

Lewis, G. B., & Pathak, R. (2014). The employment of veterans in state and local government service. State and Local Government Review, 46(2), 91–105. https://doi.org/10.1177/0160323X14537835

McCanlies, E. C., Gu, J. K., Andrew, M. E., Burchfiel, C. M., & Violanti, J. M. (2017). Resilience mediates the relationship between social support and post-traumatic stress symptoms in police officers. Journal of Emergency Management, 15(2), 107–116. https://doi.org/10.5055/jem.2017.0319

National Institute of Mental Health. (2017). Post-traumatic stress disorder. https://www.nimh.nih.gov/health/statistics/post-traumatic-stress-disorder-ptsd.shtml

Osman, A., Bagge, C. L., Gutierrez, P. M., Konick, L. C., Kopper, B. A., & Barrios, F. X. (2001). The Suicidal Behaviors Questionnaire–revised (SBQ-R): Validation with clinical and nonclinical samples. Assessment, 8(4), 443–454. https://doi.org/10.1177/107319110100800409

Prosek, E. A., & Ponder, W. N. (2021). Validation of the Adapted Response to Stressful Experiences Scale (RSES-4) among veterans [Manuscript submitted for publication].

Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for assessing generalized anxiety disorder (The GAD-7). Archives of Internal Medicine, 166(10), 1092–1097.
https://doi.org/10.1001/archinte.166.10.1092

Steinkopf, B., Reddin, R. A., Black, R. A., Van Hasselt, V. B., & Couwels, J. (2018). Assessment of stress and resiliency in emergency dispatchers. Journal of Police and Criminal Psychology, 33(4), 398–411.
https://doi.org /10.1007/s11896-018-9255-3

Substance Abuse and Mental Health Services Administration. (2018, May). First responders: Behavioral health concerns, emergency response, and trauma. Disaster Technical Assistance Center Supplemental Research Bulletin. https://www.samhsa.gov/sites/default/files/dtac/supplementalresearchbulletin-firstresponders-may2018.pdf

Weiss, D. S., Brunet, A., Best, S. R., Metzler, T. J., Liberman, A., Pole, N., Fagan, J. A., & Marmar, C. R. (2010). Frequency and severity approaches to indexing exposure to trauma: The Critical Incident History Questionnaire for police officers. Journal of Traumatic Stress, 23(6), 734–743.
https://doi.org/10.1002/jts.20576

White, B., Driver, S., & Warren, A. M. (2010). Resilience and indicators of adjustment during rehabilitation from a spinal cord injury. Rehabilitation Psychology, 55(1), 23–32. https://doi.org/10.1037/a0018451

Wild, J., El-Salahi, S., Degli Esposti, M., & Thew, G. R. (2020). Evaluating the effectiveness of a group-based resilience intervention versus psychoeducation for emergency responders in England: A randomised controlled trial. PLoS ONE, 15(11), e0241704.  https://doi.org/10.1371/journal.pone.0241704

Windle, G., Bennett, K. M., & Noyes, J. (2011). A methodological review of resilience measurement scales. Health and Quality of Life Outcomes, 9, Article 8, 1–18. https://doi.org/10.1186/1477-7525-9-8

 

Warren N. Ponder, PhD, is Director of Outcomes and Evaluation at One Tribe Foundation. Elizabeth A. Prosek, PhD, NCC, LPC, is an associate professor at Penn State University. Tempa Sherrill, MS, LPC-S, is the founder of Stay the Course and a volunteer at One Tribe Foundation. Correspondence may be addressed to Warren N. Ponder, 855 Texas St., Suite 105, Fort Worth, TX 76102, warren@1tribefoundation.org.

Whiteness Scholarship in the Counseling Profession: A 35-Year Content Analysis

Hannah B. Bayne, Danica G. Hays, Luke Harness, Brianna Kane

 

We conducted a content analysis of counseling scholarship related to Whiteness for articles published in national peer-reviewed counseling journals within the 35-year time frame (1984–2019) following the publication of Janet Helms’s seminal work on White racial identity. We identified articles within eight counseling journals for a final sample of 63 articles—eight qualitative (12.7%), 38 quantitative (60.3%), and 17 theoretical (27.0%). Our findings outline publication characteristics and trends and present themes for key findings in this area of scholarship. They reveal patterns such as type of research methodology, sampling, correlations between White racial identity and other constructs, and limitations of White racial identity assessment. Based on this overview of extant research on Whiteness, our recommendations include future research that focuses on behavioral and clinical manifestations, anti-racism training within counselor education, and developing a better overall understanding of how White attitudes and behaviors function for self-protection.

Keywords: Whiteness, White racial identity, counseling scholarship, counseling journals, content analysis

 

Counselors are ethically guided to understand and address the roles that race, privilege, and oppression play in impacting both themselves and their clients (American Counseling Association [ACA], 2014). Most practitioners identify as White despite the population diversity in the United States (U.S. Census Bureau, 2020), which holds implications for understanding how Whiteness impacts culturally competent counselor training and practice (Helms, 1984, 1995, 2017). It is important, then, to understand the role of racial identity within counseling, particularly in terms of how Whiteness can be deconstructed and examined as a constant force impacting power dynamics and client progress (Helms, 1990, 2017; Malott et al., 2015). Whiteness models (i.e., Helms, 1984) describe how White people make meaning of their own and others’ racial identity as a result of personal and social experiences with race (Helms, 1984, 2017). The Helms model, along with other constructs, such as color-blindness (Frankenberg, 1993), White racial consciousness (Claney & Parker, 1989), and White fragility (DiAngelo, 2018), implicates the harmful impacts of Whiteness and invites critical reflection of how these constructs impact the counseling process.

Though much has been theorized regarding Whiteness and its impact within the helping professions, the contributions of Whiteness scholarship within professional counseling journals are unclear. An understanding of the specific professional applications and explorations of Whiteness within counseling can help identify best practices in counselor education, research, and practice to counter the harmful impacts of Whiteness and encourage growth toward anti-racist attitudes and behaviors.

White Racial Identity and Related Constructs
     The Helms (1984) model of White racial identity (WRI) presents Whiteness as a developmental process centering on racial consciousness (i.e., the awareness of one’s own race), as well as awareness of attitudes and behaviors toward other racial groups (Helms, 1984, 1990, 1995, 2017). According to Helms, White people have the privilege to restrict themselves to environments and relationships that are homogenous and White-normative, thus limiting their progression through the stages (DiAngelo, 2018; Helms, 1984). The initial model (Helms, 1984) contained five stages (i.e., Contact, Disintegration, Reintegration, Pseudo-Independence, and Autonomy), each with a positive or negative response that could facilitate progression toward a more advanced stage, regression to earlier stages of the model, or stagnation at the current stage of development. Helms (1990) later added a sixth status, Immersion/Emersion, to the model as an intermediary between Pseudo-Independence and Autonomy. These final three stages of the model (i.e., Pseudo-Independence, Immersion/Emersion, Autonomy) involve increasing levels of racial acceptance and intellectual and emotional comfort with racial issues, which in turn leads to the development of a positive and anti-racist WRI (Helms, 1990, 1995).WRI requires intentional and sustained attention toward how Whiteness impacts the self and others, with progression through the stages leading to beneficial intra and interpersonal outcomes (Helms, 1990, 1995, 2017).

Since Helms (1984), several additional components of Whiteness have been introduced, primarily within psychology, counseling psychology, and sociology scholarship. White racial consciousness is distinct from the WRI model in its focus on attitudes toward racial out-groups, rather than using the White in-group as a reference point (Choney & Behrens, 1996; Claney & Parker, 1989). Race essentialism refers to the degree to which a person believes that race reflects biological differences that influence personal characteristics (Tawa, 2017). Symbolic/modern racism refers to overt attitudes of White people related to their perceived superiority (Henry & Sears, 2002; McConahay, 1986). A fourth Whiteness component, color-blind racial ideology, enables color-evasion (i.e., “I don’t see color”) and power-evasion roles (i.e., “everyone has an equal chance to succeed”), which allow White people to deny the impact of race and therefore evade a sense of responsibility for oppression (Frankenberg, 1993; Neville et al., 2013). White privilege refers to the systemic and unearned advantages provided to White people over people of color (McIntosh, 1988). There are also psychosocial costs accrued to White people as a result of racism that include (a) affective (e.g., anxiety and fear, anger, sadness, guilt and shame); (b) cognitive (i.e., distorted views of self, others, and reality in general related to race); and (c) behavioral (i.e., avoidance of cross-racial situations or loss of relationships with White people) impacts (Spanierman & Heppner, 2004). White fragility (DiAngelo, 2018) reflects defensive strategies White people use to re-establish cognitive and affective equilibrium regarding their own Whiteness and impact on others.

Whiteness concepts are thus varied, with different vantage points of how White people might engage in the consideration of power, privilege, and racism, and what potential implications these constructs might have on their development. These constructs also seem largely rooted in psychology research, and it is therefore unclear the extent to which counselor educators and researchers have examined and applied these constructs to training and practice. Such an analysis can assist in situating Whiteness within the specific contexts and professional roles of counseling and can identify areas in need of further study.

The Present Study
     Because of the varied components of Whiteness, as well as its potential impact on counselor development and counseling process and outcome (Helms, 1995, 2017), there is a need to examine how these constructs have been examined and applied within counseling research. We sought to identify how and to what degree Whiteness constructs have been explored or developed within the counseling profession since the publication of the Helms (1984) model. We hope to summarize empirical and theoretical constructs related to Whiteness in national peer-reviewed counseling journals to more clearly consider implications for training and practice. Such analysis can highlight the saliency of WRI, demonstrating the need for continued focus on the influences and impacts of Whiteness within counseling. The following research questions were addressed: 1) What types of articles, topics, and major findings are published on Whiteness?; 2) What are the methodological features of articles published on Whiteness?; and 3) What are themes from key findings across these publications?

Method

We employed content analysis to identify publication patterns of national peer-reviewed counseling journals regarding counseling research on Whiteness in order to understand the scope and depth of this scholarship as it applies to fostering counselor training and practice. Content analysis is the systematic review of text in order to produce and summarize numerical data and identify patterns across data sources regarding phenomena (Neuendorf, 2017). In addition, content analysis has been used to summarize and identify patterns for specific topics, including multicultural counseling (e.g., Singh & Shelton, 2011).

Data Sources and Procedure
     The sampling units for this study were journal articles on Whiteness topics published in national peer-reviewed journals (N = 24) of the ACA and its divisions, the American School Counselor Association, the American Mental Health Counselors Association, the National Board for Certified Counselors, and Chi Sigma Iota International. We used the following search terms: White supremacy, White racial identity, White privilege, White fragility, White guilt, White shame, White savior, White victimhood, color-blindness, race essentialism, anti-racism, White racism, reverse racism, White resistance, and Whiteness. We selected a 35-year review period (i.e., 1984–2019) to correspond with Helms’s (1984) foundational work on WRI.

We reviewed article abstracts to identify an initial sampling unit pool (N = 185 articles; 29 qualitative [15.6%], 56 quantitative [30.3%], and 100 theoretical [54.1%]). In pairs, we reviewed the initial pool to more closely examine each sampling unit for inclusion in analysis. We excluded 122 articles upon closer inspection (e.g., special issue introductions, personal narratives or profiles, broader focus on social justice issues, ethnic identity, multiculturalism, or primary focus on another racial group). This resulted in a final sample of 63 articles—eight qualitative (12.7%), 38 quantitative (60.3%), and 17 theoretical (27.0%; see Table 1).

Research Team
     Our team consisted of four researchers: two counselor education faculty members and two counselor education doctoral students. We all identify as White. Hannah B. Bayne and Danica G. Hays hold doctorates in counselor education, and Luke Harness and Brianna Kane hold master’s degrees in school counseling and mental health counseling, respectively. We were all trained in qualitative research methods, and Bayne and Hays have conducted numerous qualitative research projects, including previous content analyses. Bayne and Hays trained Harness and Kane on content analysis through establishing coding protocols and coding together until an acceptable inter-rater threshold was met.

 

Table 1

Exclusion and Inclusion of Articles by Journal and Article Type

Journal Excludeda Included Total

Sample

% of

Final

Sample

Quant Qual Theory Quant Qual Theory
Journal of Counseling & Development 5 0 11 16 4 5 24 38.1%
Journal of Multicultural Counseling and
Development
3 3 14 14 3 8 24 38.1%
Counselor Education and Supervision 1 0   1 4 1 2   7 11.1%
The Journal of Humanistic Counseling 1 2 14 1 1 1   3 4.8%
Journal of Mental Health Counseling 0 0   2 1 0 3   2 3.2%
Counseling and Values 0 0   0 1 0 0   1 1.6%
The Family Journal 1 1   5 0 0 2   1 1.6%
Journal of Creativity in Mental Health 0 2   4 0 0 1   1 1.6%
Adultspan Journal 0 0   0 0 0 0   0 0%
The Career Development Quarterly 0 0   0 0 0 0   0 0%
Counseling Outcome Research
and Evaluation
0 2   0 0 0 0   0 0%
Journal for Social Action in Counseling
and Psychology
0 0   3 0 0 0   0 0%
The Journal for Specialists in Group Work 0 1   6 0 0 0   0 0%
Journal of Addictions & Offender
Counseling
0 0   0 0 0 0   0 0%
Journal of Child and Adolescent Counseling 0 0   0 0 0 0   0 0%
Journal of College Counseling 2 0   0 0 0 0   0 0%
Journal of Counselor Leadership
and Advocacy
1 5   6 0 0 0   0 0%
Journal of Employment Counseling 2 0   4 0 0 0   0 0%
Journal of LGBTQ Issues in Counseling 0 1   2 0 0 0   0 0%
Journal of Military and Government
Counseling
0 0   0 0 0 0   0 0%
Measurement and Evaluation in
Counseling and Development
1 0   2 0 0 0   0 0%
Professional School Counseling 0 0   2 0 0 0   0 0%
Rehabilitation Counseling Bulletin 3 1   2 0 0 0   0 0%
The Professional Counselor 0 1   0 0 0 0   0 0%
Professional School Counseling 0 0    2 0 0 0   0 0%

Note. Quant = quantitative research articles; Qual = qualitative research articles; Theory = theoretical articles.
aArticles were excluded from analysis if they did not directly address Whiteness or White racial identity (e.g., special issue introductions, personal narratives or profiles, broader focus on social justice issues, ethnic identity, multiculturalism, or primary focus on another racial group).

 

Coding Frame Development
Dimensions and categories for our coding frame included: journal outlet, publication year, author characteristics (i.e., name, institutional affiliation, ACES region), article type, sample characteristics (e.g., composition, size, gender, race/ethnicity), research components (e.g., research design, data sources or instrumentation, statistical methods, research traditions, trustworthiness strategies), topics discussed (e.g., WRI attitudes, counselor preparation models, intervention use, client outcomes, counseling process), article implications and limitations, and a brief statement of key findings. Over the course of research team meetings, we reviewed and operationalized the coding frame dimensions and categories. We then selected one empirical and one conceptual article to code together in order to refine the coding frame, which resulted in further clarification of some categories. 

Data Analysis
     To establish evidence of replicability (Neuendorf, 2017), we coded eight (12.7%) randomly selected cases proportionate to the sample composition (i.e., two conceptual, four quantitative, two qualitative). We analyzed the accuracy rate of coding using R data analysis software for statistical analysis (LoMartire, 2020). Across 376 possible observations for eight cases, there was an acceptable rate of coding accuracy (0.89). In addition, pairwise Pearson-product correlations among raters indicated that coding misses did not follow a systematic pattern for any variable (r = −.10 to .65), and thus there were no significant variations in coding among research team members. After pilot coding, we met to discuss areas of coding misses to ensure understanding of the final coding frame.

For the main coding phase, we worked in pairs and divided the sample equally for independent and consensus coding. Upon completion of consensus coding of the entire sample, we extracted 29 keywords describing the Whiteness topics discussed in the articles. Bayne and Hays reviewed the 29 independent topics and collapsed the topics into eight larger themes. To identify themes across the key findings, Bayne and Harness reviewed 125 independent statements based on coder summaries of article findings, and through independent and consensus coding collapsed statements to yield three main themes.

Results

Article Characteristics
     We focused on several article characteristics (Research Question 1): article type (conceptual, quantitative, qualitative); number of relevant articles per journal outlet; the relationship between journal outlet and article type; and frequency of Whiteness topics within and across journal outlets. Of the 24 national peer-reviewed counseling journals, eight journals (33.3%) contained publications that met inclusion criteria (i.e., contained keywords for Whiteness from our search criteria and focused specifically on WRI). The number of publications in those journals ranged from 1 to 24 (M = 2.5; Mdn = 7.88; SD = 10.15) and are listed in order of frequency in Table 2). There was not a significant relationship between the journal outlet and article type (i.e., quantitative, qualitative, conceptual) for this topic (r = 0.04, p = .39).

 

Table 2

Articles Addressing Whiteness and Associated Keywords in National Peer-Reviewed Counseling Journals

Journal Articles Addressing Whiteness Percent of Total Sample
Journal of Counseling & Development 24 38.1%
Journal of Multicultural Counseling and Development 24 38.1%
Counselor Education and Supervision  7 11.1%
The Journal of Humanistic Counseling  3 4.8%
Journal of Mental Health Counseling  2 3.2%
Counseling and Values  1 1.6%
The Family Journal  1 1.6%
Journal of Creativity in Mental Health  1 1.6%
Adultspan Journal  0   0%
The Career Development Quarterly  0   0%
Counseling Outcome Research and Evaluation  0   0%
Journal for Social Action in Counseling and Psychology  0   0%
The Journal for Specialists in Group Work  0   0%
Journal of Addictions & Offender Counseling  0   0%
Journal of Child and Adolescent Counseling  0   0%
Journal of College Counseling  0   0%
Journal of Counselor Leadership and Advocacy  0   0%
Journal of Employment Counseling  0   0%
Journal of LGBTQ Issues in Counseling  0   0%
Journal of Military and Government Counseling  0   0%
Measurement and Evaluation in Counseling and
Development
 0   0%
Professional School Counseling  0   0%
Rehabilitation Counseling Bulletin  0   0%
The Professional Counselor  0   0%
Professional School Counseling  0   0%

 

    Additionally, we identified eight themes of topics discussed within counseling research on Whiteness (see Table 3). For qualitative research, the three most frequently addressed topics were theory development, intrapsychic variables, and multicultural counseling competency (MCC). The most frequent topics discussed in theoretical articles were theory development, counselor preparation, Whiteness and WRI expression, cultural identity development, and counseling process.

 

Table 3 

Themes in Topics Discussed Within Whiteness and WRI Articles

Theme Description N

%

Quant

n / %

Qual

n / %

Theory

n / %

Examples
Whiteness and WRI Expression Attitudes and knowledge related to WRI and Whiteness constructs, with some (n = 5) examining pre–posttest changes

 

43

68.3%

32 74.4% 3

7.0%

8

18.6%

WRI attitudes, color-blind racial attitudes, racism and responses, White privilege and responses, and developmental considerations

 

Cultural Identity Development Cultural identities and developmental processes outside of race

 

27

42.9%

21

77.8%

1

3.7%

5

18.5%

Ethnic identity, womanist identity, cultural demographics such as gender and age

 

Counselor Preparation Training implications, with some presenting training intervention findings (n = 6)

 

23

36.5%

17

73.9%

1

4.3%

5

21.8%

Pedagogy, training interventions, and supervision process and outcome

 

Theory Development Development or expansion of theoretical concepts 18

28.6%

5

27.8%

5

27.8%

8

44.4%

White racial consciousness versus WRI, prominent responses to White privilege, psychological dispositions of White racism

 

Multicultural Counseling Competency Measurements of perceived multicultural counseling competency

 

12

19.0%

10

83.3%

2

16.7%

0

0.0%

Perceived competency,
link with WRI
Counseling Process Counseling process and outcome variables

 

11

17.5%

8

72.7%

1

9.1%

2

18.2%

Client perceptions, working alliance, and clinical applications

 

Intrapsychic Variables Affective and cognitive components that influence Whiteness and WRI

 

11

17.5%

8

72.7%

2

18.2%

1

9.1%

Personality variables, cognitive development, ego development

 

Assessment Characteristics Development and/or critique of Whiteness and WRI measurements

 

9

14.3%

8

88.9%

0

0.0%

1

11.1%

Limitations of WRI scales, development of White privilege awareness scales
Totala 154

 

111

72.1%

15

9.7%

30

19.5%

Note. Quant = quantitative research articles; Qual = qualitative research articles; Theory = theoretical articles.
aPercentage total exceeds 100% because of rounding and/or topic overlap between articles.

 

Methodological Features
     To address Research Question 2, we explored the methodological features of articles. These features included sample composition, research design, data sources, and limitations as reported within each empirical article (n = 46).

Sample Composition
     For the 45 studies providing information about the racial/ethnic composition of their samples, White individuals accounted for a mean of 91% of total participants (range = 55%–100%; SD = 14). An average of 14% Black (SD = 6.7), 7.1% Latinx (SD = 4.7), 5.4% Asian (SD = 2.3), and less than 5% each of multiracial, Arab, and Native American respondents were included across the samples. Of studies reporting gender (n = 44), women accounted for an average of 68% of total participants (range = 33–100; SD = 14.7), and men accounted for 31% of total samples (range = 12–67; SD = 14). The age of participants, reported in 71.7% of the empirical studies, ranged from 16 to 81 (M = 29, SD = 8.2).

Of the 61 independent samples across the articles, a majority focused on student populations, with master’s trainees (n = 20, 32.8%), undergraduate students (n = 14, 21.9%), and doctoral trainees (n = 10, 16.4%) representing over 70% of the sample. The remainder of the samples included practitioners (n = 8, 13.1%), unspecified samples (n = 3, 4.9%), university educators (n = 2, 3.3%), educational specialist trainees (n = 2, 3.3%), site supervisors (n = 1, 1.6%), and general population adult samples (n = 1, 1.6%). The target audience of the articles (N = 63) focused primarily on counselor trainees (n = 34, 49.3%) or clients in agency/practice settings (n = 12, 17.4%). Other audiences included practitioners (n = 9, 13%), researchers (n = 3, 4.3%), general population (n = 6, 8.7%), counselor educators (n = 1, 1.4%), and general university personnel (n = 1, 1.4%).

Research Design and Data Sources
     Of the 38 quantitative articles, 10 (26.3%) included an intervention as part of the research design. The majority employed a correlational design (n = 27, 71.1%), with the remainder consisting of four (10.5%) descriptive, four (10.5%) quasi-experimental, one (2.6%) ex post facto/causal comparative, one (2.6%) pre-experimental, and one (2.6%) true experimental design. In recruiting and selecting samples, most researchers used convenience sampling (n = 27, 57.4%), while the rest used purposive (n = 12, 31.6%), simple random (n = 5, 10.6%), stratified (n = 2, 4.3%), and homogenous (n = 1, 2.1%) sampling methods.

Regarding study instrumentation, 37 quantitative studies utilized self-report forced-choice surveys, with one study employing a combination of forced-choice and open-ended question surveys. Across the 38 quantitative studies, 13 of 50 (26%) assessments were used more than once. The most frequently used assessment was the White Racial Identity Attitudes Scale (n = 24; Helms & Carter, 1990). The 50 assessments purported to measure the following targeted variables: race/racial identity/racism (n = 17, 34%); MCC (n = 9, 18%); cultural identity (n = 6, 12%); counseling process and outcome (n = 5, 10%); social desirability (n = 2, 4%); and other variables such as personality, anxiety, and ego development (n = 11, 22%). Finally, data analysis procedures included ANOVA/MANOVA (n = 25, 30.9%), correlation (n = 23, 28.4%), regression (n = 17, 21%), t-tests (n = 7, 8.6%), descriptive (n = 5, 6.2%), exploratory factor analysis (n = 1, 1.2%), confirmatory factor analysis (n = 1, 1.2%), SEM/path analysis (n = 1, 1.2%), and cluster analysis (n = 1, 1.2%).

We identified the research traditions of the eight qualitative studies as follows: phenomenology (n = 3, 37.5%), grounded theory (n = 2, 25%), and naturalistic inquiry (n = 1, 12.5%); two were unspecified (25%). The most common qualitative recruitment method was criterion sampling (n = 5, 62.5%), followed by convenience (n = 3, 37.5%), homogenous (n = 2, 25%), snowball/chain (n = 2, 25%), intensity (n = 2, 25%), and stratified purposeful (n = 1, 12.5%) sampling procedures. (Several studies used multiple recruitment methods, resulting in totals greater than 100%.) There were 12 data sources reported across the eight qualitative studies, falling into the following categories: individual interviews (n = 7, 58.3%), focus group interviews (n = 2, 16.7%), artifacts/documents (n = 2, 16.7%), and observations (n = 1, 8.3%). Trustworthiness strategies included prolonged engagement (n = 7, 13.7%); use of a research team (n = 6, 11.8%); researcher reflexivity, triangulation of data sources, thick description, and simultaneous data collection and analysis (n = 5 each, 9.8%); peer debriefing, audit trail, and member checking (n = 4 each, 7.8%); theory development (n = 3, 5.9%); and one each (2%) of external auditor, memos and/or field notes, and persistent observation.

Limitations Within Sampled Studies
     Of the 46 empirical studies, 44 (95.7%) reported limitations. Limitations included design issues related to sampling/generalizability (n = 38, 82.6%); self-report/social desirability (n = 23, 50.0%); instrumentation (n = 20, 43.5%); research design concerns related to the ability to directly measure a variable of interest (e.g., clinical work, training activities; n = 7, 15.2%); experimenter/researcher effects (n = 3, 6.5%); use of less sophisticated statistical methods (n = 3, 6.5%); and use of an analogue design (n = 2, 4.3%). Within identified limitations, researchers most often cited limited generalizability with regard to sample composition (i.e., lack of diversity, small sample sizes, homogenous samples). Social desirability was noted as a potential limitation given the nature of the topics (i.e., racism, prejudice, privilege). Instrumentation issues pertained to weak reliability for samples, limited validity evidence, and disadvantages of self-administration. Researchers also acknowledged the difficulty of conceptualizing WRI constructs as distinct, noting the multidimensional nature of WRI and the challenge in discriminating between complex constructs.

Key Findings
     There were three main categories of key findings. The largest category (i.e., 51 codes) consisted of identification of correlates and predictors of Whiteness/White racial identity. Findings related to gender and WRI were mixed, with several articles (n = 7) noting differences in WRI stages among men and women (i.e., women more frequently endorsing Contact and Pseudo-Independent stages, men more frequently endorsing Disintegration and Reintegration), and others determining gender differences were not significant in predicting WRI (n = 2). Additional findings included significant positive correlations and predictive effects between WRI, racism, MCC, personality variables (i.e., Openness linked with higher WRI and Neuroticism linked with lower WRI), and working alliance. Other constructs, such as ego defenses, emotional states, social–cognitive maturity, fear, and religious orientation, also demonstrated significant alignment with WRI stages. White guilt, the impact of personal relationships with communities of color, and lower levels of race salience (i.e., race essentialism) were also linked to Whiteness.

The next largest category (i.e., 32 codes) related to critiques of White racial identity models and measures. Most of the conceptual articles focused in some way on this category, often criticizing WRI models as subjective and lacking in complexity, or critiquing WRI measurement and previous research because of issues of reliability and validity. Several stressed caution for interpreting WRI according to existing models, suggesting a more nuanced approach of contextualizing individuals and accounting for within-group variation. Empirical articles also suggested that achieving and maintaining higher levels of WRI, particularly anti-racist identities and attitudes, may be more difficult than originally conceptualized and may require levels of engagement that are difficult to maintain in a racist society.

     Training implications and impact (i.e., 24 codes), noted within empirical and conceptual studies, included tips for addressing Whiteness in counselor education (e.g., offering courses focused on Whiteness and anti-racism) and in supervision (e.g., openly discussing race, privilege, and oppression; matching supervisors and supervisees by racial identity when possible). Empirical studies noted mixed improvement in WRI stages and MCC as a result of both general progression through a counselor training program as well as specific multicultural training: Training was linked to increased White guilt and privilege awareness (n = 15), though others did not find significant effects of training (n = 2). Conceptual articles emphasized focusing training on anti-racist development. Collectively, these findings and subsequent implications encourage further research and reflection on the correlates of WRI and MCC, factors facilitating growth, and ways to improve research and measurement to enhance critical engagement with these topics.

Discussion and Implications

In this content analysis of 63 articles covering a 35-year period across eight national counseling journals, we found that a third of counseling journals featured scholarship specifically related to Whiteness, with the Journal of Counseling & Development and the Journal of Multicultural Counseling and Development accounting for more than 76% of the total sampling units. The majority of the articles were quantitative, followed by theoretical and qualitative articles. Topical focus was centered on correlates of Whiteness with variables such as racism and color-blindness, other non-racial components of cultural identity, training implications, and theory development (see Table 3). Interestingly, many Whiteness constructs discussed in the general literature (e.g., White fragility, modern racism, psychosocial costs) were not addressed in counseling scholarship; the primary constructs discussed were WRI and White privilege.

The sample composition across empirical studies was primarily White and female with a mean age in the late 20s and with undergraduate students comprising on average 22% of the article samples. In addition, practitioners, site supervisors, the general population, and EdS trainees only comprised between 1.6% and 13.1% of the samples. Schooley et al. (2019) cautioned against the overuse of undergraduate students when measuring Whiteness constructs because of the complexities and situational influences of WRI development, and this warning seems to hold relevance for counseling scholarship. Methodological selection mirrored previously found patterns in counseling research (Wester et al., 2013), with most quantitative studies relying upon convenience sampling and correlational design with ANOVA/MANOVA as the selected statistical analyses. In addition, 26.3% of the articles included an intervention. For the qualitative studies, the most frequently used tradition and method was phenomenology and individual interviews.

Overall, findings from the sample support theoretically consistent relationships with Whiteness and/or WRI, including their predictive nature of MCC, social desirability, working alliance, and lower race salience. However, findings were mixed on the role of gender and MCC in connection to a training intervention. Additionally, some studies in our sample critiqued WRI models, cautioning against oversimplification of a complex model and highlighting issues in measurement due to subjectivity and social desirability. This critique aligns with previous researchers who have suggested that WRI is more complex than previously indicated (see Helms, 1984, 1990, 2017). WRI may be highly situational and affected by within-group differences and internal and external factors that complicate accuracy in assessment and clinical application. Of particular concern in previous research is the ability to properly conceptualize and measure the Contact and Autonomy stages (Carter et al., 2004). Both stages have demonstrated difficulty in assessment due to an individual’s lack of awareness of personal racism at each stage (Carter et al., 2004; Rowe, 2006). The Autonomy status, in particular, could be impacted by what DiAngelo (2018) referred to as “progressive” or “liberal” Whiteness, in which efforts are more focused on maintaining a positive self-image than engaging with people of color in meaningful ways (Helms, 2017). Therefore, although there are some consistencies and corroborations within counseling literature and other scholarship on Whiteness, the critiques and complexities of the topic suggest further inquiry is needed.

Implications for Counseling Research
     Based on our findings, we note several directions for future research. First, future studies could include greater demographic diversity as well as more participation from counselor educators, site supervisors, practitioners, and clients across the ACES regions. Including counselor educators in empirical studies can highlight aspects of Whiteness that influence their approach to training and scholarship. With regard to increasing scholarship involving site supervisors, practitioners, and clients, Hays et al. (2019) highlighted several strategies for recruiting sites to participate as co-researchers as well as obtaining clinical samples through strengthening research–practice partnerships. Additionally, recruiting more heterogenous samples—in terms of sample composition and demographics—could provide much-needed psychometrics for available measures as well as refined operationalization of Whiteness. Additional research can further explore individual correlates and predictors to enhance counselor training, supervision, and practice by identifying opportunities for assessment and development at each level of WRI.

Second, most reports of empirical studies in our sample noted concerns with sampling and generalizability, social desirability, and instrumentation. Given these concerns, researchers are to be cautious about the interpretation and application of previous study findings using the White Racial Identity Attitudes Scale (WRIAS). In particular, scholarship within counseling and related disciplines reveals substantial psychometric concerns with the WRIAS’s Contact and Autonomy stages (Behrens, 1997; Carter et al., 2004; Hays et al., 2008; Malott et al., 2015). The complex nature of assessing WRI-related behaviors that may run counter to a person’s intentions (Carter et al., 2004; DiAngelo, 2018) needs further study. Additionally, given the concerns with self-report measures due to socially desirable responses, it seems problematic that none of the current quantitative articles used performance measures, which could help to compare self-report with behaviors and client outcomes. Future research can therefore emphasize behavioral assessments and clinical outcomes to correlate findings with WRI models.

Third, the use of intervention-based research could explore core components of instruction, awareness, and experience to identify facilitative strategies for enhancing WRI in both counselor trainees and within client populations. Because White people are negatively impacted by racism and restricted racial identity, encouraging growth in WRI in both clinical and educational settings can be a means of promoting wellness for counselors and clients. Thus, research is needed that can carefully examine the complexities of WRI development and address difficulties in assessment due to defensive strategies such as White fragility and lack of insight into the various intra- and interpersonal manifestations of racism.

Finally, though the research examined within this analysis advances the application of WRI theory and practices within the counseling profession, opportunities exist for further exploration of WRI development and the intersection with multiple constructs of Whiteness discussed across the helping professions (e.g., White fragility, color-blindness, race essentialism). The articles analyzed for the present study reflect an assumption that more advanced WRI attitudes, lower color-blind attitudes, greater anti-racism attitudes, and greater awareness of White privilege can yield more positive clinical outcomes. However, given some of the aforementioned limitations, this assumption has not been empirically tested in counseling. Because clients’ and counselors’ affective, cognitive, and behavioral responses to Whiteness can affect the counseling relationship, process, and treatment selection and outcomes (Helms, 1984, 2017), it is imperative that this assumption is properly tested. Empirical and conceptual work should therefore further explore Whiteness constructs to elucidate how White attitudes and behaviors at each stage function for self-protection and move toward aspirational goals of anti-racism and ethical and competent clinical application.

Implications for Counseling Practice, Training, and Supervision
     In addition to future research directions related to Whiteness and WRI, findings allow for recommendations for counseling practice, training, and supervision. For example, extant literature emphasizes the importance of racial self-awareness, including an understanding of White privilege and racism. The practice of centering discussions on the harmful impacts of Whiteness, as well as the various ways Whiteness can manifest in therapeutic spaces, allows counselors to examine racial development within and around themselves. White counselors who are able to reflect on their own racial privileges and begin the conversation (i.e., broaching) about racial differences can increase the working alliance quality with clients of color (Burkard et al., 1999; Day-Vines et al., 2007; Helms, 1990).

Furthermore, counselors should heed the themes within the key findings of our sample, following recommendations for taking a broad, contextual, and critical view when understanding and applying WRI models. Counselors can be encouraged to view WRI as Helms (2019) intended—as a broad and complex interplay of relational dynamics, connected with other Whiteness constructs, and following an intentional progression toward anti-racism and social justice. Counselors should take particular caution with viewing the Autonomy stage as a point of arrival, given conflicting findings and the possibility that White people in higher stages may engage in behaviors to assuage guilt rather than to be true allies for people of color. The Helms model associates such attitudes and actions with the Pseudo-Independence stage (Helms, 2019), yet findings cast some doubt as to whether White people who score within the Autonomy stage have actually reached that level of WRI development. Counselors should thus interpret assessment scores with caution and ensure they are also assessing their own level of development and subsequent impact on others through continued and honest reflection and positive engagement in cross-racial relationships.

Regarding training, course content focusing on exploring Whiteness, WRI, and other racial identities through use of an anti-racism training model integrated throughout the curriculum can help students become comfortable with potential cross-racial conflicts and broaching Whiteness (Malott et al., 2015). The Council for Accreditation of Counseling and Related Educational Programs (CACREP) can similarly stress these desired student outcomes when updating standards for counselor training, specifically mentioning the importance of WRI as part of multicultural preparation. It is imperative to begin conversations about race and identity development to create opportunities for growth for any student who may be challenged with their racial identity and how it might impact their clients. Furthermore, counselor educators and supervisors can ask counselors in training to brainstorm how counseling and other services might be developed or adapted in order to contribute toward anti-racist goals and outcomes.

Limitations

The current findings are to be interpreted with caution, as the scope of our study presents some limitations. First, we chose to limit inclusion criteria to national peer-reviewed counseling journals in order to focus on scholarship within professional counseling journals, and therefore our results cannot be generalized to similar disciplines, dissertation research, book chapters, or more localized outlets such as state journals. Our coding sheet was also limited in the information it collected, including sample demographics. Though not all studies included the same demographic variables, we did not capture specifics related to a sample’s political affiliation, religious orientation, ability status, socioeconomic status, diversity exposure, or other details that could have better conceptualized the samples and findings. Additionally, we limited our search to the keywords related to Whiteness that we had identified in related literature but may have missed studies employing constructs outside of our search criteria. Our own identities as White academics may also have influenced the coding process as well as the subsequent interpretation of findings.

Conclusion

This content analysis provides a snapshot of Whiteness scholarship conducted in the counseling profession during a 35-year period. Patterns of study design and analysis were noted, and key findings were summarized to provide context and comparison within the broader literature. Identified themes and relationships highlight theoretically consistent findings for some Whiteness constructs, as well as showcase research gaps that need to be addressed before counselors can apply findings to practice and training. Finally, this content analysis demonstrates the need for a greater understanding of Whiteness and related constructs in counselor education, training, and practice.

 

Conflict of Interest and Funding Disclosure
The authors reported no conflict of interest
or funding contributions for the development
of this manuscript.

 

References

American Counseling Association. (2014). ACA code of ethics.

Behrens, J. T. (1997). Does the White Racial Identity Attitude Scale measure racial identity? Journal of Counseling Psychology, 44(1), 3–12. https://doi.org/10.1037/0022-0167.44.1.3

Burkard, A. W., Ponterotto, J. G., Reynolds, A. L., & Alfonso, V. C. (1999). White counselor trainees’ racial identity and working alliance perceptions. Journal of Counseling & Development, 77(3), 324–329. https://doi.org/10.1002/j.1556-6676.1999.tb02455.x

Carter, R. T., Helms, J. E., & Juby, H. L. (2004). The relationship between racism and racial identity for White Americans: A profile analysis. Journal of Multicultural Counseling and Development, 32(1), 2–17. https://doi.org/10.1002/j.2161-1912.2004.tb00357.x

Choney, S. K., & Behrens, J. T. (1996). Development of the Oklahoma Racial Attitudes Scale Preliminary Form (ORAS-P). Multicultural Assessment in Counseling and Clinical Psychology. https://digitalcommons.unl.edu/burosbookmulticultural/10

Claney, D., & Parker, W. M. (1989). Assessing White racial consciousness and perceived comfort with Black individuals: A preliminary study. Journal of Counseling & Development, 67(8), 449–451. https://doi.org/10.1002/j.1556-6676.1989.tb02114.x

Day-Vines, N. L., Wood, S. M., Grothaus, T., Craigen, L., Holman, A., Dotson-Blake, K., & Douglass, M. J. (2007). Broaching the subjects of race, ethnicity, and culture during the counseling process. Journal of Counseling & Development, 85(4), 401–409. https://doi.org/10.1002/j.1556-6678.2007.tb00608.x

DiAngelo, R. (2018). White fragility: Why it’s so hard for White people to talk about racism. Beacon Press.

Frankenberg, R. (1993). White women, race matters: The social construction of Whiteness. University of Minnesota Press.

Hays, D. G., Bolin, T., & Chen, C.-C. (2019). Closing the gap: Fostering successful research-practice partnerships in counselor education. Counselor Education and Supervision, 58(4), 278–292. https://doi.org/10.1002/ceas.12157

Hays, D. G., Chang, C. Y., & Havice, P. (2008). White racial identity statuses as predictors of White privilege awareness. The Journal of Humanistic Counseling, Education and Development, 47(2), 234–246. https://doi.org/10.1002/j.2161-1939.2008.tb00060.x

Helms, J. E. (1984). Toward a theoretical explanation of the effects of race on counseling: A Black and White model. The Counseling Psychologist, 12(4), 153–165. https://doi.org/10.1177/0011000084124013

Helms, J. E. (Ed.) (1990). Black and White racial identity: Theory, research, and practice. Praeger.

Helms, J. E. (1995). An update of Helms’ White and people of color racial identity models. In J. G. Ponterotto, J. M. Casas, L. A. Suzuki, & C. M. Alexander (Eds.), Handbook of multicultural counseling (1st ed.; pp. 181–196). SAGE.

Helms, J. E. (2017). The challenge of making Whiteness visible: Reactions to four Whiteness articles. The Counseling Psychologist, 45(5), 717–726. https://doi.org/10.1177/0011000017718943

Helms, J. E. (2019). A race is a nice thing to have: A guide to being a White person or understanding the White persons in your life (3rd ed.). Cognella.

Helms, J. E., & Carter, R. T. (1990). Development of the White Racial Identity Inventory. In J. E. Helms (Ed.), Black and White racial identity: Theory, research, and practice (pp. 67–80). Greenwood Press.

Henry, P. J., & Sears, D. O. (2002). The Symbolic Racism 2000 Scale. Political Psychology, 23(2), 253–283. https://doi.org/10.1111/0162-895X.00281

LoMartire, R. (2020). Rel: Reliability coefficients. R package version. 1.4.1. https://cran.r-project.org

Malott, K. M., Paone, T. R., Schaefle, S., Cates, J., & Haizlip, B. (2015). Expanding White racial identity theory: A qualitative investigation of Whites engaged in antiracist action. Journal of Counseling & Development, 93(3), 333–343. https://doi.org/10.1002/jcad.12031

McConahay, J. B. (1986). Modern racism, ambivalence, and the Modern Racism Scale. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 91–125). Academic Press.

McIntosh, P. (1988). White privilege and male privilege: A personal account of coming to see correspondences through work in women’s studies (Wellesley College, Center for Research on Women Working Paper, No. 189). Wellesley College. https://www.wcwonline.org/images/pdf/White_Privilege_and_Male_Privilege_Personal_Account-Peggy_McIntosh.pdf

Neuendorf, K. A. (2017). The content analysis guidebook (2nd ed.). SAGE.

Neville, H. A., Awad, G. H., Brooks, J. E., Flores, M. P., & Bleumel, J. (2013). Color-blind racial ideology: Theory, training, and measurement implications in psychology. American Psychologist, 68(6), 455–466. https://doi.org/10.1037/a0033282

Rowe, W. (2006). White racial identity: Science, faith, and pseudoscience. Journal of Multicultural Counseling & Development, 34(4), 235–243. https://doi.org/10.1002/j.2161-1912.2006.tb00042.x

Schooley, R. C., Debbiesiu, L. L., & Spanierman, L. B. (2019). Measuring Whiteness: A systematic review of instruments and call to action. The Counseling Psychologist, 47(4), 530–565. https://doi.org/10.1177/0011000019883261

Singh, A. A., & Shelton, K. (2011). A content analysis of LGBTQ qualitative research in counseling: A ten-year review. Journal of Counseling & Development, 89(2), 217–226. https://doi.org/10.1002/j.1556-6678.2011.tb00080.x

Spanierman, L. B., & Heppner, M. J. (2004). Psychosocial Costs of Racism to Whites Scale (PCRW): Construction and initial validation. Journal of Counseling Psychology, 51(2), 249–262. https://doi.org/10.1037/0022-0167.51.2.249

Tawa, J. (2017). The Beliefs About Race Scale (BARS): Dimensions of racial essentialism and their psychometric properties. Cultural Diversity and Ethnic Minority Psychology, 23(4), 516–526. https://doi.org/10.1037/cdp0000151

U.S. Census Bureau. (2020). Public Use Microdata Sample data. https://www.census.gov/programs-surveys/acs/data/pums.html

Wester, K. L., Borders, L. D., Boul, S., & Horton, E. (2013). Research quality: Critique of quantitative articles in the Journal of Counseling & Development. Journal of Counseling & Development, 91(3), 280–290. https://doi.org/10.1002/j.1556-6676.2013.00096.x

 

The authors would like to thank Cheolwoo Park for his invaluable assistance in this study. Hannah B. Bayne, PhD, LMHC (FL), LPC (VA), is an assistant professor at the University of Florida. Danica G. Hays, PhD, is a dean and professor at the University of Nevada Las Vegas. Luke Harness is a doctoral student at the University of Florida. Brianna Kane is a doctoral student at the University of Florida. Harness and Kane contributed equally to the project and share third authorship. Correspondence may be addressed to Hannah B. Bayne, 140 Norman Hall, Gainesville, FL 32611, hbayne@coe.ufl.edu.