TPC Journal-Vol 11-Issue-1

The Professional Counselor | Volume 11, Issue 1 5 Cook, eliminate all duplicate, unclear, fragmented, or unrelated statements from the 240 concourse statements, which resulted in 160 statements. Baltrinic and Cook then used a structured sample design (Brown, 1980) to reduce the 160 concourse statements to a representative 48-item Q sample (Brown, 1980; see Appendix). Representativeness of a Q sample refers to whether the subset of items represent the broader population of statements in the concourse. Third, the 48-item Q sample was then evaluated by three experts (two supervision experts and one Q methodology expert) using a content validity index (Paige & Morin, 2016). The expert reviewers rated each of the 48 items on a 4-point scale using three criterion items: 1) Is the statement clear and unambiguous for counselor educators? 2) Is the statement clear and unambiguous for counseling practicum students? and 3) Is the statement distinct from the other statements? Scores across expert reviewers’ item ratings were averaged with only scores of 3 ( mostly ) or 4 ( completely ) indicating consensus on the content validity index. Items receiving a score of 3 or 4 were included, items receiving a score of 2 ( somewhat ) were reviewed and modified by our research team for appropriateness, and items receiving a score of 1 ( not at all ) were discarded from the sample. Accordingly, 45 items received scores of 3 or 4. Baltrinic completed additional Q sample refinements for the remaining three items that received scores of 2 ( n = 2) and 1 ( n = 1); two items were rewritten to improve clarity, one duplicate item was eliminated, and one new item was added. All refinements were confirmed by the second author before accepting the items in the final Q sample. For the final step, two of the experts completed Q sorts to ensure the final Q sample facilitated the expression of views on supervisee roles. The results of these two pilot Q sorts were not included in the data analysis. Participant Sample We followed McKeown and Thomas’s (2013) recommendations for selecting an intensive participant sample. Therefore, we purposefully selected an intensive participant sample composed of seven master’s- level clinical mental health counseling practicum supervisees, one doctoral co-instructor, and one faculty instructor; all of whom represented a purposeful sample of individuals (Patton, 2015) holding similar theoretical interests and having the ability to provide insight into the topic of investigation (Brown, 1980; McKeown & Thomas, 2013). Three of the master’s-level counseling students identified as male and four identified as female, and their ages ranged from 23 to 37 years old (M = 30, SD = 10.06). Regarding race/ethnicity, five of the counseling students identified as European American and two identified as African American. The counselor educator and course instructor identified as a European American male. He holds a PhD in Counselor Education with 5 years of counseling experience and 6 years of supervision experience. Additionally, the instructor is a licensed professional counselor and an Approved Clinical Supervisor, and he publishes regularly on the topic of clinical supervision. The doctoral student co-instructor identified as a European American female who has 3 years of clinical experience as a school counselor and 1 year of supervision experience. Data Collection After receiving IRB approval, Baltrinic collected the initial consents, demographics, Q sorts, and post–Q sort interview data. The students and course instructors ( N = 9) were asked to rank-order the 48 items under the following condition of instruction: “Select the statements with which you most agree (+4) to those with which you most disagree (-4) that represent a beginning counselor practicum student’s supervisee roles.” After completing the Q sorts, each participant was asked to provide written responses for the top three items with which they most and least agreed and were asked to comment on any other items of significance. Baltrinic obtained these post-sort questionnaires in person. The purpose of gathering post-sort data is to provide qualitative context for the factor interpretations (Brown, 1996).