TPC Journal V8, Issue 2 - FULL ISSUE

178 The Professional Counselor | Volume 8, Issue 2 factor against mental distress (A. E. Williams & Green, 2016). Therefore, the authors incorporated the perceived self-competence dimension of SDT to formulate CMHPCS items. To summarize, the purpose of the present study was to develop and validate a measurement instrument for appraising student and faculty members’ perceived competence for supporting college student mental health through recognizing and referring student peers to resources for mental wellness. The following research questions were posed: (1) What is the underlying factor structure of the CMHPCS using a large sample of college faculty and are the emergent scales reliable? (2) Is the emergent factor structure from the CMHPCS confirmed in a new sample of undergraduate students? and (3) To what extent do participants’ CMHPCS scores have predictive validity for whether or not they have made a student referral to the counseling center? Method Participants and Procedures Data were collected from students and faculty members at a large mid-Atlantic public university. G*Power was used to conduct a priori power analysis for the hierarchical logistic regression analyses described below (Faul, Erdfelder, Lang, & Buchner, 2007). A minimum sample size of 264 (132 in each sample) would provide a 95% power estimate, α = .05 (two tailed), with an odds ratio of 2.0. Based on the recommendations of Mvududu and Sink (2013), the researchers ensured that the ratio of respondents to each estimated parameter for the student sample (26:1) and for the faculty sample (11:1) was sufficient for factor analysis. The CMHPCS was administered to 513 university community members, including a sample of 201 faculty members and 312 undergraduate students. The sampling procedures and demographic profiles of the two samples are described in the following subsections. Faculty. Potential faculty participants ( N = 1,000) were solicited via an email list provided by the university’s Office of Institutional Research. The measure was administered to this sample using a well-known e-survey platform, Qualtrics (2017). Overall, the response rate was 21%, consistent with the response rates of previous survey research with faculty members (e.g., Brockelman & Scheyett, 2015). Of faculty respondents, 59% ( n = 118) identified as female, 40% ( n = 81) identified as male, 0.5% ( n = 1) identified as “other gender,” and 0.5% ( n = 1) did not specify their gender. The majority of participants, 81% ( n = 162), identified as Caucasian or White, followed by African American, 4% ( n = 8); Hispanic or Latinx, 4% ( n = 8); Asian, 3% ( n = 6); and multiethnic, 2% ( n = 3); while 8% ( n = 14) did not specify their ethnic background. Faculty members comprised a variety of different ranks, including adjunct instructor (29%, n = 59), lecturer (19%, n = 39), assistant professor (17%, n = 35), associate professor (18%, n = 37), and full professor (8%, n = 16), while 7.5% ( n = 15) did not specify their rank. Students. Data were collected from 312 undergraduate college students using a nonprobability sampling procedure. Over 34 days (four data collection sessions lasting 2.5 hours), the questionnaire was administered to students in the student union. These respondents ranged in ages from 18–51 ( M = 21, SD = 5), with 95% of participants under the age of 29 at the time of data collection. Furthermore, 64% ( n = 201) were females, 34% ( n = 107) were males, 1% ( n = 3) identified as “other gender,” and 0.3% ( n = 1) did not specify their gender. The college generational status of these respondents was 37% ( n = 116) first, 40% ( n = 124) second, and 23% ( n = 72) third and beyond. Ethnicities were distributed as follows: 48% ( n = 150) African American, 30% ( n = 95) Caucasian or White, 10% ( n = 30) multiethnic, 6% ( n = 19) Hispanic or Latinx, 4% ( n = 12) Asian, 1% ( n = 3) Native Hawaiian or Pacific Islander, and 0.3% ( n = 1) American Indian or Alaska Native, while 0.6% ( n = 2) did not report their ethnic identity.