TPC Journal V8, Issue 1 - FULL ISSUE

48 The Professional Counselor | Volume 8, Issue 1 number given that Q methodology simply seeks to establish, understand, and compare individuals’ self-referent views expressed through the Q sort process (Brown, 1980). Participants were both conveniently sampled (n = 10) from counselor educators attending a workshop on Q methodology and purposefully sampled (n = 15) through recruitment emails sent to faculty members at several prominent counselor education doctoral programs in the Eastern (n = 7), Midwestern (n = 10), and Southern (n = 8) regions of the United States. Data were collected from participants by mailing packets that contained an informed consent, basic demographic questionnaire, Q sort, post–Q sort questionnaire, and a postage-prepaid return envelope. (Additional participant demographics are shown in Table 1). Note, we abstained from collecting certain demographic data (e.g., race, ethnicity, university type) from participants in response to their stated concerns about anonymity during data collection. Also, participants in this study were those that completed Q sorts (N = 25) versus those (N = 54) counselor educators used to generate the concourse described below. Concourse Generation and Selecting Items for the Q Sample Q methodology studies begin with creating a concourse, or a collection of thoughts or sentiments about a topic (Stephenson, 1978), which serves as the source material for selecting items for the Q sample. To generate the concourse for this study, 54 counselor educators, each with a minimum of one year of experience mentoring doctoral students in graduate teaching, were solicited on a counseling listserv (see Table 2). Counselor educators each provided 5–10 opinion statements on teacher mentorship approaches for working with CEDS in response to one open- ended question: What are your preferred approaches to mentoring CEDS in teaching? This process resulted in 432 opinion statements. However, this was too many statements for participants to rank order during the Q sort process. Accordingly, a 2 x 2 factorial design based on Kram’s (1985) career and psychosocial mentorship types and Borders et al.’s (2012) formal and informal mentoring styles Table 1 Demographics of Participants (N = 25) Age n (%) Rank n (%) 25–30 1 (4%) Full Professor 5 (20%) 31–40 7 (28%) Associate Professor 8 (32%) 41–50 5 (20%) Assistant Professor 12 (48%) 51–60 9 (36%) 61–65+ 3 (12%) Gender n (%) Tenure Status n (%) Female 13 (52%) Tenured 13 (52%) Male 12 (48%) Untenured 12 (48%) Years of Teaching Mentorship Experience n (%) 1–5 9 (36%) 6–10 3 (12%) 11–15 6 (24%) 16–20 4 (16%)