TPC Journal V8, Issue 1 - FULL ISSUE

50 The Professional Counselor | Volume 8, Issue 1 applicable to the study. The final Q sample was given to participants for rank ordering during the Q sort process. Q Sort Process After Institutional Review Board approval was obtained, 25 participants completed the Q sort process. During the Q sort process, participants were prompted to reflect on their personal experiences of mentoring teaching to CEDS and then asked to rank order the 48 items in the Q sample on a forced- choice frequency distribution, shown in Table 3. Participants indicated a conscribed number of items with which they most agreed (+4) to items with which they least agreed (-4) along the distribution. Items placed in the middle of the rank order indicated statements about which participants were neutral or ambivalent. After finishing the rank ordering of items, participants were asked to provide brief post–Q sort written responses for the top two or three statements with which they most and least agreed, which were incorporated into the factor interpretations found in the results section below. Data Analysis Twenty-five completed Q sorts were entered into the PQMethod software program V. 2.35 (Schmolck & Atkinson, 2012). The PQMethod software creates a by-person correlation matrix (i.e., the “intercorrelation of each Q sort with every other Q sort”) used to facilitate factor analysis and subsequent factor rotation (Watts & Stenner, 2012, p. 97). The purpose of factor analysis in Q methodology is to group small numbers of participants with similar views into factors in the form of Q sorts (Brown, 1980). Factor analysis helps researchers rigorously reveal subjective patterns that could be overlooked via qualitative analysis. A 3-factor solution was selected to provide the highest number of significant factor loadings associated with each factor (Watts & Stenner, 2012). Factors were then rotated using varimax criteria with hand rotation adjustments in order to best reveal groupings of individuals with similar Q sorts. The factor rotations increased the total number of significant factor loadings from 17 to 20 of 25 participants, shown in Table 4. We approached analyzing and interpreting each factor in the context of all other factors to provide a holistic factor interpretation, versus favoring specific items (i.e., factor scores, +4 or -4) over others within a particular factor (Watts & Stenner, 2012). To do so, a worksheet was created from the factor array (see Table 5) for each individual factor containing the highest and lowest ranked items within the factor and those items ranked lower within the factor compared to other factors. Second, items in the worksheets were compared to participants’ demographic and qualitative responses associated with that factor in order to add depth and detail before the final step. Finally, the finished worksheets were used for constructing the factor interpretation narratives, which are written as a story containing the viewpoint of the factor as a whole. Table 3 Q Sort Forced-Choice Frequency Distribution Ranking Value - 4 -3 -2 -1 0 +1 +2 +3 +4 Number of Items 3 4 6 7 8 7 6 4 3