TPC-Journal-V6-Issue2

The Professional Counselor /Volume 6, Issue 2 179 participants (2, 8, 9 and 17) loaded significantly onto both rotated factors and were therefore dropped from the study and excluded from further analysis (Brown, 1980; Watts & Stenner, 2012). Two rotated factors were retained, which accounted for 60% of the variance in perspectives on advocacy behaviors. Fifteen of the original 19 participants were retained in this factor solution. Q methodology uses only orthogonal rotation techniques, meaning that all factors are zero- correlated. Even so, it is possible for factors to be significantly correlated but still justify retaining separate factors (Watts & Stenner, 2012). The two factors in this study are correlated at 0.71. This correlation indicates that the perspectives expressed by the two factor arrays share a point of view but are still distinguishable and worthy of exploration as long as the general degree of consensus is kept in mind (Watts & Stenner, 2012). Constructing Factor Arrays After the two rotated factors were identified, factor arrays were constructed in PQMethod. A factor array is a composite Q sort and the best possible estimate of the factor’s viewpoint using the 25 Q sample items. First, a factor weight was calculated for each of the 15 Q sorts that loaded onto a factor. Next, normalized factor scores ( z scores) were calculated for each statement on each factor, which were finally converted into factor arrays (see Table 1). In Q methodology, unlike traditional factor analysis, attention is focused more on factor scores than factor loadings. Since factor scores are based on weighted averages, Q sorts with higher factor loadings contribute proportionally more to the final factor score for each item in a factor than those with relatively low factor loadings. Finally, factors were named by examining the distinguishing statements and interview data of participants that loaded onto the respective factors. Factor one was labeled focus on clients and factor two was labeled focus on multiple roles . Factor Characteristics Factor one was labeled focus on clients and accounted for 32% of the variance in perspectives on advocacy behaviors. It included nine participants. The demographic breakdown on this factor was: six females, three males; eight White individuals and one person who identified as multi-racial. The average age on this factor was about 51 ( SD = 10.33), ranging from 37 to 66. Persons on this factor had on average 11 years of post-master’s counseling experience ( SD = 8.6), ranging from one and a half to 31 years. Fifty-six percent of participants on this factor worked in 4-year colleges or universities, 33% in non-profit agencies, and one person worked at a community college. Factor two was labeled focus on multiple roles and accounted for 28% of the variance in career counselors’ perspectives on advocacy behaviors. It included six participants. Five participants on this factor identified as female and one identified as male. Five persons were White; one was Black. The average age of participants on this factor was almost 35 ( SD = 6.79), ranging from 29 to 48, and they had an average of just over seven years of post-master’s experience ( SD = 3.76), ranging from three and a half to 14 years. Four worked in higher education, and two worked in non-profit settings. Factor Interpretation In the factor interpretation phase of data analysis, the researcher constructs a narrative for each factor by incorporating post-sort interview data with the factor arrays to communicate the rich point of view of each factor (Watts & Stenner, 2012). Each participant’s interview was considered only in conjunction with the other participants on the factor on which they loaded. I read post-sort interview transcripts, looking for shared perspectives and meaning, in order to understand each factor array and enrich each factor beyond the statements of the Q sample. Thus, the results are reported below