TPC-Journal-V2-Issue1

The Professional Counselor \Volume 2, Issue 1 35 Narrowing the scope, the present study updates this important research inquiry. Specifically, our interests were to explore how college students, as potential consumers of mental health services, came to their understandings of counselors’ professional competence. That is, how effective do students perceive professional counselors to be? Obviously, since such perceptions are tied to students’ ultimate use of counselors’ services, the answers to the research question have significant implications for personnel working in college counseling settings. Method In accomplishing the study’s aim of assessing how college students generate their perceptions of professional counselors, we considered a number of potential research designs. While quantitative methods such as surveys would provide us with a relative breadth of understanding in this area (Patten, 1998) and this would be valuable, we believed such an approach would not be as apt as a qualitative design. Generally, quantitative approaches answer “what” or “how many” types of questions (Sarafino, 2005). However, we were more interested in knowing answers to “how” and “why” types of questions. These, by and large, are best answered via qualitative designs (Atkinson, Coffey, & Delamont, 2003). At the outset, we are explicit regarding our decision in using an atheoretical approach to the qualitative method. Significant and heated debate presently exists in qualitative circles regarding whether one should or should not use theory—and if so, what that role should be. Originally Glaser and Strauss (1967) advocated that atheoretic, inductive approaches were the only means of generating a grounded theory. Later, Strauss and Corbin (2008) purported that theory was legitimate and useable for some qualitative research designs. Glaser (1992), however, vehemently opposed this departure from the classical approach indicating that researchers must exercise disciplined restraint in holding back theory when generating or interpreting results. Obviously, we are not going to abate the controversy in this article, but we do wish to be explicit in reporting our commitment to classical grounded theory. That is, philosophically we believe that phenomenological studies such as the present one should be conducted inductively—holding theory at bay. While we understand the implications and even potential limitations of this approach, we believe it to be most apt, nonetheless. A sample of 26 students was drawn from a general psychology course (16 females and 10 males). The institution was a selective, private comprehensive university located in the Midwest. Departing from traditional criterion or purposeful sampling most often used with qualitative research (Seidman, 2006), we used random sampling for this study because we wished to enhance the external validity of our findings as much as feasible. That is, the trade-off of expanded generalizability was believed to be worth the expense of potentially less rich descriptions through specifically selected students. Since the general psychology course was part of the liberal arts core curriculum at the institution, our sample reflected a relatively wide cross-section of majors, included students who were freshmen through seniors, and participants’ ages ranged from 18 to 22 years of age. Interviews were tape recorded and later transcribed for analysis. In writing the present article, we used respondent pseudonyms for reading clarity. We utilized a semi-structure method in conducting the in-depth interviews. We used staple constructs for generating questions, but also allowed students to deviate in their replies, enabling them to tell their own stories and share perceptions inductively. Following Firmin (2006a), two waves of interviews were conducted. That is, all participants were interviewed twice—with transcription and coding occurring in between the interviews. This allowed for constant comparison of the data and dialogue among the researchers for generating potential codes. When analyzing the interviews, we used an open (Maxwell, 2005) coding, inductive process (Marshall, 1999). Since we located no studies published on this topic, axial coding was not practical and open coding was more consistent with the study’s exploratory aim. Frequent meetings among the article’s authors occurred and this process facilitated coding, providing verifications for consistency of analysis. When generating potential codes, we read through the transcripts, utilizing constant comparison methods (Bogdan & Biklen, 2007). This involved continuously comparing the transcripts to one another, looking for any repeating words, phrases, or constructs that were common among the participants. In order to keep the data manageable, some similar categories were collapsed into major categories. NVIVO qualitative research software also was used to help analyze data. This program helps to manage relatively large amounts of transcript