Measuring and Modeling Cultural Change in Higher Education
The study presented here is an exploratory effort within a larger evaluation project of the SCALE MSP. The goal of this smaller study is to advance theory and discover fundamental processes of cultural change within IHEs. The analysis of cultural change was not included in the original research design, but instead was identified as a valuable and necessary addition to the analytical framework in order to explain emerging findings and themes. Given the small non-random sample (see below), this research is not generalizable beyond the 29 individuals included in this study.
The research design for the overall evaluation employed a cross-sectional longitudinal design within the overarching framework of a qualitative case study. This design was selected as appropriate to the research questions, which focus on multi-dimensional change processes within a complex institutional environment. These questions are:
How does the institutional context influence STEM instruction, STEM and education faculty collaborations on pre-service programs, and IHE and K-12 collaborations on in-service programs?
Have SCALE activities contributed to change in SCALE's primary goal areas? If so, how?
Under what conditions are change initiatives, including SCALE, accepted and incorporated at the institution?
In addition, the qualitative case study design was selected due to its utility in conducting empirical inquiry into a "contemporary phenomenon within its real life context, especially when the boundaries between phenomenon and context are not clearly evident" (Yin, 2003:23). The unit of analysis for this research is the individual embedded within a multi-dimensional institutional context. The data collected include in-depth semi-structured interviews and official university documents. Interviews were conducted with 29 faculty and administrators, who were selected non-randomly using snowball and purposive sampling procedures. These 29 individuals were interviewed at two points in time, some only at T1 or T2 and others at both points in time. At T1 in mid-2005, a total of 22 interviews were conducted, and at T2 in early 2007, a total of 26 interviews were conducted for a total of 40 interviews. Documents were collected and analyzed throughout the duration of the research.
The analytic methods originally proposed in this evaluation were based on a grounded theory approach to the thematic analysis of interview transcripts. After analyzing the T1 interviews, the data needed a more structured approach to classification and categorization. Thus, a coding paradigm approach from Strauss & Corbin (1990) was employed, which included a structured coding scheme comprised of the following: 1) institutional context categories (including pervasive values); 2) topics considered barriers or enablers to SCALE; and, 3) changes attributed to SCALE. Using this coding paradigm all T1 and T2 interviews were coded. I then conducted matrix queries that indicated which of the institutional context categories were most frequently cross-coded with barrier or enabler, or change attributed to SCALE codes. These frequently cross-coded passages were then reviewed to ensure the accuracy of their content, and themes that were reported by at least three respondents were included in the final analysis.
A simple reporting of the most recurrent themes is not sufficient to answer the research questions. Since the outcomes had been cross-coded with contextual factors, it was apparent that the effects of SCALE could be fruitfully situated within the institutional context. The finished causal network is a time-ordered display that posits mechanisms of change within the IHE context by linking the data points to a larger network of other variables, including SCALE program effects.
Finally, based on the recurrence of cultural norms as an important aspect of institutional life, and the interest in culture change shown by the SCALE PI and the MSP program, the processes of cultural change related to a specific SCALE activity was investigated. This activity was a series of professional development workshops for STEM faculty led by a faculty from the College of Education, focused on improving their skills in course management, student engagement, assessment, and pedagogical content knowledge.
I analyzed data on this activity using Quinn's interpretive method (1985), in which cultural models are inferred from interview transcripts. This analysis was also based on the cultural model approach of Shore (1996),which posits that cultural models exist at the collective level in the form of social behavior and artifacts and are comprised of schema that refer to specific information or propositions, and that individuals differentially internalize these "messages" in the form of personal mental models. I first analyzed in detail the codes for "pervasive values" that pertained to STEM instruction; this constituted the original cultural model for STEM instruction at the collective level. Next, I analyzed how individual participants in the professional development workshops responded to the activity, and compared their responses to the original cultural model in order to infer, in broad strokes, the change processes associated with their personal mental models. As a result, any changes are attributed not to cultural models that pertain to STEM instruction, but to personal mental models of individual respondents.
