Page 61

ACEF Journal Vol 3 Issue 1 December 2012

Klocko designed to measure the ecological worldview of the superintendent as well as to gather rich demographic data used to formulate the underpinnings of the interview. The survey instrument was pilot tested and provided necessary background information regarding the superintendents’ attitudes and knowledge regarding sustainability. Results were analyzed to support the qualitative themes that emerged from the data and expressed in narrative. Face-to-face interviews enabled me to collect rich, contextual information that may not have been possible through other methods. By conducting multiple interviews at each district in this case study, I was able to construct historical patterns that took into account that the recollections may have been filtered through each interviewee’s personal lens. All interviews were digitally recorded and manually transcribed in advance of coding and categorizing. Data Inventory and Code Development Data collection and data analysis took place simultaneously, the constant comparison data analysis method (Glaser & Strauss, 1967). The interview questions thereby addressed the underlying constructs of the research question: knowledge of sustainability, attitude about sustainability, and influence in design decision-making. Data were coded through the use of NVivo 7 (see Appendix B). The analysis of these data was an ongoing and recursive process (Creswell, 2008). Patterns were grouped and clustered into major categories and subcategories. This strategy of pattern-matching “can help a case study to strengthen its internal validity” (Yin, 1984, p. 103). Drawing upon established methods of qualitative inquiry as a guide, I used coding and categorizing processes that made use of both inductive and deductive approaches to establish emergent themes from the data and make assertions about the case (Miles & Huberman, 1994; Stake, 2010) (see Appendix C). Validity The primary strategy used to ensure internal validity in this study was the use of rich, thick descriptions. Merriam (1988) advocated this type of description in order to facilitate transferability. Creswell (2008) indicated that rich, thick descriptions used to convey the findings add to the validity of the study, giving readers an element of shared experiences. Olsen (2004) argued that triangulation of multiple data sources enhances the scope and clarity of the interpretation of the data. Finally, member checking was used to validate the interpretation of the interviews. The concurrent triangulation strategy was used to integrate the results of the quantitative and qualitative data, thus promoting a means to cross-validate the findings. The quantitative data were examined through the Dunlap and VanLiere’s (1978)New Ecological Paradigm Scale (NEP). A rating was calculated on a scale of 1-5, where 3.0 would indicate a neutral response. Scores below 3.0 suggested a negative ecological worldview and scores from 3.1 to 5.0 indicated a positive worldview as determined by their responses to the NEP that was embedded in the survey instrument (see Appendix A). Because of the exhaustive use of the NEP within scientific research, it was not essential to conduct a factor analysis to establish validity of these measures included in the pre-interview survey, questions 21 through 35. Dunlap, VanLiere, Mertig, and Jones (2000) tested to determine if the 15 items can legitimately be treated as measuring a single construct and found internal consistency between item responses sufficient enough to warrant combining a set of items into a single measure of ecological worldview (Mueller, 1986). 61 Vol. 3, No. 1, 2012


ACEF Journal Vol 3 Issue 1 December 2012
To see the actual publication please follow the link above