Higher education system governance: an exploratory study of presidential decision making in the kentucky community and technical college system

Validity and Reliability

Multiple data collection procedures, including both quantitative and qualitative, aided in reducing threats to validity, especially considering the small sample size.

Likewise, collecting data from the system president and college presidents facilitates a
comparison of presidential decision making from multiple perspectives that can highlight rival explanations.

Quantitative Validity and Reliability

The survey instrument that Ingram and Tollefson (1996) used was based on a list of 39 decisions classified as either academic, administrative, or personnel in nature by the Carnegie Foundation for the Advancement of Teaching. The commission considered the list of decisions to be representative of key policy areas in higher education governance based on a study conducted in 1982. Ingram and Tollefson obtained face validity of the instrument by comparing survey items with examples of decisions described in the literature. Furthermore, they used a modified Delphi technique in which an expert panel validated the importance of the items on the survey instrument to community college governance. The six panel members included current and former presidents of community colleges, as well as former chief executives of a state community college systems. Using a five-category Likert scale, panel members rated the importance of each item to community college governance.

Ingram and Tollefson (1996) calculated mean ratings for each item, which then served as its assigned value. The overall mean value of items was then compared to the assigned mean value for each item. Of the 37 survey items, the panel responses validated the importance of 36 items. Furthermore, the mean values for each category of decisions were calculated to determine whether panel members assigned different levels of significance to the different categories of academic, administrative, and personnel decisions. Ingram and Tollefson conducted a Kruskal-Wallis analysis of variance to test the null hypothesis that there is no difference in importance among the academic,
administrative, and personnel decision categories. The result of this test indicated that it cannot be assumed that the expert panel viewed any decision category as more important in the operation of community colleges than any of the other categories. Based on the use of an expert panel and an analysis of variance to determine whether certain types of decisions were relatively more important in the operation of community colleges that were other types of decisions, Ingram and Tollefson established validity and reliability of the survey instrument. For this reason, selection and use of the instrument was appropriate for this study.