Contextual Analysis and Multilevel Models

Abstract

The political sociology of higher education is enjoying a renaissance in the United States, but there are few if any contextual analyses. The findings reported by Gross and Simmons (2007) in their comprehensive review of contemporary research may be paraphrased roughly as follows: Numerous recent studies focus on political differences among professors and on how the attitudes of academics differ from those of the public at large. Some on the right believe that the left has captured academia and that instruction is consequently biased. Others question these assertions, believing that academics exhibit a variety of political beliefs and that many professors do not express their personal politics in the classroom. Given the research aim of either documenting or debunking such assertions, the typical study follows the logic of a public opinion survey: It samples individual professors and compares their responses to those of the general public; the effects of institutional and departmental contexts on political beliefs are not examined in depth. Some studies exhibit methodological flaws of sampling design, questionnaire construction, measurement, data analysis, and interpretation; most often, the investigators do not examine closely the influence processes and social mechanisms that shape faculty opinion.