The relationship between teaching beliefs and teaching practices is understudied; often one or the other is studied independently but seldom are both interrogated simultaneously. Assuming that an instructor's teaching practices are related to their teaching beliefs, a probable assumption, the relationship between them could be more clearly evaluated if they were investigated contemporaneously.

Fishbein and Ajzen (1975) proposed a theory of planned behavior in which behavior is mediated by intention, which in turn is the product of attitude, subjective norms, and perceived behavioral controls. Upon this framework, Luft and Roehrig (2007) developed the semi-structured Teacher Beliefs Interview (TBI) protocol. While the name may suggest the protocol only interrogates a teacher's attitudes, their specific guidelines for classifying interview responses, on a continuum from traditional to reform-based, indicate that teaching beliefs are more than attitudinal and include subjective norms and perceived behavior controls.

Through a combination of interviewing, classroom observation, and instructor self-reflections, this research sought to better understand the relationship between the teaching practice and teaching beliefs of introductory geographic information systems (GIS) instructors. Six such instructors from two universities participated in this research. Each instructor was interviewed using the TBI before and after the semester. Throughout the semester, their teaching was observed at least three times using either the Reformed Teaching Observation Protocol (RTOP) or the Classroom Observation Protocol for Undergraduate STEM (COPUS). Following each classroom observation, the instructors reflected upon their teaching used a series of guiding question designed to assess the same constructs as the TBI. These data were integrated on a per-instructor basis and subsequently analyzed holistically to identify areas where teaching beliefs and teaching practices appeared to align or where they seemed to be inconsistent. The longitudinal nature of these data allowed for changes in teaching beliefs and behaviors to be identified as well.