Describing undergraduate STEM teaching practices: a comparison of instructor self-report instruments

Background: Collecting data on instructional practices is an important step in planning and enacting meaningful initiatives to improve undergraduate science instruction. Self-report survey instruments are one of the most common tools used for collecting data on instructional practices. This paper is an instrument-and item-level analysis of available instructional practice instruments to survey postsecondary instructional practices. We qualitatively analyzed the instruments to document their features and methodologically sorted their items into autonomous categories based on their content. The paper provides a detailed description and evaluation of the instruments, identifies gaps in the literature, and provides suggestions for proper instrument selection, use, and development based on these findings. Results: The 12 instruments we analyzed use a variety of measurement and development approaches. There are two primary instrument types: those intended for all postsecondary instructors and those intended for instructors in a specific STEM discipline. The instruments intended for all instructors often focus on teaching as well as other aspects of faculty work. The number of teaching practice items and response scales varied widely. Most teaching practice items referred to the format of in-class instruction (54 %), such as group work or problem solving. Another important type of teaching practice items referred to assessment practices (35 %), frequently focusing on specific types of summative assessment items used. Conclusions: The recent interest in describing teaching practices has led to the development of a diverse set of available self-report instruments. Many instruments lack an audit trail of their development, including rationale for response scales; whole instrument and construct reliability values; and face, construct, and content validity measures. Future researchers should consider building on these existing instruments to address some of their current weaknesses. In addition, there are important aspects of instruction that are not currently described in any of the available instruments. These include laboratory-based instruction, hybrid and online instructional environments, and teaching with elements of universal design.