The impact of change strategies on faculty practice
Before we delve into a discussion of how to improve reform efforts, it is helpful to first consider the
impact of past efforts. Until recently there has been very little knowledge about the degree to which
STEM faculty know about or use instructional strategies that are consistent with the research base.
However, survey‐based research in Engineering [3], Geoscience [4], and Physics [5,6] now allows us to
estimate the impact of the current wave of change efforts, which we roughly characterize as the last two
decades. During this time the NSF and other funding sources have provided substantial support for these efforts. Table 1 shows the results of these three independent surveys in terms of faculty
knowledge about alternative teaching strategies and faculty use of these strategies. On the surface
these results indicate that faculty are largely aware of research‐based instructional strategies and,
although there is a significant gap between knowledge and use, many faculty report using these
strategies.
While we are encouraged by these results, there are strong indications [5,6] that these results
significantly overestimate the actual situation, especially with respect to the level of use. For example,
in the physics survey we not only asked respondents to identify instructional strategies that they used,
but we also asked them to describe their frequency of use of a number of more specific instructional
activities (e.g., the use of small groups). For several innovative instructional strategies we had significant
numbers of self‐reported users to be able to compare the frequency of use of specific instructional
activities to the frequency suggested by the strategy developer. It was very unusual (6% ‐21%,
depending on the level of fidelity sought) for self‐reported users to report instructional activities
consistent with those advocated by the developer [6]. We also conducted follow‐up interviews with 72
survey respondents to the physics survey [7,8]. Although the analysis process has not yet been
completed, we have clearly found that an instructor’s characterization of their level of knowledge and
use (or non‐use) of a particular strategy is not reliable. For example, we have found self‐described users
of the Peer Instruction instructional strategy who knew little to nothing about the strategy [8] and selfdescribed
non‐users who knew much about the very specific ideas advocated by the developer and used
the majority of features of a formal implementation. Thus, we need to find better ways to talk about
these research‐based instructional strategies and we should be cautious about giving much weight to
self reports of knowledge or use of a particular strategy.