Principles regarding the use of LIFT results for summative evaluation of faculty teaching
- Integrate evidence from various sources. Student feedback should be a component of the evidence for each faculty member's teaching effectiveness, but it cannot be the only evidence. The GVSU University Academic Senate has declared that student feedback should make up no more than 33% of the teaching component of personnel evaluations.
- Consider evidence from multiple LIFT surveys together, but not combined. Whether you focus on quantitative results, student comments, or both, the potential for bias is much greater in a single report than in multiple reports. Consider how each report contributes to the picture of the instructor as an effective teacher. Do not merely aggregate reports into a single "superscore" or meta-section. Look for the pattern that emerges across sections. Are their worst results in sections that students would rather not be taking? Is there evidence of improvement over time, or strength in some course types over others?
- Consider benefits and drawbacks of both quantitative and qualitative evidence. Current practice at GVSU ranges from purely quantitative to purely qualitative interpretations of student feedback, along with various hybrid stances in between.
- Quantitative - There is considerable evidence that quantitative scales from student evaluations of teaching are valid measures of instructional effectiveness -- that is, there's a nontrivial relationship between scale scores and measures of learning. The scales have some advantages over qualitative feedback in ease of use, ease of documentation, and duplicability. On the other hand, they clearly reduce a very complex concept - effective instruction - to a one-dimensional construct, and one that is subject to both random error and systematic bias. A purely quantitative interpretation bears a risk of making decisions based on random or biased factors.
- Qualitative - Compared to ordinal scales, student comments can offer much richer information about what students appreciated or disliked about courses and instructors. Comments frequently draw attention to issues or nuances that the boilerplate LIFT questions cannot get to. At the same time, open-ended comments present severe challenges related to balancing countervailing perceptions (e.g. do two "helpfuls" offset an "arrogant"?) and interpreting absent information (e.g. what does it mean if three students in a class of 40 say the instructor was "confusing"?). It falls to decision-makers to create implicit weighting schemes to decide which comments are important and which are not. Decision-makers make value judgments like that almost constantly in daily life, but for high-stakes decisions the absence of a defined criterion raises the specter of bias.
- Remember that each faculty member has the right to request that comments that they deem prejudicial be redacted from LIFT documents used for summative evaluations.