Some Canadian universities participate in the National Survey of Student Engagement (NSSE). I should know but don’t which Canadian colleges participate in the Community College Survey of Student Engagement (CCSE). However, CSSE is administered by a different organization anyway and no session was on offer at CIRPA so I thought I’d hear what Shimon Sarraf of the Center for Postsecondary Research in Bloomington, Indiana had to say about NSSE.Side note: I’d stepped outside at lunch to take some pictures of the snow-covered mountains and met a few people who wanted me to take their photos. One turned out to be Shimon Sarraf who presented in this session. His comment: “We don’t have mountains like this in Indiana!”
NSSE focuses on behaviours more than perception than has a few about student perceptions. The intention (criticized by some) is for indirect process measures of student learning and development. Shimon had top-notch visual displays of data and concepts. Because about 95% of variation is within each institution (Maclean’s take note!), within-institution variation is informative. NSSE likes to use Box and Whisker charts to indicate not only equivalence of means but also the amount of variation between units in an institution (for example schools, departments or faculties). For example, on one measure, the means might be equivalent, but the range between 50th and 75th percentiles could be much wider for one of the units compared to another. A box and whisker chart displays this handily.
Shimon also provided a nifty display of valid (and not-so-valid) methods of multi-year analysis. based on a type of flow-chart with coded arrows and colour-coded smiley (or neutral or frowning) faces.
In terms of effect size, NSSE has moved to the following scale: 0.1=small; 0.3=medium; 0.5=large; 0.7=very large.
NSSE also provides some useful documents for interpreting results, such as an effect size guide, “lessons from the field“, and examples of university websites that display NSSE results.