PBA Home > Institutional Research & Analysis > Surveys > NSSE 2002 > Guide to Graphs

National Survey of Student Engagement 2002

Guide to Interpreting the NSSE Trend Graphs

The symbols in the NSSE trend graphs represent mean (average) scale or item responses for a particular group of students. For example, the first row labeled SC: Cognitive skills in Changes over Time, As Reflected in the NSSE Scales represents mean scale responses, by institution and class level, for 2000 and 2002 NSSE participants. The scales used in the 2002 analyses are similar to those used in 2000. However, because some items used in the 2000 NSSE administration differ from those in the 2002 administration, the 2000 and 2002 scales differ somewhat. For example, a question about examinations differed considerably in 2000 and 2002 and was therefore not included in the 2002 cognitive skills scale, even though it was included in the 2000 cognitive skills scale. All scale scores for 2000 participants were recalculated using 2002 methods for the changes over time analyses. Therefore, the results for 2002 will not necessarily match those previously posted for the 2000 analysis. To view a table containing item numbers and names for each item within each of the scales, see NSSE Item Numbers and Variable Names, by Scale.

The red diamond in the graphs denotes University of Colorado at Boulder seniors, and the blue circle denotes CU-Boulder freshmen. The blue and red letters denote freshmen and seniors, respectively, at other participating AAU public institutions. The key linking participating AAU public institutions to the letters used to denote them in the graphs is provided in a separate document, which may be released only to authorized users at CU-Boulder, and to NSSE-AAUDE liaisons at other participating institutions. For permission to access this document, please send an email to IR@colorado.edu.

NSSE items have several different response formats. For example, some items have a five-point scale, whereas others have a four-point scale. Consequently, response values for all items reported here have been re-scaled. This allows items with different response formats to be plotted on one graph that has a horizontal axis ranging from "low" to "high." The vertical bar in the center of the graph simply indicates the midpoint of that axis.

One question that people sometimes ask concerning the graphs is whether a difference between two group means is significant. The answer is not simple; it depends on whether statistical significance, practical significance, or some combination of the two is under consideration. In the graphs of the NSSE results, the difference between the 2000 and 2002 means for seniors on the item "Students not run around here" (part of the general satisfaction scale) is significant both statistically (p < .05) and in a practical sense (as measured by an effect size of .34, or 1/3 of a standard deviation). It might be helpful, when considering other results displayed in the graphs, to use the distance between the 2000 and 2002 senior means on this item as a guide: If other items have means for 2000 and 2002 that are spaced about as far apart as these are, then the change over time in mean student response that they represent may be considered noteworthy. A summary of items that have noteworthy differences from 2000 to 2002 is provided in Table 3 of Comparing CU-Boulder with Other Participating AAU Institutions, 2000 Versus 2002.

CU-Boulder NSSE 2002 Main Page

Last revision 01/08/03



PBA Home  |  Strategic Planning  |  Institutional Research & Analysis |  
Budget & Finances |  Questions? Comments? 
15 UCB, University of Colorado Boulder, Boulder, CO 80309-0015, (303)492-8631
 © 2001, The Regents of the University of Colorado

l:\ir\survey\NSSE\02\Report\graphguide.doc