I read the article Schneider article with interest. The article mostly criticizes how NSSE is used. Others' criticisms of the inherent reliability and validity of the test are mentioned, but are not the focus of the article.
One issue I have with NSSE is that it doesn't relate what students think and say is happening to what the college actually is doing. For example. Suppose a particular item is measuring how often students participate in class discussions, and 50% of our freshmen report low participation and 50% report high participation. What does that tell us about what the college is doing to encourage participation in class discussion? Not much.
Suppose we chose several courses that are structured to encourage class discussion, and several that are not, and compare the NSSE class discussion responses of students in the two sets of classes. That would tell us much more about how effective our efforts are to induce participation in class discussion, when we are making an effort to do so.
Imagine, for example, that 90% of the students in the encourage-participation classes report high participation, whereas only 10% of the students in the comparison classes do. These data say: (1) You know how to get students engaged in participating in discussion; and (2) You know where you would have to intervene in order to get more students participating in class discussions.
That's when the NSSE becomes useful -- at least to these tired old eyes!
Steve.