Active v. Passive

It is important to note that passive responses from customers are welcome, and may often highlight an area for library action. However, it is also the case, that such responses are like the proverbial "Ann Landers Survey." This means that it is not statistically valid, for it is not a random sample. Yet, like a suggestion box, libraries should allow such feedback.

Active surveys, on the other hand, can be quite useful. Would it be possible to select samples from the registered library users? If so, a random sample of 400 would yield statistically significant responses. Also, good surveys are not easy to produce. It would be good to do such, but with care, and as part of a learning organization. Learning about valid survey techniques, therefore, is quite important. A cycle of plan (learn how to survey), do (post the survey), check (analyze the survey), and act ( respond to results) could be useful.