Research comparing paper and online student evaluations of instruction

What research exists (point me toward references) that compares the effectiveness, validity, and security (or at least a subset of these) of paper and online student evaluations of instructors?

At my institution, a working group of faculty and administrators have proposed switching from a paper-based student evaluation system to an online system. I do not doubt that an online system will save time and money. The current system involves printing, sorting, distributing, collecting, sorting again, reading, and processing three pieces of paper for each student enrolled in each course, followed by data entry and number crunching. I lose an entire period of class time when these happen so that we can guarantee most students participate. According to the office that coordinates this effort, the time and cost involved are equivalent to one full-time employee above and beyond what people are already paid to do.

However, many faculty members have raised concerns, among which are the following:

• Participation will decrease. Only the motivated (very conscientious and very irritated) will participate, potentially skewing the results.


• In order to increase participation, instructors will be required to offer "carrots" that amount to grade inflation.


• Since all students on the roster will be notified by email, even students enrolled in the course, but who have never come to class, will be evaluating instructors. These students who have not participated, do have a right to evaluate instructors, but usually do not because it occurs during class time.


• Student evaluation of instructor surveys will no longer be anonymous. Students will have to log in with a certified log in to verify that they are who they say they are and to match them to surveys for specific classes. While this data will be "anonymized" prior to processing and distribution back to instructors, it will always exist somewhere with identifying information attached to it.

Members of the working group have responded to these and other concerns with statements beginning "Research shows..." However, as an academic, I do not believe "research" until I have seen it myself. The working group has not yet provided a list of references.