10.9. A UNIVERSITY DEPARTMENT CASE

In one university department, each faculty member writes an annual report that includes two pages on each of the following topics: teaching, research, impact, and service. Under teaching, there is information about the quantity and the quality (student comments, syllabi, textbooks written) of the professor's work. Under research, there is information about the particular problem the professor is attacking and the success so far; two reprints of papers from the current year are appended. The impact section indicates what reviewers had to say about the work, who is working on the same topic, how often the professor has been quoted in the refereed literature and by whom, and the number of invitations for major lectures or participation at major conferences or colloquia. The service shows membership on the editorial boards of journals, on national and international committees, and on university committees. The information that is relevant to two categories is entered twice. For example, the invitation to become an editor of a journal is an index of impact, but doing that job is also recorded under service.

The reports for the whole faculty, based on these four categories, are read by a committee of nine elected from among the faculty. Having at least that many members on the committee ensures that the biases of a single judge will be avoided. Getting a substantial panel means balancing different perspectives and provides a range of expert opinion.

The ratings on each of the four criteria are converted into z-scores (z = X¯/σ), which have a mean of zero and a standard deviation of 1.0. The sum of these four z-scores constitutes the professor's evaluation. However, one can use any number of criteria (see Chapter 3 for a discussion of criteria of effectiveness, especially the following section on implementation strategy). If in a given year the department of the particular university receives, let us say, a 5 percent raise, then those with a sum of z-scores of 0.000 get 5 percent. Those with z-scores around 1.0 get 10 percent, and those with –1.0 get nothing. Of course, the formula that translates z-scores to percentages should be arrived at participatively by the laboratory. The point of this example is that this model can be adapted in a number of ways for the use of a particular lab. Members of the lab should be encouraged to take part in the discussions. For example, one might introduce job evaluation as a factor by computing the z-scores only within the same range of job evaluation scores. One might introduce gainsharing by paying some bonus to all on the basis of the whole lab. The 5 percent figure in the example might be adjusted to take lab performance into account.

One might also provide flexible benefits since different employees have different needs for money, security, status, and so on, at different points of their careers. People should also be given some choice of fringe benefits, since not everyone wants the maximum of cash as opposed to vacation time, insurance policies, and so on.