Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
It would be useful to have a goodness-of-fit measure that would both facilitate comparisons (1) among Tobit equations and (2) between Tobit equations and OLS equations. This paper suggests a number of possibilities and then provides Monte Carlo results that permit comparisons of the values of different measure. Also it is noted that point (2) suggests that a Pseudo-R[superscript]2 be chosen which is best able to mimic the OLS-R[superscript]2. Under this criterion, a modification of a measure developed by R. McKelvey and W. Zavoina is consistently superior. Copyright 1994 by Blackwell Publishing Ltd