| Abstract: | Spector and Mazzeo assert that ordinary least squares regression analysis has been misused by many economics education researchers. They explain that OLS is inappropriate for the analysis of discrete dependent variables, and they suggest the use of probit analysis instead. They then show how probit analysis can be employed in an economics education research project and compare the results of this approach with the results obtained by using OLS. (Those who want to know more about probit analysis might be interested in a new book, Carlos Daganzo's Multinomial Probit: The Theory and Its Application to Demand Forecasting, published in 1979 by Academic Press, Inc. Also see the bibliography provided by Spector and Mazzeo.) |
