help scoregof


scoregof -- Score test of normality for probit and bivariate probit


scoregof [, bootstrap(#)]


scoregof is a postestimation command for probit, dprobit, and biprobit. It computes a goodness-of-fit score test.


bootstrap(#) computes the p-value by bootstrapping the test using the specified number of bootstrap replications. If this option is not specified, p-values are computed using an asymptotic chi-squared distribution.


scoregof cannot be used if probit, dprobit, or biprobit was called with any of the following options: partial, offset, constraints.

After biprobit, the test still works if one of the left-hand-side variables appears on the right side of the other equation.

The test is performed on the estimation sample.

Saved results

scoregof saves the following in r():

Scalars r(scorestat) test statistic r(p) p-value


Chiburis, R. C. (2009). "Score tests of normality in bivariate probit models: Comment and implementation," Working paper, University of Texas at Austin.

Murphy, A. (2007). "Score tests of normality in bivariate probit models," Economics Letters 95(3): 374-379.

Also see

Online: [R] probit, [R] dprobit, [R] biprobit, [R] probit postestimation, [R] biprobit postestimation