Pooled estimate = $$\frac{\mbox{total number of correct responses}}{\mbox{Total number of possible correct responses}}$$