•Cross validation example
In our example of predicting twelfth-grade achievement test scores from eighth-grade variables a sample of 700 students (a subset of the larger National Education Longitudinal Survey of 1988) were randomly split into two groups

The results of this cross-validation analysis with the two groups sample of 700 students indicates that the model (therefore the variables used) is good at predicting twelfth-grade achievement test scores from eighth-grade variables. The first group produced an R-squared of .55, indicating that the model explains 55% of the variance in the observed twelfth-grade achievement test scores. The second group produces a cross-validity coefficient of .53, indicating that the model explains 53% of the variance in the observed twelfth-grade achievement test scores. Since the cross-validity coefficient is close to the R-squared produced by the first group, we can see that the model was able to predict twelfth-grade achievement test scores in the second group with a degree of accuracy comparable to the first group. Additionally, the shrinkage was only 2%, indicating that the model performed even better in the second group than in the first group.