Estimating Stanine Scores from a Non-Random Sample: A Methodology Discussion

A methodology for estimating descriptive statistics, specifically the mean and the variance, from a sample that is not normally drawn is described. The method involves breaking the sample down into subgroups and weighting the descriptive statistics associated with each subgroup by the proportion of the population that the subgroup represents. This methodology is demonstrated using a data set of almost 4,000 children in Maine associated with the Reading Recovery program. After the calculation of means and variances, stanine scores for the spring administration of six literacy assessments are computed. (Author/RS)