Data from 787 international undergraduate students at an urban university in the United States were used to demonstrate the importance of separating a sample into meaningful subgroups in order to demonstrate the ability of an English language assessment to predict the first-year grade point average (GPA). For example, when all students were pooled…

Automated scoring models for the "e-rater"® scoring engine were built and evaluated for the "GRE"® argument and issue-writing tasks. Prompt-specific, generic, and generic with prompt-specific intercept scoring models were built and evaluation statistics such as weighted kappas, Pearson correlations, standardized difference in…

If a student writes two essays, the score reliability can be estimated from the correlation between essays. However, if the essays are in different modes or require different skills, the reliability may be underestimated from the correlation. In Advanced Placement history examinations, students wrote one standard essay and one essay that required…