Predicting Academic Achievement Growth among Low-Income Mexican American Learners Using Dynamic and Static Assessments.

Dynamic assessment methods, initially developed by Feuerstein in the 1970s, have been recommended as being more equitable for identifying the academic abilities of students who may not perform well on traditional assessments due to these learners' cultural, linguistic, or economic differences from the population for whom the traditional measures were developed. In this exploratory study we examined seven years' follow-up performance on a state proficiency test to determine the extent to which dynamic and static measures administered in second grade predicted academic growth among a low-income population of students of Mexican American heritage. Based on a series of multilevel models, we concluded that neither the dynamic nor static test performance was a statistically significant predictor of individual differences in growth, though confidence intervals are suggestive that a larger sample with more power might identify such a relationship. We discuss findings in the context of identification for gifted education programming.