Science Performance Assessment and English Learners: An Exploratory Study
This study examined the science achievement of English Learners as measured by three performance assessments embedded within inquiry-based units of instruction at the fifth grade level. Participants in a multi-district, NSF-funded science education reform effort developed the assessments and their associated rubrics. Scores from a member district yielded a sample of 589 students identified as either English Learners or Non-English Learners, with the former category further subdivided into the mutually exclusive groups of EXIT (recently reclassified), LEP (limited English proficient) and NEP (non English proficient). Comparison of means from the teacher-generated scores showed that, on the whole, English Learners underperformed in relation to Non-English Learners. However, contrary to previous studies using scores from traditional assessments, these differences were not found to be significant. In addition, except for the NEP subgroup on one of the assessments, all student groups performed at the proficient level on each of the three assessments. Dissimilar reading comprehension demands are discussed as a potential source of variance for student performance. While issues such as the small sample size of the English Learner subgroups and lack of inter-rater reliability information on the scores limit the strength of these findings, this study highlights the need for continued research on the use of performance assessments as viable measures of inquiry-based science for English Learners.
