Middle School Reading Assessment: Measuring What Matters Under a RTI Framework

Abstract

This article presents a prototype reading assessment designed to measure and support component and integrated reading skills. The work is motivated by the need to create a new generation of assessments that are better aligned with the research in cognitive science, reading and learning. An assessment that synthesizes advances in these areas may prove to be more useful for educators to make informed decisions about who needs help, what help is needed, and whether the help is effective. This article describes our initial efforts to conceptualize, prototype, and test aspects of our framework in light of Response to Intervention.

Acknowledgments

This research was funded in part by grants from the Institute of Education Sciences (R305G04065 and R305F100005), and ETS Research and Development. We would also like to acknowledge collaborations and support from Strategic Educational Research Partnership (SERP) Institute and the Design Team at Harvard University and Boston University for working with our ETS team to help develop, collect, and apply assessment information in support of struggling readers. We would like to thank Rick Morgan for providing technical assistance with the analyses. We also would like to acknowledge the ETS team working on cognitively based assessments as, of, and for learning initiative. Any opinions expressed in this article are those of the authors and not necessarily of Educational Testing Service.

Notes

1. For more information on the cognitive model and the initiative that created it, “Cognitively Based Assessments, of, for and as Learning (CBAL),” please visit http://www.ets.org/research/topics/cbal/initiative

2. In order to facilitate interpretation and discussion of the results, we present the following results for all students in the three grades combined, rather than separately by grade. This is justifiable from a statistical standpoint, because the Screener scale scores were created based on the relationship between the Screener raw scores and the state test scale scores, using the same procedure for each grade. In any event, the patterns of results were similar when separated by grade.