Date of Award

2-6-2013

Document Type

Dissertation

Degree Name

Doctor of Education (Ded)

Department

Educational and School Psychology

First Advisor

Timothy J. Runge, Ph.D.

Second Advisor

William F. Barker, Ph.D.

Third Advisor

Joseph F. Kovaleski, D.Ed.

Fourth Advisor

Mark R. McGowan, Ph.D.

Abstract

The advent of No Child Left Behind led to increased teacher accountability for student performance and placed strict sanctions in place for failure to meet a certain level of performance each year. With instructional time at a premium, it is imperative that educators have brief academic assessments that accurately predict performance on high-stakes state assessments. Curriculum-based measurement (CBM) is a widely used type of assessment that can be used for benchmarking and progress monitoring. The majority of research about CBM is focused on reading, so additional research on CBM in written expression (CBM-WE) is warranted. This study examined the predictive power of brief measures of writing accuracy (CBM-WE metrics Correct Word Sequences [CWS] and Correct minus Incorrect Word Sequences [CIWS]) and writing complexity (Flesch-Kincaid Grade Level readability index) with respect to the PSSA Writing assessment. Other factors in PSSA Writing score prediction included type of writing prompt (expository or narrative), length of sample (three or five minutes), and time of year (fall or winter). Demographic variables of sex and resource availability were also considered. Archival data were collected from one elementary school in a rural district in the northeastern United States that conducted universal screenings of writing. The sample consisted of 196 students in fifth grade receiving regular education. Data were analyzed by Pearson correlation, MANOVA, ANOVA, mixed between-within subjects ANOVA, and path analysis. The main conclusions of the study were, (1) female participants performed significantly than males higher on CBM-WE metrics, (2) resource availability was not a significant factor in performance on any writing measure, (3) participants performed significantly higher when given expository prompts, (4) 5-minute expository CBM-WE measures were highest correlated with PSSA Writing performance and should be considered in combination with the indirect effect of sex mediated by CBM-WE score, and (5) F-K Grade Level readability scores are not significantly positively correlated with PSSA Writing scores and should not be used as a predictor of high-stakes writing assessment scores.

Share

COinS