The relationship between curriculum-based measures in written expression and quality and completeness of expository writing for middle school students.

This study extended the work on curriculum-based measurement to examine the criterion-related validity of curriculum-based measures in written expression for middle school students, the differences in validity coefficients for various lengths of text, and the sensitivity of curriculum-based measures to change in student performance. Curriculum-based measures were the number of correct word sequences (CWS) and correct minus incorrect word sequences (CIWS) written in expository essays. Criterion measures were the number of functional elements in and quality ratings of student essays. Results revealed a strong relationship between curriculum-based and criterion measures.

**********

Curriculum-based measurement (CBM) is a system of measurement that can be used by teachers to monitor student progress over time and to evaluate the effects of instructional programs (Deno, 1985). Research on CBM at the elementary school level has demonstrated that simple and efficient measures can be used as general indicators of student performance in an academic area (Deno, 1985). For example, in written expression, the number of words written, the number of words spelled correctly, and the number of correct word sequences (i.e., two adjacent correctly spelled words acceptable within the context to a native English speaker) written in 3 minutes all correlate at a moderate to moderately strong level with other measures of students' writing performance. These measures include scores on standardized achievement tests, holistic evaluations of writing, and teacher evaluations of writing ability (see Marston, 1989). Further, when CBM procedures are used by teachers to monitor student progress and evaluate the effects of instructional programs, students achieve more (Fuchs, 1998).

Research on CBM has revealed that the measures used at the elementary level are not necessarily reliable and valid and at the secondary level (see Espin & Tindal, 1998). For example, in the area of written expression, simple scoring metrics such as the number of words written and the number of words spelled correctly in a limited time frame (e.g., 3-6 minutes) have not been shown to be valid and reliable indicators of general writing proficiency for secondary students. Instead, somewhat more complex scoring systems involving the use of correct word sequences (CWS) seem to be required (Espin, Scierka, Skare, & Halverson, 1999; Espin et al., 2000; Fewster & MacMillan, 2002; Parker, Tindal, & Hasbrouck, 1991 a, 1991b; Tindal & Parker, 1989; Watkinson & Lee, 1992).

CBM Writing Research at the Secondary Level

Tindal, Parker, and colleagues conducted the initial research on the development of CBM measures in written expression for students at the secondary level (Parker et al., 1991a, 1991b; Tindal & Parker, 1989). Their research pointed to the use of either the number (Parker et al., 199 lb; Tindal & Parker, 1989) or the percentage (Parker et al., 1991a) of correct word sequences as valid indicators of student performance in written expression. The CWS scores were valid at both middle and high school levels, although correlation coefficients were somewhat stronger at the middle school level than at the high school level (Parker et al., 1991a). Neither the number nor the percentage of CWS resulted in regular increases across the school year (Parker et al., 1991b). Percentage measures were seen to present unique problems with respect to growth monitoring because, by their nature, percentage measures are not sensitive to change in performance (Parker et al., 1991b; Tindal & Parker, 1989). If a student writes 10 word sequences at the beginning of the year with 5 correct, the percentage score is 50%. If that same student writes 50 word sequences at the end of the year with 25 correct, the percentage score remains 50%. No change in performance is reflected in the score.

Because of the inherent problems associated with use of percentage scores for growth monitoring, subsequent research focused on the number rather than the percentage of correct word sequences. Espin and colleagues (Espin et al., 1999; Espin et al., 2000) confirmed the validity and reliability of the number of CWS as an indicator of general writing performance and introduced a new scoring procedure, the number of correct minus incorrect word sequences (CIWS; Espin et al., 2000). In the Espin studies, both CWS and CIWS were found to correlate at moderate to moderately strong levels with holistic ratings of writing performance. Similar to the findings of Parker et al. (1991a), correlation coefficients were somewhat stronger at the middle school level (Espin et al., 2000) than at the high school level (Espin et al., 1999). At the middle school level, both CWS and CIWS were found to have acceptable alternate-form reliabilities, and validity and reliability were found not to differ as a function of type of text (story writing vs. descriptive writing) or writing time (3 min vs. 5 min; Espin et al., 2000). Effects of type of text and writing time were not examined at the high school level.