Doctor of Philosophy (PhD)



Document Type



Curriculum-based measurement (CBM) is a well established tool for formative assessment. CBM has also been used for prediction of state test scores. However, validity coefficients between math-CBM and state tests have been moderate at best (Skiba, Magnusson, Martson, and Erickson, 1986; Martson, 1989; Putnam, 1989). The purpose of the present investigation was to develop and evaluate a set of math assessments designed to measure the type of application and problem-solving objectives required on state tests. The "application" type assessments constructed for this study combined characteristics of CBM, accuracy-based curriculum-based assessment (CBA) and criterion-referenced assessment (CRA). Theses assessments were derived from state standards and matched to local district curriculum. The methodology involved obtaining validity coefficients for (a) traditional CBM assessments versus (b) newly developed assessments which incorporated applications/problem-solving with regard to "state tests" which served as a standard against which (a) and (b) above could be evaluated. The assessments examined included (a) one single skill computation assessment, (b) one multiple skill computation assessment, (c) one maze reading assessment, (d) a newly constructed applications assessment, (e) Woodcock-Johnson III math subtests, (f) Louisiana Education Program for the 21st Century (LEAP), and (g) the Iowa Test of Basic Skills (ITBS). Additionally, the various assessments were investigated relative to teacher-based indicators including (a) students' final grades, (b) teacher report of year end performance, and (c) teacher preference of math assessments. Participants included 172 first to fifth grade regular education students who were administered the CBM/CBA/CRA assessments one month following the state tests. State testing included use of the LEAP in fourth grade and the ITBS in second, third and fifth grades. Results indicated that the newly developed application assessment exhibited a stronger relationship with the criterion assessments, with students' final grades, and teacher report of year-end performance. In addition, the application type assessments were preferred over the computation assessments by all teachers. Results and limitations are discussed with regard to the construction and use of an application-oriented CBM/CBA/CRA for users needing assessment which might combine the power of formative evaluation with the ability to accurately predict performance on state tests.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Joseph C. Witt

Included in

Psychology Commons