Purpose: Designed to identify specific mathematical concepts in which a student is lacking.
Population: Grades 1.5 to 6.5.
Score: Four scores.
Time: (95-110) minutes.
Authors: Leslie S. Beatty, Richard Madden, Eric F. Gardner, and Bjorn Karlsen.
Publisher: The Psychological Corporation.
Description: The Stanford Diagnostic Mathematics Test (SDMT) was designed to identify those specific mathematical concepts and skills on which a student is making less than satisfactory progress. Covering the mathematical content of grades 1 through 8, the test is divided into four levels, (two grades per level) and three areas: Number System and Numeration, Computation, and Applications. It may be administered to single individuals or to groups, and permits both norm-referenced and content-referenced interpretations.
Scoring: The three subtests produce raw scores that can be converted into percentile ranks, stanines, grade equivalents, and scaled scores for each of the three subtests and for the total score.
Reliability: Reliability estimates for the subtests are presented as KR20 coefficients and alternate-form coefficients. The reliability values are acceptably high, but most users will probably be more interested in focusing on the Domain and Cluster scores for diagnostic purposes. The manual indicates that the Domain scores may be interpreted if considerable caution is used; however, few Cluster scores can be interpreted with an acceptable degree of confidence, since not many of the reliability values even reach .70.
Validity: Evidence for criterion-related validity consists of correlations between the subtests and the corresponding Mathematics Tests of the Stanford Achievement Test, all of which exceed .66. These values constitute evidence of reasonable validity for the subtests.
Norms: The national standardization sample was stratified on socioeconomic status, district enrollment, and geographical area. A comparison of the sampled districts with the total U.S. public schools on geographical area and district enrollment shows favorable agreement, but no SES data are presented in the manuals.
Suggested Uses: Recommended for use in assessing mathematical
achievement in educational and research settings.