The Reliability of Randomly Generated Math Curriculum-Based Measurements. (September 2015)

• Record Type: Journal Article
• Authors: Strait, Gerald G.; Smith, Bradley H.; Pender, Carolyn; Malone, Patrick S.; Roberts, Jarod; Hall, John D.
• Is Part Of: Assessment for effective intervention. Volume 40:Number 4(2015)
• Page Start: 247
• Page End: 253
• Publication Date: 2015-09
• Subjects: math -- curriculum-based assessment -- curriculum-based measurement; Educational tests and measurements -- Periodicals; Special education -- Periodicals; 371.2605
• DOI: 10.1177/1534508415588075
• Languages: English
• ISSNs: 1534-5084

Abstract

Curriculum-Based Measurement (CBM) is a direct method of academic assessment used to screen and evaluate students' skills and monitor their responses to academic instruction and intervention.Interventioncentral.org offers a math worksheet generator at no cost that creates randomly generated math curriculum-based measures (M-CBMs). In this study, we examined the test–retest reliability and alternate-form reliability of four parallel, randomly generated M-CBMs designed to assess multiple arithmetic skills (addition, subtraction, multiplication, and division). The participants ( N = 283 sixth-grade students) completed each M-CBM worksheet twice during a semester. According to our results, these M-CBMs have moderate test–retest and alternate-form reliability. Applying the Spearman–Brown Prophecy Formula revealed that aggregating M-CBMs increases the reliability of these measures to acceptable levels for progress monitoring (i.e., above .80).