The use of error pattern analysis in the diagnosis and remediation of whole number computational difficulties

Abstract

This investigation was a field-based study of the use of error pattern analysis to diagnose and remediate whole number difficulties. The subjects for this study were 101 third, fourth, fifth, and sixth grade teachers from the Bryan Independent School District, Bryan, Texas. The subjects were assigned by the district to one of two treatment groups. Group one received inservice instruction in using error pattern analysis and nine weeks later group two received the same inservice instruction. The 2 1/2 hour inservice treatment focused on developing in teachers the ability to replicate an error pattern, determine the type of error, and select plausible remediation approaches. The instrument used to collect the dependent variable data was an investigator-constructed, criterion-referenced instrument designed to measure teachers' ability to diagnose and remediate computational errors. Validation information was provided by three experts in mathematics education. Chi-square was used to determine if untrained teachers could use error pattern analysis. One-way analysis of variance was used to determine if the inservice was effective in training teachers to diagnose and remediate using error pattern analysis. A series of t-tests was used to evaluate the effect of the inservice over time. Regression analysis was used to determine whether or not the background characteristics of level taught, years of experience, undergraduate specialization, number of degrees, hours of specialized diagnosis/remediation inservice, and courses in mathematical diagnosis and remediation had significant influence on teachers' ability to diagnose and remediate error patterns. The analysis of the data indicates that untrained teachers could not determine the type of error made or select plausible remediation approaches. The analysis of the data relative to the inservice further indicates that the inservice was significantly effective in providing immediate improvement in teacher scores. However, this effectiveness declined over the course of the investigation. Additionally, none of the background characteristics were found to have a significant linear relationship with either the diagnosis or remediation score on the third test.