A validity study of total score versus strand scores for a multi-level curriculum-based mathematics test

Persistent Link:
http://hdl.handle.net/10150/288979
Title:
A validity study of total score versus strand scores for a multi-level curriculum-based mathematics test
Author:
Carriveau, Ronald S.
Issue Date:
1999
Publisher:
The University of Arizona.
Rights:
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Abstract:
The purpose of this study was to determine the degree to which the interpretation and use of test scores from a school district's mathematics test may be meaningful and valid for making instructional decisions, for measuring growth, and for making accountability decisions. The data used for the study came from six levels of a standardized mathematics test that grouped items into six specific categories to match the district's curriculum. The district's mathematics curriculum referred to the six item categories as "strands." The six strands were Number Sense, Data Analysis, Algebra, Geometry, Measurement, and Structure/Logic. The test items were grouped in each test booklet by item categories (strands) and thus formed six strand subtests. Factor analysis was used to examine the structure of each of the six test levels. Findings from the factor analysis indicated that there was more than one dimension (underlying construct) at each test level. Factor loadings were found to group by strand and not by item difficulty. Analysis of variance and correlation procedures were used to gather evidence that was confirmatory in nature to help verify the findings from the factor analysis. An analysis of variance found a significant difference between some of the strands in pairwise comparisons, which supported the findings from the factor analysis indicating that more than one construct (dimension) was being measured. Strand intercorrelation coefficients that were corrected for attenuation showed strong relationships among strands, which supported test unidimensionality. It was concluded that there was evidence to support an overall dimension called mathematics, but that there was also evidence to support other dimensions which reflected the six mathematics strands.
Type:
text; Dissertation-Reproduction (electronic)
Keywords:
Education, Mathematics.; Education, Tests and Measurements.
Degree Name:
Ph.D.
Degree Level:
doctoral
Degree Program:
Graduate College; Educational Psychology
Degree Grantor:
University of Arizona
Advisor:
Sabers, Darrell L.

Full metadata record

DC FieldValue Language
dc.language.isoen_USen_US
dc.titleA validity study of total score versus strand scores for a multi-level curriculum-based mathematics testen_US
dc.creatorCarriveau, Ronald S.en_US
dc.contributor.authorCarriveau, Ronald S.en_US
dc.date.issued1999en_US
dc.publisherThe University of Arizona.en_US
dc.rightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.en_US
dc.description.abstractThe purpose of this study was to determine the degree to which the interpretation and use of test scores from a school district's mathematics test may be meaningful and valid for making instructional decisions, for measuring growth, and for making accountability decisions. The data used for the study came from six levels of a standardized mathematics test that grouped items into six specific categories to match the district's curriculum. The district's mathematics curriculum referred to the six item categories as "strands." The six strands were Number Sense, Data Analysis, Algebra, Geometry, Measurement, and Structure/Logic. The test items were grouped in each test booklet by item categories (strands) and thus formed six strand subtests. Factor analysis was used to examine the structure of each of the six test levels. Findings from the factor analysis indicated that there was more than one dimension (underlying construct) at each test level. Factor loadings were found to group by strand and not by item difficulty. Analysis of variance and correlation procedures were used to gather evidence that was confirmatory in nature to help verify the findings from the factor analysis. An analysis of variance found a significant difference between some of the strands in pairwise comparisons, which supported the findings from the factor analysis indicating that more than one construct (dimension) was being measured. Strand intercorrelation coefficients that were corrected for attenuation showed strong relationships among strands, which supported test unidimensionality. It was concluded that there was evidence to support an overall dimension called mathematics, but that there was also evidence to support other dimensions which reflected the six mathematics strands.en_US
dc.typetexten_US
dc.typeDissertation-Reproduction (electronic)en_US
dc.subjectEducation, Mathematics.en_US
dc.subjectEducation, Tests and Measurements.en_US
thesis.degree.namePh.D.en_US
thesis.degree.leveldoctoralen_US
thesis.degree.disciplineGraduate Collegeen_US
thesis.degree.disciplineEducational Psychologyen_US
thesis.degree.grantorUniversity of Arizonaen_US
dc.contributor.advisorSabers, Darrell L.en_US
dc.identifier.proquest9927504en_US
dc.identifier.bibrecord.b3956941xen_US
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.