Document Type

Dissertation

Date of Degree

Spring 2015

Degree Name

PhD (Doctor of Philosophy)

Degree In

Teaching and Learning

First Advisor

John L. Hosp

Abstract

The computer to pupil ratio has changed drastically in the past decades, from 125:1 in 1983 to less than 2:1 in 2009 (Gray, Thomas, and Lewis, 2010), allowing for teachers and students to integrate technology throughout the educational experience. The area of educational assessment has adapted to the increased use of technology. Trends in assessment and technology include a movement from paper-based to computer-based testing for all types of assessments, from large-scale assessments to teacher-created classroom tests. Computer-based testing comes with many benefits when compared to paper-based testing, but it is necessary to determine if results are comparable, especially in situations where computer-based and paper-based tests can be used interchangeably.

The main purpose of this study was to expand upon the base of research comparing paper-based and computer-based testing, specifically with elementary students and mathematical fluency. The study was designed to answer the following research questions: (1) Are there differences in fluency-based performance on math computation problems presented on paper versus on the computer? (2) Are there differential mode effects on computer-based tests based on sex, grade level, or ability level?

A mixed-factorial design with both within- and between-subject variables was used to investigate the differences between performance on paper-based and computer-based tests of mathematical fluency. Participants completed both paper- and computer-based tests, as well as the Group Math Assessment and Diagnostic Evaluation as a measure of general math ability. Overall findings indicate that performance on paper- and computer-based tests of mathematical fluency are not comparable and student grade-level may be a contributing factor in that difference.

Public Abstract

With the increase in available classroom technology, it is necessary to examine how assessments administered through the use of technology compare to traditional, non-technology based methods. Paper-based and computer-based tests are often used interchangeably, depending on teacher and student comfort level with technology, therefore it is necessary to examine how paper- and computer-based tests compare.

This study examined the comparability of paper-based and computer-based tests designed to measure mathematical fluency, as well as how the factors of age, sex, and general ability in mathematics contributed to the comparability of paper-based and computer-based tests. Results showed that scores obtained through the administration of identical paper-based and computer-based tests of mathematical fluency are not comparable. There was no difference in how males and females performed, but the computer-based tests had a greater effect on fifth grade students than fourth grade students. There was also not a consistent relationship between the level of overall mathematical ability and performance on computer-based tests.

Keywords

publicabstract, Mode Effects

Pages

ix, 92 pages

Bibliography

Includes bibliographical references (pages 81-92).

Copyright

Copyright 2015 Kiersten Kenning Hensley

Share

COinS