Date of Award

2017

Degree Name

Curriculum and Instruction

College

College of Education

Type of Degree

Ed.D.

Document Type

Dissertation

First Advisor

Committee Chair, Dr. Samuel Securro, Jr.

Second Advisor

Dr. Lisa A. Heaton

Third Advisor

Dr. Ronald Childress

Fourth Advisor

Dr. Paula Nelson

Abstract

The central purpose of this study was to investigate the differences in test-taking motivation between large-scale high-stakes and low-stakes tests. The high-stakes test in this study was any college admission test, and the low-stakes test was the mandatory state assessment in West Virginia. Comparisons deduced the influence of the following independent variables on self-reported test-taking motivation: test stakes (high or low), future plan (college-bound or non-college bound), grade level (9 – 12), and sex (female or male).

The sample in this investigation (n=161) was taken from a West Virginia school district with a high school population of about 3,000 students. Participation resulted from random sampling of homeroom teachers and the return of affirmative parental consent forms. Although all subgroups were not represented proportionally, the data were aggregated for analysis.

The dataset for the analysis was the result of voluntary participation in a paper-and-pencil version of the Student Opinion Scale. On this survey, students self-rated their perception of test effort and test importance regarding the high- and low-stakes tests. The analysis revealed drastically higher motivation from students taking the high-stakes tests compared to students taking the low-stakes test.

A similar dramatic difference resulted from comparing the motivation of the college-bound subgroup on the high- and low-stakes tests. Further analyses yielded higher motivation from females and underclassmen when taking the low-stakes state exam.

The central conclusion is that the sample of high school participants in this study indicated higher motivation when taking high-stakes tests compared to low-stakes tests. An important implication is that confidence in state test score validity may be questionable because test scores may not be optimal. Further study is necessary to investigate the scale of this effect.

Subject(s)

Test-taking skills -- Research.

Study skills -- Research.

Download

Share

COinS