EFFECTS
OF
TEST ANXIETY ON PERFORMANCE, WORRY, AND
EMOTIONALITY IN NATURALLY OCCURRING EXAMS
JERRY
L.
DEFFENBACHER’,~
AND
SHEILA
R.
DEITZ
Colorado State University
Test performance and reported anxiety levels of high and low test-anxious
subjects taking either a regular exam
or
an exam containing brief, written
relaxation instructions were compared. A consistent main effect for test
anxiety was found; high test-anxious subjects performed more poorly and
reported greater worry and emotionality than did low test-anxious subjects.
Effects for the relaxation manipulation were found only on the second of three
exams where the high anxious subjects receiving the relaxation exam format
reported less worry than the high anxiety-regular exam group. Results provide
greater external validity for the Test Anxiety Scale (TAS).
Sarason
(1957, 1959, 1961, 1963)
has shown that subjects scoring high on the
Test Anxiety Scale (TAS) obtain lower scores on aptitude and classroom tests than
subjects scoring lower on the TAS. Sarason
(1972)
and Wine
(1971)
have inter-
preted this performance deterioration in terms of selective attention. They postu-
late that evaluative stress elicits higher levels of interfering “state” anxiety such
that the highly anxious spend greater portions of their time focused on heightened
self-preoccupation and physiological arousal, a process which directs time away
from efficient task solution. Two recent laboratory studies (Deffenbacher, in press;
Sarason
&
Stoops, in press) have supported this attentional interpretation by show-
ing that highly anxious subjects under evaluative stress performed more poorly,
spent less time on task, and reported higher levels of worry, cognitive distraction,
and emotionality than either the highly anxious under reassuring conditions or
than the less anxious under evaluative stress. NO studies, however, were found in
which attentional predictions were evaluated in more naturalistic settings.
The present study was designed to assess further the external validity of the
TAS
by
evaluating the predictions of attentional theory in naturally occurring
classroom exams. Additionally, this study investigated the potentiality of brief
written relaxation instructions for improving performance and reducing anxiety of
the highly anxious. Performance, worry, and emotionality levels were compared
for high and low test-anxious subjects taking either the regular exam or the experi-
mental, relaxation format.
It
was predicted that in the normal exam condition the
highly anxious would perform more poorly and would report greater worry and
emotionality than their less anxious classmates.
METHOD
Subjects
A project investigating reactions to testing was announced to
193
students in
a
sophomore level psychology of human sexuality class. Students were informed
that they would receive an additional
5
points toward their course grade if they
completed four anxiety questionnaires during the term. Of the
172 (104
females and
‘Portions of this research were funded by a grant to the senior author from the Faculty Council
Committee
for
Research, Graduate School, Colorado State University.
2The authors are grateful to Tom Michaels and Ilarleen Stahl for their assistance in data col-
lection and analysis.
Requests for reprints should be sent
to
Jerry
L.
Ileffenbacher, Dept. of Psychology, Colorado
State University, Fort Collins,
CO
80.523.
446