The Effect of Academic Stress Upon the Anxiety and Depression

Levels of Gifted High School Students*

Mary Yadusky-Holahan William Holahan

Despite the paucity of empirical data concerning the The central goal of this study was to compare the effects
incidence and severity of anxiety and depression among of living alone or with a roommate upon anxiety and depres-
gifted and motivated students, the literature suggests that sion in the achievement-oriented, high-ability student. The
three factors encourage these stress reactions. First, these project was undertaken in an effort to isolate the effects
students create their own pressure to succeed by setting of living without a roommate on students under stress from
high, sometimes unrealistic, goals for themselves (Gissrau, multiple sources. The data will be used to assist the school
1976). &dquo;The goals of these driven achievers are always one in targeting the emotional needs of students and developing
step ahead of what they can produce, so they frequently programs to meet them.
have a sense of failing to meet their impossibly high expec- It was anticipated that all of the students would report
tations. Such students are in a constant state of stress, increased anxiety and depression in relation to increased
even though they are the winners in the system&dquo; (Bovilsky, academic pressure. In addition, higher levels of anxiety
1982). and depression were expected among students in single
Secondly, parents and teachers of the academically rooms, due to reduced peer contact and support. Students
talented also exert unusual pressures to succeed (Rothney who elected to live in single rooms were expected to create
& Koopman, 1958). The student’s actual or perceived both a physical and interpersonal distance from their peers
inability to meet the expectations of these significant and thus find themselves excluded from the type of dyadic
others may increase the potential for higher levels of anxiety friendship experienced by roommates. Moreover, the
and depression (Gowan & Bruch, 1971). personality traits which initially motivated certain students
Finally, the stress that these students experience may to live alone may also contribute to their isolation. A final
be compounded if the school they attend is residential. In hypothesis was that students in single rooms would report
April of 1981, the Independent School Health Association a greater number of age-specific problems, since one con-

sponsored a conference on stress in residential schools sequence of reduced peer contact is a reduced opportunity
to examine this problem. They concluded that the primary for social comparison.
effect of a residential environment on the high-ability,
achievement-oriented student is prolonged stress due to Method
the inability of the student to escape academic and social Sample
pressures (Bovilsky, 1982; Carapetyan, 1982). Sixty twelfth-grade students from a public residential
Peer support is a crucial mechanism for dealing with the
high school served as subjects. All of these students, by
stress and subsequent anxiety and depression created by virtue of having completed a lengthy and competitive appli-
these three factors. Adolescent friendships provide a forum cation process and subsequently being invited to attend the
for ventilating feelings which result from stress (Chaikin
school, were identified as highly motivated to achieve
& Derlega, 1974; Cozby, 1973; Rubin & Shenker, 1978). academically and as having exceptional intellectual ability.
Peers also provide a vehicle for self-evaluation, through The sample in this study consisted of thirty students
a comparison of the self to similar others. Through this
(19 male, 11 female) who had selected single rooms and
process of social comparison, unrealistic expectations thirty students with chosen roommates (18 male,
may be recognized and modified (Bleda & Castore, 1973; 12 female). All the students at the school living alone were
Evans, 1974; Festinger, 1954). included, whereas those with roommates were selected
A major forum for peer contact and support at a board-
using a standard table of random numbers. Possible envi-
ing school is the residence hall. Existing evidence suggests ronmental effects were controlled by matching an equal
that college students living in dormitories form friend- number of students having roommates with those living
ships on the basis of the proximity of their rooms (Ecklund, alone, by both residence hall and floor. Also, no pairs of
Charlis, et al., 1972; Menne & Sinnett, 1971; Priest & roommates were allowed to participate in the study, so
Sawyer, 1967). In addition, roommates become closer that roommate conflicts would not be duplicated.
friends than nonroommates (Newcomb, 1961; Rubin &
Shenker, 1978), particularly when they choose each other Instruments
rather than being arbitrarily paired (Hill & Stull, 1981; Three instruments were used in this study. Each was
Wheaton, 1974). administered to students at the school in a single group
*The authors would like to gratefully acknowledge the excellent editorial setting. The data were gathered in three separate tests
assistance of H. Scott Swartzwelder and Jane Collins. over a period of five months.

42
Downloaded from gcq.sagepub.com at UNIV OF MICHIGAN on April 19, 2015
 The Depression Adjective Check Lists (Lupin, 1981) are Table 1
thirty-two and thirty-four item adjective lists used to Mean DACL Score with Standard Deviation
measure self-reported depressive mood. Two item analyses for Student Groups at Three Testing Times
on separate male and female samples resulted in the

development of seven checklists. The four equivalent

forms of the list within the female set and the three equiva-
lent forms within the male set met the criterion of alternate
forms with high reliability. The normative sample of high
school and college students indicates that a typical mean
score for males ranges from 8.02 to 8.78 with standard
deviations from 4.77 to 5.47. A typical mean score for
female students ranges from 7.32 to 8.12 with standard
deviations from 4.77 to 5.49 (Lupin, 1981).
The IPAT Anxiety Scale (Krug, Scheier, & Cattell, 1976)
is forty-item instrument designed to provide an appraisal
a
of the student’s total anxiety level. The scale is sensitive to
change over time and provides separate scores for covert three groups: males with roommates (t = 4.30, df 34, =

and overt anxiety. The normative sample of high school .01), males without roommates (t = 1.85, df 36,
p < =

students resulted in a raw score mean of 30.0 with a stand-

p < .05), and females without roommates (t 1.86, df = 20,
=

ard deviation of 11.4 for males and a raw score mean of 33.1 .05). At testing three, significantly higher DACL scores
p<
with a standard deviation of 11.6 for females. were also reported, compared to testing one, in males with
The Mooney Problem Check List (Mooney & Gordon, roommates (t 2.47, df 34, p < .05) and in females without
= =

1950 revised) is a self-administered survey of personal roommates (t 2.34, df 20, p < .05). In comparing testing
= =

concerns. The High School Form used in this study con-

two with testing three, males without roommates had
tains 330 items, thirty in each of eleven problem categories.
significantly lower depression scores in the third testing
(t 1.69, df 36, p < .05), as did females with roommates
= =

Procedure (t 2.06, df 22, p < .05).

= =

The DACL and IPAT Anxiety Scale were administered

two days before classes began in the fall, in order to deter-
mine a baseline of anxiety and depression. A second testing,
using the DACL, IPAT Anxiety Scale, and the Mooney
Problem Check List, was administered approximately two
months later, during a time characterized by normal
academic pressure. The final testing, using the DACL and
IPAT Anxiety Scale, was scheduled two days before
semester examinations began in order to determine anxiety
and depression levels during a time of extreme academic
pressure. These will hereafter be referred to as the first,
second, and third testings, respectively.
All data-gathering sessions were held in a large recrea-
tional lounge located in the basement of one of the dor-
mitories. This was a quiet area which allowed subjects to
choose various seating arrangements optimal for their
comfort and to protect the confidentiality of their
responses.

Results

Depression Adjective Check Lists

Figure 1 and Table 1 summarize the effects of academic
stressors on mean DACL scores at chosen intervals in
the four groups of students. Mean DACL scores at testing
one closely correspond to national high-school norms. Figure 1. Mean (± SEM) depression adjective check list
Compared to these baseline results, the reported depres- scores over three testings for four student groups
sion at the second testing was significantly higher for (M - Male, F - Female).

43
Downloaded from gcq.sagepub.com at UNIV OF MICHIGAN on April 19, 2015
 No significant differences between depression scores Mooney Problem Check List
of the four groups were observed within either testing one Similarities were noted between groups regarding cate-
or two. At the third testing, however, females without
gories with the greatest and least concentration of items
roommates generated higher depression scores than both marked. For all four groups the two categories expressed
females with roommates (t = 2.49, df = 21, p < .05) and as least problematic were (1) Home and
males without roommates (t = 1.70, df = 28, p < .05). Family and
(2) Curriculum and Teaching Procedures. In the four
IPAT Anxiety Scale categories having the greatest number of concerns identi-
Table 2 illustrates the relative homogeneity of reported fied, two were common f6,~ill groups: (1) Social and Recrea-
tional Activities and (2) Health and Physical Development.
anxiety scores in the four groups of students over several
As indicated by these results, the four groups were gener-
testings. None of the groups differed significantly from
national normative high-school groups and no rise in ally in concordance as to their perceived problems.
An item analysis, a portion of which is illustrated in
anxiety scores was observed as the semester progressed.
The Pearson coefficient of correlation between mean Table 3, revealed several unexpected differences both
DACL scores and mean IPAT Anxiety Scale scores was between groups and between sexes. Females without
0.56. roommates described themselves aslosing their temper,
being nervous, feeling inferior, moody, worrying, not living
up to their ideal, and wanting more time to themselves
Table 2
much more frequently than females with roommates. A
less extreme, though similar, trend was found among males.
Mean IPAT Anxiety Scale Score with Standard Deviation
for Student Groups at Three Testing Times
Discussion
Consistent with the hypothesis of this study, the depres-
sion reported by all groups, except females with room-
mates, was significantly higher during the second testing
than during baseline. Two groups, males with roommates
and females without roommates, reported
significantly
higher depression at the third testing compared to baseiine.
In addition to the three factors discussed above, viz.,
unrealistic goals, high expectations from significant others,
and the residential setting, stressors specific to this school
Table 3 situation may have contributed to this heightened depres-
Percentage of Students in Each Group Responding sion. These stressors include a demanding academic
to Specific Mooney Problem Checklist Item workload, a change in social status through being removed
from a school where they were the brightest to one in which
they are merely average, and the feeling that it would be
an admission of failure to their
community should they
be unable to meet the rigorous standards for retention.
The group of females with roommates did not respond
to these stressors as predicted. There are several possible
explanatory factors. First, persons in distress seek out
others similar in personality in order to compare reactions
(Firestone, Kaplan, & Russell, 1973; Gerard & Rabbie,
1961; Miller & Zimbardo, 1966; Rabbie, 1963; Schachter,
1959). Secondly, women report more intimate friendships
with their roommates than do men (Hill & Stull, 1981).
Finally, female friendships involve more sharing of confi-
dences than do those between males (Booth, 1972; Rubin
& Shenker, 1978). It is possible that females undergoing
stress during times of increased academic pressure
experience an elevated need for affiliation and this need
is satisfied by the emotional support and opportunities
for social comparison provided by a roommate.
Expanding this conclusion, it is possible to explain why
females without roommates meet the predictions under
the hypothesis that students in single rooms would report

44

Downloaded from gcq.sagepub.com at UNIV OF MICHIGAN on April 19, 2015

 higher levels of depression than students with roommates. of the students to the Mooney Problem Check List provide
That is, females living alone may not be able to meet their support for this prediction in that students in single rooms
increased need for emotional support and social compari- described their temperament more negatively than the
son during times of academic stress due to the lack of students with roommates. While it is unclear whether
a roommate. these differences originally influenced their choice to live
Males without roommates, however, may not respond alone or whether these responses were the result of sub-
as clearly to peer group exclusion. Douvan and Adelson sequently reduced peer contact, such dispositions would
(1966) report that adolescent males do not form restrictive seemingly sustain any alienation from peers and increase
i friendships as do females. Rubin and Shenker (1978) found the isolation of students without roommates.
that whereas intimate emotional self-disclosures are not Females without roommates reported losing their
central to male friendships, they are to female friendships. temper, being nervous, feeling inferior, moodiness and
These factors may account for the results indicating that worrying much more frequently than did any other group.
males in single rooms were not significantly more depressed These feelings seem to be unrelated to academic demands,
than males with roommates. since a strikingly low percentage of females living alone
The prediction that depression would rise relative to report feeling that they do not spend enough time in study.
greater academic stress was only supported by data from Interestingly, 73% of the females in single rooms expressed
the females in single rooms. Curiously, the other three wanting more time to themselves. Perhaps this suggests a
groups reported less depression before final exams than desire to avoid situations with the potential for social
during the time of normal academic demands. This unex- rejection or for solitude which may have originally en-
pected trend may have been caused by an environmental couraged their choice of a single room.
factor. An unexpected winter storm closed the school for The prediction that students in single rooms would
three days, and the third testing took place the evening of report more age-specific problems was not strongly sup-
the first day the school was closed. Although this condition ported by the data. Although males and females without
may have mitigated stress by providing time for extra study, roommates consistently checked more items per category
sleep and recreation, clearly it only reduced reported than males and females with roommates, respectively,
depression to a significant degree in the group of females the differences were not significant. In fact, all four groups
with roommates. The only subjects who did not report were very similar in the categories they identified as most
lower depression at this testing were the females without and least problematic.
roommates, they actually reported significantly more
depression than during their baseline test. Conclusions
A surprising finding, inconsistent with predictions, was A preliminary
response to these data was the creation
that anxiety scores at the second and third testing remained of an affective education course to be included in the
similar to the baseline and national norms. There are sev- school curriculum. The goals of this course are to teach
eral potential interpretations of these data. First, the instru- students to recognize stress and its expression in terms
ment may not have been sensitive to the particular appre- of depression, to assist students in developing constructive
hension of these students. Secondly, although the students methods for dealing with stress, and to teach students the
were assured of the confidentiality of their responses, they communication skills necessary for building supportive
may have distorted their answers to appear socially accept- peer relationships.
able. A third explanation is that anxiety truly was not In this population, a clear link has been established
present in the students. Although anxiety is an expected between academic stress and depression. Moreover, it
reaction to stress, these students may have developed has been found that students without the support of room-
successful coping mechanisms over many years of grap- mates generally describe themselves in terms of failure
pling with the demands of themselves and others for (see Table 3, e.g. &dquo;Not living up to my ideal&dquo; and &dquo;Feeling
achievement. All of the students in the sample had had inferior&dquo;) and experience the emotional trauma associated
one or more years of residential living which may have with self deprecation.
increased their tolerance for the singular stresses of this These findings have important implications for residential
living situation. Finally, although not a problem reported high schools, colleges, and universities. They suggest the
with this instrument, the students may have formed a need to alter the physical structure of residence halls to
response set to the items, resulting in little change in self increase social interaction. Since at times the students in
reported anxiety levels. Future research might be under- this study reported depression levels almost twice the
taken using the Eight Parallel Anxiety Battery (Scheier national norm, highly motivated students under academic
& Cattell, 1960), which provides alternate forms with stress in a residential setting should be the target of coun-
known interform reliability, to guard against this possibility. seling outreach. Alert counselors should be able to antici-
The theory was advanced that students in single rooms pate predictable crisis periods, characterized by increased
might be isolated because of personality factors, in addition academic demands, and enact preventative support
to the physical ones mentioned previously. The responses measures.

45
Downloaded from gcq.sagepub.com at UNIV OF MICHIGAN on April 19, 2015
 References Krug, S. E., Scheier, I.H., & Cattell, R. B. The handbook for the IPAT
anxiety scale. Champaign, IL: Institute for Personality and Ability
Bleda, P. R., & Castore, C. H. Social comparison, attraction and choice Testing, 1976.
of a comparison other. Memory and Cognition, 1973, 1, 420-424.
Lupin, B. The manual for the depression adjective check lists. San Diego,
Booth, A. Sex and social participation. American Sociological Reuiew, CA: Educational and Industrial Testing Service, 1981.
1972, 37, 183-193. Menne, J. M. C., & Sinnett, E. R. Proximity and social interaction in resi-
Bovilsky, D. Up against the ivy wall. Independent School, 1982, 41 (3), dence halls. Journal of College Student Personnel, 1971, 12, 26-31.
51-55.
Miller, N., & Zimbardo, P. G. Motives for fear induced affiliation: Emo-
Carapetyan, F. Students under siege. Independent School, 1982, 41 (3), tional comparison or interpersonal similarity? Journal of Personality,
57-60. 34, 381-403.
1966,
Chaikin, A., & Derlega, V. Self disclosure. Morristown, NJ: General Mooney, R. L., & Gordon, L. V. Manual of the Mooney problem check
Learning Press, 1974. lists revised, NYC: The Psychological Corporation, 1950.
Cozby, P. C. Self-disclosure: A literature review. Psychological Bulletin, Newcomb, T. M. The acquaintance process. NYC: Holt, Rinehart &
1973, 79, 73-91. Winston, 1961.
Douvan, E., & Adelson, J. The adolescent experience. NYC: John Wiley, Priest, R. F., & Sawyer, J. Proximity and peership: Bases of balance in
1966. interpersonal attraction. American Journal of Sociology, 1967, 72,
Ecklund, Charlis, et al. The effects of proximity, willingness to engage in 633-649.
social interaction and sorority membership on the initial formation of Rabbie, J. M. Differential preference for companionship under threat.
friendship patterns among previously unacquainted college freshmen. Journal of Abnormal and Social Psychology, 1963, 67, 643-648.
Paper read at the Southeastern Psychological Association, Atlanta, Rothney, J. W. M., & Koopman, N. Guidance of the gifted. In N. B. Henry
April, 1972, 6 pp. (Ed.), Education for the gifted, fifty-seventh yearbook. Chicago:
Evans, J. F. Motivational effects of being promised an opportunity to University of Chicago Press, 1958.
engage in social comparison. Psychological Reports, 1974,34,
175-181. Rubin, Z., & Shenker, S. Friendship, proximity, and self disclosure,
Festinger, L. A. A theory of social comparison processes. Human Rela- Journal of Personality, 1978, 46 (1), 1-22.
tions, 1954, 7, 117-140. Schachter, S. The psychology of affiliation. Stanford, CA: Stanford Uni-
Firestone,I. J., Kaplan, K. J., & Russell, J. C. Anxiety, fear and affiliation versity Press, 1959.
with similar-state versus dissimilar-state others: Misery sometimes loves
nonmiserable company. Journal of Personality and Social Psychology,
Scheier, I. H., & Cattell, R. B. The IPAT 8-parallel-form anxiety battery.
26, 409-414.
1973,
Champaign, IL: IPAT, 1960.
Wheaton, B. Interpersonal conflict and cohesiveness in dyadic relation-
Gerard, H. B., & Rabbie, J. M. Fear and social comparison. Journal of
ships. Sociometry, 1974, 37, 328-348.
Abnormal and Social Psychology, 1961,
62,
566-592.
.Gissrau, B. Test anxiety and the performance of juveniles. German Mary Yadusky-Holahan.
Journal of Child Psychology and Child Psychiatry, 1976,
25 (6), 216-221. Address: Chatham County Mental Health Center, 1105 East Cardinal
Gowan, J., & Bruch, C. B. The academically talented student and Street, Siler City, North Carolina 27344.
guidance. Boston: Houghton-Mifflin, 1971. William Holahan.
Hill, C. T., & Stull, D. E. Sex differences in effects of social and value Address: Human Development and Psychological Services, University
similarity in same-sex friendship. Journal of Personality and Social of North Carolina at Chapel Hill, Peabody Hall, Chapel Hill, North
Psychology, 1981, 41, 488-502. Carolina 27514.
Jourard, S. M. Self-disclosure: An experimental analysis of the trans-
parent self. NYC: Wiley-Interscience, 1971.

Book Reviews

Reva Jenkins-Friedman

Book Review Editor

MAKER, C. JUNE. Curriculum Development for the ings which grapple with curriculum design theory as applied
Gifted. Rockville, MD: Aspen Systems Corporation, 1982; to the education of the gifted.
392 pp. $32.50. True, most books on the education of the gifted have a
section or sections devoted to &dquo;differentiated curriculum&dquo;
The purpose of C. June Maker’s book &dquo;is to provide but these are, with very few exceptions, atheoretical and
teachers and prospective teachers of the gifted with a discuss and describe practice. It is not that program
comprehensive handbook of theoretical and practical planners are unmindful of the critical importance of differ-
approaches to teaching these special children&dquo; (p. ix). She tiated curriculum-few terms are used more frequently-
achieved her purpose although she is much shorter on but rather that the basic significant questions of curriculum
theory-particularly curriculum theory-than she is on design are ignored or dealt with inadequately. We enrich
practice. Having said this, it must be observed that the or accelerate but we do not create total curriculum designs
field of gifted education in general tends to be short on and learning environments which would provide the gifted
curriculum theory as demonstrated by the dearth of writ- with adequate and appropriate experience for nurturing

46

Downloaded from gcq.sagepub.com at UNIV OF MICHIGAN on April 19, 2015