.2 .. .1:
.,:*:.-.;.;‘
NW" ““3

v

Ti“:

w

.8 “i.

|

7.. .«Qvlu-m
-N. ..
Vow;

 

To.
«no

nwm
, Y r...-
-h- a

v
52....
>3;

a.-

L. :1

 

"QT—'12:; :

u-ws
y...

g‘

ng‘

-nw’~
mun -1: - »- ~ A
u-dm » “V

 

,,..~

A»:

 

m«—’

$6.55
9‘2" ‘ 3

J
‘ iiiéfiig‘i'i'
t. fidifiififi' °

v

Lo

. .o
«.

.Ar

22%" :

1 :
91:1.“

paw-v

3‘:

 

~m‘

mam lllllllllllllllllllllllllllllllllllllllllllillllllllllllllllll

31293 02058 6495

:7\ “l

‘1‘, {n‘ ”“1

"1.

This is to certify that the

dissertation entitled

THE ROLE OF HARDINESS AND.*‘PERCEPTIONS OF SOCIAL
SUPPORT IN THE EARLY COLLEGE ADJUSTMENT OF
INTERNATIONAL STUDENTS

presented by

S . A11 Mirzadeh

has been accepted towards fulfillment
of the requirements for

Ph.D. Counseling Psychology

 

 

degree in

Major professor

Date August 30, 1999

 

MSU i: an Affirmatiw Action/Equal Opportunity Institution 0- 12771

 

 

LIBRARY

Michigan State

Unlverslty

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECAUED with earlier due date if requested.

 

 

 

 

DATE DUE DATE DUE DATE DUE
NOV] 7 2001 (MM 2 @805
111701
APR ”‘3 “o 9033

 

m g 5 2004

___17_g_§

 

 

 

 

 

 

 

 

 

 

 

11/00 chIRC/DmoDmpfiS—p.“

 

 

THE ROLE OF HARDINESS AND PERCEPTIONS OF SOCIAL
SUPPORT IN THE EARLY COLLEGE ADJUSTMENT OF
INTERNATIONAL STUDENTS

By

S. Ali Mirzadeh

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Counseling, Educational Psychology, and Special
‘ Education

1999

ABSTRACT

THE ROLE OF HARDINESS AND PERCEPTIONS OF SOCIAL SUPPORT
IN THE EARLY COLLEGE ADJUSTMENT OF INTERNATIONAL STUDENTS

By

S. Ali Mirzadeh

Change is a constant, though often gradual, characteristic of most
environments. Environmental change is often precipitated by relational factors such
as divorce, deaths in the family, or incarceration; by social factors such as crime,
riots, and revolutions; by economical factors such as recessions and unemployment,
and by natural factors such as earthquakes and floods. It can also be produced by
experimental manipulations. Occasionally, however, significant and rapid change is
induced through the insertion of an organism into a new environment. International
students are an example of a naturally occurring population that experiences a high
degree of environmental change in just such a manner.

Change in life circumstances has been demonstrated to elicit a stress
reaction in some individuals. Research in the past twenty years has consistently
demonstrated a link between exposure to stress and the development of a variety of
physical and psychological symptoms. This association, though significant, is not
universal. The generally cited correlation index is of the order of 0.30. In the past
two decades, the roles of a range of individual differences that purportedly
influence the stress-illness link have been examined. Hardiness and perceptions of

social support have emerged in the literature as two such factors.

According to the literature, the prevalence of physical and psychological
symptoms among international students is similar to the reported rates of these
symptoms among other stress-exposed groups. This implies that although a
portion of the international student population exhibits vulnerability to the
development of stress related symptoms, another segment remains relatively
symptom free. As well as provide descriptive data on the health status of new
international students enrolled in a large mid-Westem university, this study
examined the roles of hardiness and perceptions of social support in the adjustment

of this population to environmental change.

This dissertation is dedicated to all Iranian students who left our home
country in pursuit of knowledge in the years between 1975 and 1979. As we each

stepped onto the plane at Mehrabad, little did we know there was no coming back.

ACKNOWLEDGEMENTS

Those who complete the requirements of a doctoral degree discover, both
directly and vicariously, that the role of the advisor is the single most influential
determinant of outcome. In my own case, I was blessed with an advisor who
personifies the often discussed but far less oflen observed personal qualities of
genuineness, warmth, and unconditional positive regard. Professor Robbie
Steward, by her own human-centered example, exposes the fallacy of a task-
centered approach to life.

This dissertation is yet more evidence in support of the belief that no work
is produced by a single author. The original idea of this project was conceived and
developed with the generous and expert guidance of professor Frederick G. Lopez.

During the latter stages of the project my work was immeasurably facilitated by
Professor Kenneth G. Rice’s willingness to remedy the considerable deficits in my
statistical skills. Throughout the project, I was helped, assisted, encouraged,
advised, inspired, stimulated, and ,most essentially, prodded by my wife Susan
Doyle-Mirzadeh. From permitting me to express (and thus think through) half-
baked ideas, to most ably helping me in the coding, data entry, and data analytical
stages she graciously conspired with me in my delusion that things were always
under control. In her willingness to proof-read the various drafts of the manuscript
she exhibited the same zeal and commitment to the plight of the distressed that are

the hallmarks of her work with the severely mentally ill. Finally, I wish to express

my sincere gratitude to the psychology staff at Wright-Patterson Medical Center
and in particular Lt. Col. (Dr) Ken Dobbins for graciously providing me with time

and support during my clinical residency year.

TABLE OF CONTENTS

LIST OF TABLES

CHAPTER 1

Introduction
Problem Statement

CHAPTER 2

Review of the Literature .
The Health Status of International Students .
The Relationship Between Stress and Health .
The Role of Hardiness in the Stress-Illness Link .
The Role of Social Support in the Stress-Illness Link.

The Combined Roles of Hardiness and Social Support in the

Stress-Illness ink
Questions under Investigation.
Hypotheses

CHAPTER 3

Methodology .

Participants

Procedures

Instruments
Stress .
Hardiness .
Social Support.
Adjustment .

Data Analytic Procedures

CHAPTER 4

Format.

Data Entry Issues

Sample Description

Sample Representativeness .

Variables Used 1n the Present Analyses

Internal Consistency of the Instruments Used. .
Comparisons of the Scores Obtained by Participants in the

Present Study with those Reported in Other Studies .

Frequency and Intensity of Symptoms, Syndromes, and
Critical Items . .
Descriptive Analysis of the Index of Life Stress

vii

\ju—I

18
21
23

26
28
28

30
3O
31
32
32
33
34
35
36

38
39
39
47
48
51

52

52
57

Descriptive Analysis of the Index of Social Support . . 59

Inferential Statistics and Hypothesis Testing . . . 61
Hypothesis 1 . . . . . . 61
Hypothesis 2 . . . . . . 68
Hypothesis 3 . . . . . . 73
Hypothesis 4 . . 79

The Effects of the Three Components of Hardiness upon
Adjustment . 82

Roles of Demographic Factors in the Adjustment of
International Students. . . . 98

CHAPTER 5

Discussion . . . 100
The Health Status of International Students . . 103
Significant Sources of Stress and Social Support. . 104
The Role of Hardiness in the Adjustment of New

International Students. . 105
The Role of Social Support 1n the Adjustment of New
International Students. . 109
The Roles of Native Language and Geographical Region 110
Strengths and Limitations . 1]]
Implications for Future Research and Practice. . 113
APPENDICES

Appendix A (Initial Contact Letter) . . . . 118

Appendix B (Informed Consent Form) . . . 120

Appendix C (Sign-up Form). . . 122

Appendix D (Demographic lnforrnation) . . . 123

Appendix E (SUNYA-PSC). . . . . 125

Appendix F (SCL-90-R) . . . . . 126

Appendix G (188) . . . . . . 128

Appendix H (ILS) . . . . . . 130

Appendix I (PVS- II). . . . . 132

Appendix J (Demographic Information 11) . . . 134

BIBLIOGRAPHY . . . . . . . 135

viii

LIST OF TABLES

Table
1. Sample Demographic Information.

2. Sample Social Network Information

3. Sample Education Related Information

4. Descriptive Statistics for All Variables

5.1. Correlations Among Variables Used in the Analyses

5.2. Correlations Among the Three Components of Hardiness

6. The Reliability Coefficients (Cronbach Alpha) of the Instruments .
7. [LS and 188 Scores: Present Study Vs. Other Studies.

8. SUNYA-PSC Scores: Present Study Vs. Other Studies

9. Descriptive Statistics and Paired Sample T-Tests for the Depression, Anxiety,

and Somatization Scales of the SCL-90-R
10. T-Scores on the Depression Scale of the SCL-90-R
11. T-Scores on the Anxiety Scale of the SCL-90-R
12. T-Scores on the Somatization Scale of the SCL-90-R .
13. Frequency of Critical Items
14. Frequently and Intensely Experienced Psychosomatic Symptoms
15. Salient Sources of Life Stress
15.1 Frequency of Occurrence of Items Comprising the [LS .
16. Salient Sources of Social Support.

16.1 Frequency of Occurrence of Items Comprising the 188

ix

page
41

44

46

49

50

51

51

52

52

53

54

54

55

56

57

58

58

59

60

17.1 The Amount of Variation in BDEPR Explained by Equation 1

17.2 The Evaluation of Coefficients Associated with the Predictor
Variables

18.1 The Amount of Variation in BANX Explained by Equation 2

18.2 The Evaluation of the Coefficients Associated with the Predictor
Variables

19.1 The Amount of Variation in PSCPST Explained by Equation 3 .

19.2 The Evaluation of the Coefficients Associated with the Predictor
Variables

20.1 The Amount of Variation in BSOM Explained by Equation 4 .

20.2 The Evaluation of Coefficients Associated with Predictor Variables

21.1 The Amount of Variation in BDEPR Explained by Equation 5 .

21.2 The Evaluation of Coefficients Associated with Predictor Variables

22.1 The Amount of Variation in BANX Explained by Equation 6 .

22.2 The Evaluation of Coefficients Associated with Predictor Variables

23.1 The Amount of Variation in PSCPST Explained by Equation 7

23.2 The Evaluation of Coeflicients Associated with Predictor Variables

24.1 The Amount of Variation in BSOM Explained by Equation 8 .

24.2 The Evaluation of Coefficients Associated with Predictor Variables

25.1 The Results of the Levene’s Test

25.2 The Results of the ANCOVA

26.1 The Results of the Levene’s Test

62

62

63

63

64

65

66

66

68

69

7O

70

71

71

72

72

74

74

75

26.2 The Results of the ANCOVA . . . . . 75

26.3 The Results of the Independent samples T-Test comparing the means of

BDEPR across NATLAN G1 . . . . . 75
27.1 The Results of the Levene’s Test . . . . 76
27.2 The Results of the ANCOVA . . . . . 76
28.1 The Results of the Levene’s Test . . . . 76
28.2 The Results of the ANCOVA . . . . . 76
28.3 The Results of the Independent samples T-Test comparing the means of
BANX across NATLANGI . . . . . . 77
29.1 The Results of the Levene’s Test . . . . 77
29.2 The Results of the ANCOVA . . . . . 77
30.1 The Results of the Levene’s Test . . . . 77
30.2 The Results of the ANCOVA . . . . . 77
31.1 The Results of the Levene’s Test . . . . 78
31.2 The Results of the ANCOVA . . . . . 78
32.1 The Results of the Levene’s Test . . . . 78
32.2 The Results of the ANCOVA . . . . . 78
32.3 The Results of the Independent samples T-Test comparing the means of
BANX across NATLANGI . . . . . . 79
33.1 The Results of the Levene’s Test . . . . 80
33.2 The Results of the ANCOVA . . . . . 80
33.3 The Results of the Independent samples T-Test comparing the means of
BDEPR across ZREGION . . . . . . 81
34.1 The Results of the Levene’s Test . . . . 81

34.2 The Results of the ANCOVA . . . . . 81

35.1 The Results of the Levene’s Test . . . . 81
35.2 The Results of the ANCOVA . . . . . 81
36.1 The Results of the Levene’s Test . . . . 82
36.2 The Results of the ANCOVA . . . . . 82
37.1 The Results of the Levene’s Test . . . . 83
37.2 The Results of the ANCOVA . . . . . 83
38.1 The Results of the Levene’s Test . . . . 84
38.2 The Results of the ANCOVA . . . . . 84
39.1 The Results of the Levene’s Test . . . . 84
39.2 The Results of the ANCOVA . . . . . 84
40.1 The Results of the Levene’s Test . . . . 84
40.2 The Results of the ANCOVA . . . . . 85
41.1 The Results of the Levene’s Test . . . . 86
41.2 The Results of the ANCOVA . . . . . 86
41.3 T-Test for Independent samples: Mean Depression Scores of the High
Commitment Vs. The Low Commitment Groups . . . 86
42.1 The Results of the Levene’s Test . . . . 86
42.2 The Results of the ANCOVA . . . . . 86
42.3 T-Test for Independent samples: Mean Anxiety Scores of the High
Commitment Vs. The Low Commitment Groups . . . 87
43.1 The Results of the Levene’s Test . . . . 87

43.2 The Results of the ANCOVA . . . . . 87

xii

43.3 T-Test for Independent samples: Mean Psychosomatic Symptoms Scores of

the High Commitment Vs. The Low Commitment Groups . 88
44.1 The Results of the Levene’s Test . . . . 88
44.2 The Results of the AN COVA . . . . . 88
44.3 T-Test for Independent samples: Mean Somatization Scores of the High

Commitment Vs. The Low Commitment Groups . . 88
45.1 The Amount of Variation in BDEPR Explained by Equation 25 89

45.2 The Evaluation of Coefficients Associated with Predictor Variables 90
46.1 The Amount of Variation in BANX Explained by Equation 26. 91
46.2 The Evaluation of Coefficients Associated with Predictor Variables 91
47.1 The Amount of Variation in PSCPST Explained by Equation 27 92
47.2 The Evaluation of Coefficients Associated with Predictor Variables 92
48.1 The Amount of Variation in BSOM Explained by Equation 28 . 93

48.2 The Evaluation of Coefficients Associated with Predictor Variables 93

49.1 The Results of the Levene’s Test . . . . . 94
49.2 The Results of the ANCOVA . . . . . 95
50.1 The Results of the Levene’s Test . . . . 95
50.2 The Results of the ANCOVA . . . . . 95
51.1 The Results of the Levene’s Test . . . . 95
51.2 The Results of the ANCOVA . . . . . 95

51.3 T-Test for Independent samples: Mean Psychosomatic Symptoms Scores of
the High Control Vs. The Low Control Groups . . 96

51.4 The Amount of Variation in PCSPST Explained by Equation 3 1a 96

xiii

51.5 The Evaluation of Coefficients Associated with Predictor Variables 97

52.1 The Results of the Levene’s Test . . . . . 98
52.2 The Results of the ANCOVA . . . . . 98
53.1 The Results of the Levene’s Test . . . . 99
53.2 The Results of the ANCOVA . . . . . 99

53.3 T-Test for Independent Samples: Mean Psychosomatic Symptoms Scores of
the With Acquaintance in the Lansing/East Lansing area Vs. No
Acquaintance in the Lansing/East Lansing area Participants . 99

CHAPTER 1

INTRODUCTION

Change in one or more areas of a person's life has been cited as a major source
of stress (Brett, 1980; Brown & McGill, 1989; Hinkle, 1974; Holmes & Rahe, 1967;
Miller & Sollie, 1985; Rahe & Arthur, 1978; Zeiss, 1980). Life change has been
defined as "any noticeable alteration in one's living circumstances that requires re-
adjustment” (Weiten & Lloyd, 1994, p. 73). Empirical work in this area has
consistently shown that people are more likely to exhibit physical and psychological
problems concurrent with and/or subsequent to a major life change (Dohrenwend &
Dohrenwend, 1974).

Soon after the initial demonstrations of the relationship between stress and
illness, empirical attention was focused upon explaining the repeated observation that
many individuals maintain good health even when subjected to prolonged exposure to
stress (Adler & Matthews, 1994; Hinkle, 1974; Holohan & Moos, 1985).
Accordingly, investigators have sought to study individual differences that might
explain why some people are more vulnerable to the deleterious effects of stress than
others (e. g., Cohen & Edwards, 1989; Contrada, Leventhal, & O’Leary, 1991;
Holroyd & Coyne, 1987; Suls & Rittenhouse, 1987). In the late seventies, Suzanne
Kobasa observing the small but significant correlation between stress and illness, began
to conceptualize and investigate personality factors that according to a set of

theoretical reasons, she believed, explained some of the variation in the stress-illness

relationship.

Kobasa’s work was grounded in at least two theoretical domains. First, her
conceptualization tendencies were based upon her own and her graduate school
advisor, Salvatore Maddi’s allegiance to an existential theory of personality (Kobasa &
Maddi, 1977). A basic tenet of existential personality theory is that the development of
a healthy personality is a function of the capacity to stand on one’s own convictions
(May, 1969c, p.243). Furthermore, according to this theory, an individual’s ability to
mobilize his or her resources in order to achieve personally meaningful goals (a
concept known as “will” or “intentionality”) is also an important factor in the
development of a healthy personality (May, 193971967a). Second, Kobasa’s work was
influenced by Richard Lazarus’ (1966) formulation of stress as a reaction mediated by
an individual’s appraisal of environmental change. More specifically, Lazarus
conceived of the stress reaction, a term first coined by Henry Selye in 1956, as a
physiological response to the perception of a threat to one’s well-being. Whether or
not an event was perceived as threatening was based upon personal belief systems
(Lazarus, 1966). Lazarus’ work was partially based upon Magda Amold’s (1945)
theory which suggested that organisms instinctively and rapidly appraise their
environments “as a fundamental act of perception, producing tendencies to act” (cited
in Fiske & Taylor, 1991, p. 436). In a refinement of Amold’s work, Lazarus (1966)
introduced the concepts of primary appraisal, evaluation of the relevance of a stimulus,
and secondary appraisal, evaluation of available coping resources.

In line with these theoretical underpinnings, Kobasa (1979) predicted that

psychologically “hardy” individuals would be less vulnerable to the harmful effects of
stress. Principle component anaylses conducted subsequent to a series of studies in
which she compared personality characteristics of hardy individuals with those
exhibiting vulnerability to stress, revealed a set of three personality factors which
together explained a significant portion of the variation in the adjustment of her sample
to stress. She, therefore, defined hardiness in terms of these three personality
characteristics which she labeled as commitment, challenge, and control. Commitment
is defined as the “tendency to involve oneself in (rather than experience alienation
from) whatever one is doing or encounters” (Kobasa, Maddi, & kahn, 1982, p. 169).
Challenge is defined as a “belief that change rather than stability is normal in life and
that the anticipation of changes are interesting incentives to growth rather than threats
to security” (Kobasa et al., 1982, pp. 169-170). Control is defined as a “tendency to
feel and act as if one is influential (rather than helpless) in the face of the varied
contingencies of life” (Kobasa et al., 1982, p. 169). Maddi and Kobasa (1984) have
suggested that hardiness is an acquired personality characteristic, and as such is subject
to change.

Although the current empirical status of hardiness will be reviewed elsewhere in
this manuscript, it is appropriate to state here that since its introduction the hardiness
construct has received a considerable amount of empirical attention. Kobasa’s own
initial work demonstrated the influential role of hardiness in the stress-illness link
(Kobasa, Maddi, & Courington, 1981; Kobasa, Maddi, & Kahn, 1982;1(obasa, Maddi,

& Puccetti, 1982b; Kobasa & Puccetti, 1983). In a review of the literature, Funk

(1992) found that the role of hardiness had been studied in such diverse areas as
cardiovascular reactivity, development of depressive symptoms, burnout, and the levels
of immunoglobulin-A in the breast milk of nursing mothers; and among such
populations as nurses, teachers, single parents, attorneys, and bankers. Although the
bulk of empirical findings have been consistent with Kobasa’s own discoveries, in
recent years several conceptual and methodological criticisms have been leveled against
this line of research (Benishek, 1994; Funk & Houston, 1987; Hull, Van Treuren, &
Vrrnelli, 1987). These criticisms will be reviewed in chapter 2.

A parallel line of research has examined the role of social support as a
mitigating factor in the experiencing of harmful effects of stress (Anotonsovsky, 1979;
Cassel, 1976; Gentry & Kobasa, 1984; Johnson & Sarason, 1979; Leavy, 1983; Rabkin
& Streuning, 1976). The possible role of social support in psychological health was
first discussed within the sociological literature (Mechanic, 1974). Early empirical
work in this area consisted mainly of studying the role of social support as a mitigating
factor against stress-related physical and psychological disorders (e. g., Cassel, 1976;
Cobb, 1976; Dean & Lin, 1977). The definition of social support has been the subject
of some disagreement (Gentry & Kobasa, 1984). House (1984), for example,
conceived of social support as a “flow of one or more of four things between people”
(cited from Gentry & Kobasa, 1984, p. 93). These consisted of emotional support
(e. g. expressions of affection, interest, and concern), appraisal support (e. g. helping
people think through their problems and generate solutions), informational support

(e. g. offer advice and new c0ping strategies), and instrumental support (e. g. offering a

place to stay, lending money, etc). Recent studies, however, have tended to
operationalize social support in terms of the individual’s perceptions of the availability
and utility of such support (Cohen & Hoberman, 1983; Gore, 1978). With some
exceptions (e. g. Kobasa & Puccetti, 1983) the main weight of empirical evidence
supports the view that absence of social support is associated with increased
psychological distress. Less certain, however, is the buffering properties of social
support against the adverse effects of stress (Leavy, 1983).

Although research with populations such as those reported by Funk (1992) is
of general interest to counseling psychologists, research concerning the adjustment of
college students is of particular interest because a) counseling psychologists routinely
help normal functioning individuals who are experiencing mild, moderate, or severe
distress in response to life change; b) this line of research places emphasis upon person-
environment interactions, rather than the exclusive focus on either the person or the
environment; and c) counseling psychologists are often involved in helping
relationships with both domestic and international students who have, in pursuit of
educational goals, geographically relocated.

As well as representing a large source of firnds for various institutions of higher
education in Europe and the United States (Annual Report of the Office for
International Students and Scholars, 1996), international students, by and large,
represent the intellectual and social elite of their countries and frequently return to
positions of authority and influence (Hull, 1978). In terms of size, international

students continue to evolve into a significant segment of the US. college student

population. Torrey (1970) reported that whereas in 1930 the number of international
students in the US. was less than 10,000, by the end of the war this number had
already climbed to about 20,000, and it exceeded 100,000 in 1968, representing
students from over 150 countries. More recent data indicates that over 416,000
international students were enrolled in U. S. colleges and universities in 1991 (U .8.
Dept. of Education, 1993 as reported in Yang & Clum, 1995). Furthermore, 11% of
all master’s degrees and 23.4% of all the doctoral degrees awarded by US. colleges
and universities are earned by international graduate students (US. Dept. of
Education, 1993 as reported in Yang & Clum, 1995).

Studies addressing the overall health status and the prevalence of stress related
symptoms among international students have been rare and flawed. The meager
available data have been equivocal: studies have variably described this population as
one particularly vulnerable to the development of a range of physical and
psychological symptoms (e.g., Sam & Eide, 1991; Fumham & Trezise, 1983), or
alternatively as a resilient population, which in many ways, may be better adjusted than
its domestic cohort (e.g., Parr, Bradley, & Bingi, 1992; Allen & Cole, 1987). Studies,
however, have not attempted to explain this variation. Indeed the present author was
unable to locate an article which gave a systematic review of the available literature.
Few studies have focused on the adjustment characteristics of international students in
the early stages of their sojourn. This is of particular interest in this study since the
stressors associated with this early stage are almost exclusively due to environmental

changes, whereas stressors present during later stages may be due to such factors as

pressures to conform or perform, conflicts, frustration, etc. Finally, no studies have
attempted to explain the variation in international student health by investigating
personality factors.
Emblem Statement

The constant increase in the size of the international student population and the
paucity of empirical data regarding the health status of this population in their new
environment have created a need for a more systematic investigation of this area.
There is also a need to examine, in this population, the role of factors that have been
demonstrated to influence the stress-illness association in domestic samples. Given the
current state of uncertainty about the role of individual differences in the association
between stress and illness, there is a need for the kind of data that might allow for
added conceptual and empirical clarity. The present study will address some of the
gaps in the literature concerning the adjustment of international students by examining
the role of factors, such as hardiness and perceptions of social support. Specifically,
the primary objectives of this study are to: a) assess the adjustment of new
international students in terms of their physical and psychological health; b) examine
the role of hardiness in the adjustment of new international students; c) examine the
role of perceptions of social support in the adjustment of new international students; (1)
examine the combined effects of hardiness and stress in the adjustment of new
international students; e) examine the combined effects of perceptions of social
support and stress in the adjustment of new international students; and f) assess the

possible of effects of such demographic factors as language and ethnicity upon the

adjustment of new international students to their new environments.

Should the results of this study reveal high prevalence of physical and
psychological distress among new international students, and should the results of this
study support the view that individual differences explain a significant amount of the
variation in the health status of this population, research can then turn to the creation
of empirically based programs designed to facilitate the adjustment of new international
students. In case, however, that the results of this study reveal low prevalence of
physical and psychological distress among this population, firture research can focus on
comparing the characteristics of this apparently stress-resistant population with other
stress—exposed groups who have traditionally not fared as well. Finally, should the
results of this study reveal high prevalence of physical and psychological distress
among this population, and should the results of this study indicate an insignificant role
to hardiness and perceptions of social support, research can then turn to the
examination of the role of other variables, such as neuroticism, self-esteem, or

openness to experience, in the stress-illness link of this population.

CHAPTER 11
REVIEW OF THE LITERATURE
The Hg_al_th Status 21' International Students

Travel to intellectually dominant countries by scholars and students has been a
constant feature of scholarly life since the time of the Greeks (Pruitt, 1978). Indeed,
as Wehrly (1988) points out such universities as Paris and Salerno have been enrolling
international students since the 12"‘ century. The United States has received and
hosted students from other countries since 1784 (Hendricks & Skinner, 1977).
According to Marion (1986), the appeal of the US. as a popular educational
destination is based upon a) the advanced state of science and technology; b) English
as the language of instnrction; c) the generally high quality of instruction; and (1) social
and political stability.

“fith a few notable exceptions, the subject of the health and adjustment of
intemational students began to draw empirical attention in the early sixties. At the
1961 meeting of the American College Health Association, Lyle Ward described his
observation of a set of physical and psychological symptoms prevalent among
international students. These symptoms included somatic complaints, adoption of a
passive and withdrawn attitude, and a marked reluctance to communicate. In more
severe cases, according to Ward, the symptoms mentioned above were accompanied
by a general disheveled appearance along with a restriction of physical movements.
Although these observations are consistent with signs and symptoms of clinical

depression, Ward applied the label of Foreign Student Syndrome to this cluster of

symptoms.

Ward’s observations were preceded by Lysgaard’s (1955) U-curve hypothesis.
Though not exclusively intended for the international student population, the U-curve
hypothesis suggests that the process of adjustment of a sojourner to his or her new
environment is described, graphically, by a curve resembling the letter U: a sojoumer’s
initial experiences in his or her new environment are often pleasant and exciting. This
state of affairs, however, begins to deteriorate as the sojourner encounters difficulties
arising out of his or her unique situation. Finally, as the sojourner expands his or her
repertoire of adaptive skills, his or her adjustment along with his or her sense of well-
being improves. In a continuation of Lysgaard’s work, Oberg (1960) proposed the
term “culture shock” to describe a series of four stages that, according to his
observations, people abroad experienced. Oberg suggested that the initial period of a
sojoumer’s experience, the “Honeymoon” stage was characterized by a sense of
excitement and hopeful anticipation. This short lived state of affairs is followed by a
period of “crisis” when the realities and nuances of the new environment lose their
sense of wonder and are instead, are perceived by the sojourner as hurdles with which
the sojourner must grapple and, in order to adjust, overcome. Oberg offers a list of
these hurdles, or potential sources of stress: racial discrimination, language difficulties,
housing difficulties, separation reactions, financial problems, dietary differences, and
loneliness. During the third, or the “Recovery” stage, the sojourner’s initial affective
reactions begin to give way to a sense of purposefirl mobilization of psychological and

physical resources. Finally, Oberg suggests that a sojoumer’s recovery from culture

10

shock is concluded during the “Adjustment” stage.

In a review of empirical literature on the viability of the U-Curve hypothesis
Church (1982), concluded that the evidence for the U-Curve hypothesis was “weak,
inconclusive, and overgeneralized” (Church, 1982, p. 571). More recently, Nash
(1991) has suggested that the failure to find empirical support for the U-Curve is more
due to the methodological flaws in the studies than the validity of the U-Curve
concept. The main thrust of Nash’s criticism is directed towards sampling procedures
employed by the early studies. First, Nash suggests that the U-Curve was intended to
describe neither the experiences of tourists (since, for obvious reasons, this group is
never obliged to adjust to the demands of the new environment), nor those who
terminate their stay prematurely. Second, Nash claims that the cross-sectional designs
used by the earlier investigators simply do not yield the kind of data necessary for the
accurate description of the course of individual sojoumer’s experience in his or her
new environment. This criticism was echoed by Zheng and Berry (1991), who also
proposed that longitudinal data, collected through more sensitive instruments, were
necessary for the definitive testing of the U-Curve hypothesis.

In his own study of the U-Curve hypothesis, Nash (1991) employed a
longitudinal design to investigate the course of adjustment of a group of American
college students enrolled in a “well-established Junior-Year—Abroad program” (Nash,
1991, p.284). His control group consisted of a group of students enrolled in the same
university, but who were not making the trip abroad. Nash’s findings did not support

the U-Curve hypothesis: there were no significant differences between the

11

psychological well-being of the students studying abroad and those studying at home.
In his discussion of the results, Nash questions the representativeness of his sample in
that their stay abroad was firlly coordinated and organized by the staff of the Junior-
Year-Abroad program to such a degree that the students were either not obliged to
adjust and adapt to their new environment, or any adjustments were greatly facilitated.
In line with Ward’s observation and description of the Foreign Student
Syndrome, Ray (1966), presented data suggesting that the hospital admission rates
among the international students enrolled at a major mid-Westem university were
significantly higher than those of the domestic students. She also presented data
indicating that her figures were in line with those reported fi'om other universities
including those in the United Kingdom. Nearly two decades later, Gunn (1985)
described the effects of education abroad on international student health as so grave as
to be worthy of immediate attention and intervention by the World Health
Organization. Largely as a response to Gunn’s claims, Allen and Cole (1987) reviewed
the medical charts of a group of international students enrolled in Australian
universities and compared them to those of a group of Australian students enrolled in
the same universities. Their results did not yield any significant differences in the
severity or nature of health complaints between the two groups. In fact, interestingly,
this study yielded a borderline significant indication that the international students may
enjoy better health than their domestic cohort. Among the most serious limitations of
this study was its use of charts as a source of data: it is possible that, due to cultural

factors, the international student group was more reluctant to seek health care than its

12

domestic cohort (F umahm & Trezise, 1983). Furthermore, the group of international
students used in this study consisted mainly of students from Singapore, Malaysia, and
Hong Kong; countries which, by virtue of their historical ties to the British
Commonwealth, share a number of cultural and linguistic characteristics with Australia.
The role of cultural differences in the adjustment of international students was
hypothesized, though not investigated, by Akka (1967). He predicted that higher
levels of cultural discrepancy between that of the international student and that of the
host nation would be associated with higher levels of pathology among this population.
As examples of areas of cultural discrepancy, Akka cited difliculties in communication
arising from both language skills and norms of interpersonal interaction, family ties,
academic problems, social behavior, and religion.
The bulk of the empirical evidence has been supportive of Akka’s prediction

(e. g., DeArmond, 1983; Domingues, 1970; Miller & Harwell, 1983; Zurin & Rubin,
1967; and Sam & Eide, 1991, Klineburg & Hull, 1979; Wehrly, 1986). Specifically, in
a survey of the health complaints of international students enrolled at 27 large
universities across the US, Brislin (1981), found that students from European
countries, and students with better language skills reported fewer health concerns as
compared with students from Third World countries, and students with poorer
language skills. In a review of the literature, Church (1982) found that nationality and
prior travel experience were important variables in predicting the adjustment of
international students. Some research, however, has failed to reveal any such

relationship (e. g. Fumham and Trezise, 1983). Though the results of Fumham and

13

Trezise’s study did indicate a significantly higher level of psychological disturbance
among international students as compared to domestic students, the within group
analysis of the international student data according to nationality, did not indicate any
significant differences.

In a study of the mental health of international students enrolled in Taiwanese
universities, Ko (1979) administered a questionnaire designed to measure such
variables as suspiciousness, hypochondriacal tendencies, asocial tendencies, self-
esteem, feelings of inferiority, anxiety, obsessive-compulsive tendencies, psychosexual
inhibition, hostility, ego-strength, dependent tendencies, and life satisfaction. Ko
compared the results obtained from his sample with that of its domestic cohort. His
results indicated significantly higher levels of suspiciousness, feelings of inferiority,
obsessive-compulsive tendencies, psychosexual inhibition, and significantly lower levels
of satisfaction with life among the male international students. Among the female
international students, Ko’s results indicated only significantly elevated rates of
suspiciousness, obsessive-compulsive tendencies, and dependent tendencies. It is
important to note that Ko obtained these results even though his sample of
international students, while from abroad, consisted of ethnic Chinese students sharing
much cultural and linguistic characteristics with their host country. An obvious
limitation of Ko’s study was its failure to assess the pre-departure levels of these
characteristics among his sample.

Ko’s findings were replicated in a study by Oei and Notowidjojo (1990). They

compared the effects of life change in the adjustment of international students enrolled

14

in an Australian university with that of a group of Australian students. Their results
indicated that international students with lengths of stay greater than one year were
significantly more likely to experience clinical depression as compared to either the
domestic students or newly arrived students. This finding, though cross-sectional in
nature, can be construed as supportive of Lysgaard’s U-curve hypothesis.

Miller and Harwell (1983), conducted a survey of the health problems of
international students enrolled at a small mid-Western university. The results of this
study indicated that fatigue, homesickness, headaches, colds, and insomnia were the
most common health problems among the sample. This study did not employ a control
group of domestic students, nor did the investigators compare their results with the
prevalence rates of these complaints on the campus as a whole and so no conclusions
regarding the significance of the findings can be drawn.

Other authors have conducted similar, though more internally valid studies.
Ebbin and Blankenship (1986) for example, reviewed and compared the medical
records of both international and domestic students enrolled at a major urban university
located in western United States between 1980 and 1983; a time span during which
this university, reportedly, had the highest rate of enrollment of international students
among US. colleges and universities. As well as reporting the top 27 presenting
problems of international students (the top five being common cold, pharyngitis, acne,
sprain/strain, and patient education, respectively), this study found that the prevalence
of such potentially stress related disorders as chest pain, cough, gastritis, back pain,

anxiety and depression (e. g., Dohrenwend & Dohrenwend, 1974) was significantly

15

greater among international students as compared to their domestic cohort. The latter
. finding was fiirther supported when these investigators conducted a survey of 476
college health center directors (Ebbin & Blankenship, 1988). A response rate of 38%
to the survey, from institutions with an average enrollment of 11,120 and a minimum
international student enrollment of 200 yielded results that were generally in line with
these authors’ earlier findings from their 1986 study: whereas the prevalence of such
potentially stress-related disorders as anxiety, gastritis, headache, constipation,
insomnia, depression, chest pain, abdominal pain, and peptic ulcers was found to be
significantly higher among the international students; the prevalence of other stress-
related disorders such as hypertension, low back pain, amenorrhea, and neurodermatitis
was found to be lower among the international students enrolled in the surveyed
institutions as compared to their domestic cohort.

A prevalent methodological issue in this literature is that most investigators fail
to assess the international students’ pre-departure or its immediate post-arrival health.
Observing this problem, Sam and Bide (1991) conducted a survey designed to assess
the mental health of international students taking into account their pre-departure
symptoms. The participants in this study consisted of all international students enrolled
at a Norwegian university. In addition to a demographic questionnaire, the survey
package consisted of a an inventory of adjustment and health derived fi'om Goldberg’s
(1972) General Health Questionnaire and Derogatis’ (1973) Symptom Check List. A
return rate of 67%, yielded 400 completed packages fi'om a wide range of students

including 118 females, 190 males, and 50 nationalities. Results of this study included

16

the following. First, a t-test of the means of the students’ self evaluation of their
current and prior health indicated a significant deterioration of this group’s sense of
well-being. More specifically, analysis of responses to the inventory revealed
significant change in Zlof the 32 items. There were significant increases in the levels of
nervousness, chest pains, suspiciousness, guilt, loneliness, sadness, worry, anhedonia,
heart palpitations, insomnia, feelings of worthlessness, and fatigue; as well as poorer
appetite. Second, it was found that four factors, paranoia, anxiety, depression, and
somatic complaints, accounted for more than 45% of the variance. All four syndromes
exhibited significant elevations in their pre and post-arrival levels. Finally, their results
indicated that higher levels of mental health were found among the Scandinavian and
North American students, whereas being single or married but living away from
spouse, being younger, being female, being an undergraduate, being Asian, and being
Arabic speaking were all found to be associated with deterioration of mental health.
Overall, research on the health status of international students tends to suffer
from the following limitations. First, the lack of control groups in these studies
severely limits the range of conclusions that could drawn. Second, a large number of
these studies suffer from sampling problems that severely limit the generalizability of
their results. These include narrow range of nationalities represented; the presence of
high levels of similarities between the host culture and the cultural characteristics of the
sample; and the problems that often plague survey data e. g. volunteer bias. Third,
with the exception of the Sam and Bide (1991) study, most studies designed to assess

the health status of international students fail to take into account pre-anival

l7

symptoms.

The Reggignship Begun Stress and Health

The nature of the relationship between mind and body has been of interest to
philosophers since antiquity (Contrada, Leventhal, & O’Leary, 1991). Accounts of
charismatic individuals (e.g. Indian fakirs, faith healers, shamans, voodoo
practitioners) capable of influencing physiological functioning through psychological
manipulations are widely available. Walter Cannon (cited in Dohemewend &
Dohrenwend, 1984, p. 2) was among the first scientists to undertake a systematic
investigation of the effects of psychological stress upon physiological functioning. He
suggested that vital bodily organs would be irreparably damaged if the autonomic
nervous system was maintained in a highly aroused state through prolonged
psychological stress. Hans Selye’s (193 6) introduction of the general adaptation
syndrome allowed for more clarity in Cannon’s assertions by suggesting that prolonged
exposure to stress leads, inevitably, to the exhaustion of physical and psychological
defensive resources. This exhaustion leaves the organism vulnerable to a variety of
physical and psychological disorders, and in severe cases it can be fatal. It is firrther
suggested that although the experiencing of sudden and severe stressors that result in
fatality are nowadays relatively rare, certain characteristics of modern life do lead to a
sustained state of autonomic arousal characteristic of a mild stress reaction. This type
of a mild stress reaction has been associated with poor academic performance (Lloyd,
Alexander, Rice, & Greenfield, 1980); insomnia (Hartmann, 1985); nightmares

(Cernovslry, 1989); sexual difficulties (Malatesta & Adams, 1984); drug abuse

18

(krueger, 1981); and anxiety and depression (Weiten, 1988). Holmes and Rahe (1967)
were among the first investigators to cite changes in life circumstances as sources of
this kind of mild stress reaction.

More recently, researchers have been successfirl in inducing an array of illnesses
in laboratory animals through exposing them to severe stress. Sklar and Anisman
(1979) for example, found that mice that were administered electric shock, afler having
had cancerous cells surgically planted into them, were more likely to grow tumors
more rapidly and die as compared to those animals with cancerous cells that were not
administered electric shock.

Studies with humans have, for obvious reasons, been of a quasi-experimental
nature. Although these designs lack the degree of internal validity necessary for strong
causal statements, they do, nevertheless, allow the investigator to make causal claims
of a somewhat weaker kind. Kasl and Cobb (1970) for example, studied a group of
workers beginning two months before their jobs were to be terminated and for two
years subsequent to loss of employment. Their design included a control group
consisting of men in similar occupations who did not lose their jobs. For the group of
workers who lost their jobs, the results indicated elevated blood pressure both with
anticipation of job loss and afler termination of employment. No such rise in blood
pressure was found in the control group. The inclusion of a control group adds to the
internal validity of this study. Interestingly, however, the authors do not account for
those workers who lost their jobs but did not experience a rise in blood pressure.

Jemmott and Magloire (1988), examined the efi‘ects of academic stress upon

19

salivary concentrations of Immunoglobulin A (IgA) among 15 healthy undergraduates
5 days before, during, and 14 days after their final exams (IgA is an immune system
protein responsible for the protection of mucus membranes fiom pathogens). Their
results supported the hypothesis that academic stress can temporarily suppress immune
system activity through reducing the concentrations of IgA in the saliva. A strength of
this study was its collection of data regarding the levels of subjects’ social support.
These data were used to account for within sample differences in the concentrations of
IgA. Other studies have established a link between chronic stress and onset of such
diseases as the common cold (Totman, Kiff, Reed, & Craig, 1980), vaginal infections
(Williams & Deffenbacher, 1983), cardiovascular disease (Rosengren, Tibblin, &
Wilhelrnsen 1991); infectious disease (Stone, Reed, & Neale, 1987); and pregnancy
complications (Pagel, Smilkstein, Regen, & Montano, 1990).

Studies have also established an association between chronic exposure to stress
and psychological symptoms. In a widely cited study, Seligrnan (1974) initially
exposed a group of dogs to inescapable electric shock. He subsequently found that
these dogs were less likely to learn to escape the electric shock in situations from
which escape was possible. He cited this kind of “learned helplessness” as a possible
antecedent of depression. More recently, Bodnar and Kielcot-Glaser (1994), using a
longitudinal design, discovered significantly higher prevalence of depression among
caregivers of persons with Alzheimer's disease as compared to a control group.

Although the existence of a stress-illness link has been consistent and

significant, it has rarely been found to be greater than .30 (Holohan & Moos, 1985).

20

The usual range of correlations has been reported as between 0.20 and 0.40 with a
standard deviation that is about eight times the mean (Rabkin & Struening, 1976a,
1976b). This finding has provided the impetus for researchers to explain these within
group variations through the investigations of personality factors.

The Role of Harding; in the Stress-Illness Link

The role of hardiness as a mitigating factor in the stress-illness link was first
studied by Suzanne Kobasa in the course of her doctoral dissertation research (Kobasa,
1979). In this seminal work, Kobasa predicted that individuals high in commitment,
challenge, and perceived control would be less vulnerable to the deleterious effects of
life change. Her results significantly supported this prediction. An immediate criticism
of Kobasa’s findings was that data from her first study did not include information
regarding individual levels of hardiness prior to exposure to stress. This was deemed
important because in order to demonstrate that hardiness acted as a buffer against the
adverse effects of stress, it was essential to show that symptomatic individuals were
low in hardiness prior to exposure to stress.

Kobasa acknowledged the need for a prospective study to determine the
directionality of her findings. She and her colleagues published this prospective study
in 1982 (Kobasa, Maddi, & Kahn, 1982). Their results confirmed that hardiness acted
prospectively as a buffer against the harmfirl effects of stress. (Kobasa et al., 1982, p.
174)

Since Kobasa's introduction of hardiness, other researchers have studied its role

in the stress-illness link of other populations. Rhodewalt and Zone (1989) for example,

21

conducted a survey study of adult women to examine whether psychological hardiness
acted as a buffer against stressful life change. Their results indicated a significant
difference between hardy and nonhardy subjects' appraisal of their life events as either
negative or positive. Hardy and nonhardy women also differed in their reports of the
average amount of adjustment required for each event. Although Rhodewalt and Zone
discovered an interaction between hardiness and appraisal of undesirable life events in
the prediction of depression and illness, this interaction was found to be much stronger
for nonhardy women than it was for hardy women. On the basis of this finding
Rhodewalt and Zone suggested that appraisal processes are perhaps a major
determinant of most hardiness effects reported in the literature. Accordingly, other
researchers have examined the role of variables such as neuroticism (Bensihek, 1993)
or maladjustment (Bernard and Belinsky, 1993) as possible confounds in this line of
research.

In their review of the literature, Bernard and Belinsky (1993 ), reported that the
dimensionality of the hardiness construct has been questioned along with the methods
used in its measurement. Specifically, with regards to the dimensionality criticism, they
found that contrary to Kobasa’s claims, the three components of hardiness appeared to
possess differential effects upon the stress-illness link. Furthermore, with regards to
the measurement criticism, Bernard and Belinsky found a general consensus in the
literature to the effect that measuring the converse of a variable, as most hardiness
instruments did, was not a satisfactory way of measuring that variable (It was claimed,

for example, that the presence of alienation was perhaps not quite the same thing as the

22

absence of commitment and so on).

Criticism has also been directed at the statistical procedures employed in the
analyses of hardiness data (Benishek, 1993). It has been reported, for example, that in
many hardiness studies, analyses of variance (ANOVA) have been carried out using
continuous data. Finally, there has been a criticism of hardiness samples with respect
to their generalizability characteristics (Funk, 1992).

Although it would be a gross distortion to suggest that the role of hardiness as
a factor in the stress-illness link has been rejected, it would be equally inaccurate to
state that its role has been fully understood. Further research in this area is needed.
The Role of Soeial Suemrt in the Stress-Illneee link

The study of the role of social support as a means of ofi'setting stress induced
psychological and psychosomatic disorders dates back to at least the seventies. As
Heller and Monahan (1977; reported in Leavy, 1983) noted "in general, the evidence is
sparse concerning the ability of supportive social structure to moderate the impact of
stressful life events ..... how support operates, or how its beneficial effects can be
optimized is a matter for firture research" (p. 3). A large number of studies in this area
have been done in the twenty years since Heller and Monahan’s observation.

One of the early problems of this line of research was the conceptualization of
social support. Various definitions have been offered. House (1981) for example,
defined social support in terms of emotional support, consisting of the exchange of
affection, admiration, and affirmation; appraisal support, consisting of listening to

people and helping them in the assessment of their problems and the evaluation of

23

options; informational support, consisting of guidance and direction in the solution of
practical matters; and instrumental support, consisting of, for example, lending money
or providing transportation to one in need of such services. Other researchers have
conceptualized social support in terms of an individual‘s perceptions regarding its
availability, utility, and meaning (e. g., Yang & Clum, 1995).

Gore (1978) used a longitudinal design to study the efi‘ects of social support in
moderating the health consequences of unemployment. Her sample consisted of 54
rural and 46 urban married blue collar men with an average age of 49. Gore also used
a matched control group consisting of 74 men who were employed in similar
occupations that did not lose their jobs. In this study, social support was
operationalized in terms of a) subject’s perception of his wife, fiiends, and relatives as
supportive or unsupportive; b) fi'equency of activity outside the home with the same
individuals; and c) subject's perceived opportunity for engaging in social activities
which are satisfying and which allow him to talk about his problems. Her results
indicated a significant difference between the depression, self-blame, and illness
symptoms of the tenninees according to their levels of measured social support. Data
obtained fiom the control group, however, was highly suspect and therefore not
analyzed, since the subjects in this group reported a significant amount of health related
concerns. As Gore notes, these subjects were apparently motivated by the prospects of
being visited by a registered nurse on a regular basis.

Cohen and Hoberman (1983) studied the combined and separate roles of

positive life events and social support in the development of psychological and

24

physical symptoms of stress due to life change. Their subjects consisted of a group of
college students enrolled at a large university. The instruments were administered
during several class periods and student attendance influenced the number of subjects
completing each instrument. In this study, social support was defined as perceived
availability of support and as perceived support received during the past month.
Results indicated the following: a) perceived availability of social support provided
some protection fiom the adverse effects of high levels of stress; and b) increases in the
perceived availability of support were associated with decreases in depressive
symptoms but not with decreases in physical symptoms.

Yang and Clum (1994) studied the combined and separate roles of life stress,
social support and problem solving skills as predictors of depression, hopelessness, and
suicidal ideations among a group of international students from Asia enrolled in a US.
university. In this study, social support was operationalized in terms of scores on the
UCLA Loneliness Scale (Russell, Peplau, & Ferguson, 1978 as reported in Yang &
Clum, 1994). This instrument measures perceived social support and degree of social
integration. Yang and Clum hypothesized that levels of life stress, problem-solving
deficits, and social support would both independently and interactively predict
depressive symptoms, hopelessness, and suicide ideation. Results of this study
affirmed both the separate and combined utility of social support, problem-solving
deficits, and life stress in the prediction of depression, hopelessness, and suicidal
ideation.

In a review of forty-six studies investigating the role of social support in the

25

stress-illness link, Leavy (1983) concluded that ”regardless of research methods, one
finding is consistently reported: the absence of social support is associated with
increased psychological distress" (p. 15). Leavy, however, found that his literature
review did not allow him to make a similarly conclusive assertion regarding the
bufi’ering effects of social support against the adverse effects of life stress. Kobasa and
Puccetti (1983), for example, found that for male business executives perceptions of
support from their boss did produce a buffering effect, whereas perceptions of support
from their family did not produce such an effect. Indeed Kobasa and Puccetti found
that family-rooted support was associated with greater number of symptoms in subjects
low in hardiness.

Although the literature on the role of social support is equivocal, no researcher
has argued for the wholesale dismissal of its role. Instead there seems to be a need for
refinements in both conceptual and methodological approaches. Leavy identified some
possible reasons for the inconsistent findings including the use of inappropriate or
insensitive instruments, inappropriate statistical procedures, problems with how social
support is operationalized, and variations in sample characteristics.

Th om ined Roles f Hardiness and Social Su in he Stress-nines Link

As early as 1983, Kobasa and Mark Puccetti were investigating the combined
roles of hardiness and social support in the stress-illness link. In this relation, they also
studied the roles of various types of social support (e. g. support received at home vs.
support received at work). They hypothesized that a) high levels of hardiness along

with high levels of access to social resources will predict better health status in male

26

executives experiencing high levels of stress; and b) higher levels of hardiness along
with low levels of social resources will be correlated with better health status as
compared with lower levels of hardiness and higher levels of social resources.

The subjects in this study consisted of 240 middle and upper level business
executives. The results were mixed. First, although hardiness combined with social
support received at work served as a buffer against the adverse effects of stress, the
interaction of low levels of hardiness and social support received at home led to
adverse health consequences. Second, social support received from the family
functioned as a resistance resource only among hardy participants. Subjects who were
high in hardiness but low in family social support showed significantly lower illness
scores than executives who were low in hardiness but high in family social support.
This study is subject to some of the same criticisms as were leveled against Kobasa's
earlier work.

Kobasa (1979) had suggested that the three components of hardiness
(commitment, challenge, and control) are so highly interrelated that a composite score
can be used to represent all three components. This suggestion has been contradicted
in several studies. Ganellen and Blaney (1984), for example found that the three
components of hardiness were differentially related to social support. Specifically, they
found strong correlations between the commitment and challenge dimensions of
hardiness and social support, but they failed to find such a correlation for the control
component. With respect to the roles of hardiness and social support, this study

revealed that although life stress, social support, and hardiness were all directly related

27

to depression, social support was not found to buffer the effects of life stress.
Mummies

This study is designed to address the following questions: a) In terms of
physical and psychological health, how do international students adjust to their new
environment; b) does hardiness play a role in the adjustment of new international
students; c) do perceptions of social support play a role in the adjustment of new
international students; d) does the interaction of perceptions of social support and
stress play a role in the adjustment of new international students; e) does the
interaction of hardiness and stress play a role in the adjustment of new international
students; 1) is there a relationship between immediate post-arrival health and the 12-
week post-arrival health of international students; g) are there any demographic
characteristics that help predict the adjustment of new international students; and h) are
there any differential effects due to the three components of hardiness.

Me

To address the above questions, the present study is designed to test the
following hypotheses:

A The hardiness level of new international students will significantly
moderate the relationship between stress and adjustment. Specifically, among new
international students with higher levels of hardiness, stress will be unrelated to
adjustment; whereas among new international students with lower levels of hardiness,

stress will be significantly related in a negative direction with adjustment.

28

B. New international students’ perceptions of social support will significantly
moderate the relationship between stress and adjustment. Specifically, among new
international students with more positive perceptions of social support, stress will be
unrelated to adjustment, whereas among new international students with less positive
perceptions of social support, stress will be significantly related in a negative direction
with adjustment.

C. The adjustment of new international students will be significantly
negatively correlated with the level of discrepancy between U. 8. culture and the new
international students’ native culture. Specifically, the mean of the adjustment scores
of the new international students for whom English is the native language will be
significantly lower than the mean of the adjustment scores of the new international
students for whom English is not the native language.

D. The adjustment of new international students will be significantly
negatively correlated with the level of discrepancy between US. culture and the new
international students’ native culture. Specifically, the mean of the adjustment scores
of the new European (or from European extraction) international students will be
significantly lower than the mean of the adjustment scores of the new non-European

(or from non-European extraction) international students.

29

CHAPTER II I
METHODOLOGY
Participants

The sample for this study consisted of a group of new international students.
For the purposes of this study “new international student” was defined as: a) born
outside of the US; b) Fall 1997 enrollment at Michigan State University (MSU)
represented the first and only enrollment at a US. college or university; c) arrived in
the US. on or after August 1, 1997; (1) other than the present stay, has not resided
continuously in the US. for more than a month in the last 10 years; and e) has never
resided in the US. for more than 2 months.

A power analysis (Cohen, 1977), using an alpha level of.05, R2 of .2, power of
.8, and with 7 predictor variables ( base-line depression, base-line anxiety, base-line
psychosomatic symptoms, base-line somatization, hardiness, social support, and
stress), yielded a required sample size of 102. In order to compensate for the effects of
attrition in our sample, and in order to enhance the randomness of the sample all “new”
international students were contacted and solicited to participate.

The pool of potential participants for this study was identified through
cooperation with the university’s office of the Registrar. This office generated a list of
412 names and addresses of first-time-enrolled international students. In order to
preserve the privacy of the new international students, stamped envelopes containing a
cover letter (see appendix A), a sign-up sheet (see appendix C), and the initial survey

package along with a stamped return envelope were submitted to the Registrar’s office,

30

The cover letter described the purpose of the study and extended an invitation to those
interested in participation, to complete and return the sign-up sheet. The registrar’s
office affixed the address labels and mailed the envelopes.

Of the 412 mailed packages, 9 were returned by the post office as
undeliverable. Of the remaining 403 presumably delivered packages, 218 were
returned (54% response rate). Of this number 92 were deemed unusable for the
following reasons: 71 were excluded because the respondent did not meet all
participation requirements; 14 were excluded because the respondent had not
completed the survey package; and 7 were excluded because the respondent had not
completed the sign-up sheet and thus could not be contacted for the final data
collection session. Of the remaining 126 respondents, who had satisfied all the
inclusion requirements of the study, 102 (81% response rate) completed the final
survey package and thus formed the final sample of this study.
medium:

The initial survey package consisted of the following: an informed consent
form, the first demographics questionnaire (see appendix D), and two instruments
measuring the physical and psychological health of the participants (see appendices E
and F). The measurement of these variables in the first two weeks after arrival allows
for the assessment of the effects of the predictor variables in this study while
controlling for baseline levels of symptoms. The design of this study required that the
initial survey package be completed on or before 30 September, 1997.

The second survey package consisted of a second demographics questionnaire

31

(see appendix J), the two instruments measuring the physical and psychological health
of the participants, the hardiness instrument (see appendix 1), the measure of stress (see
appendix H), and the measure of social support (see appendix G). In order to minimize
fatigue effects, four versions of this package were produced. The packages differed
only in the order in which the instruments appeared. These packages were
administered in person on three evenings in mid-November, 1997.

Lamar:

The present study included one measure of stress (The Index of Life Stress);
one measure of hardiness (the Personal Views Survey-II); one measure of social
support (The Index of Social Support); and two measures of adjustment (the SUNYA-
Psychosomatic Symptoms Checklist and the Symptoms Check List-90-R); a
demographic questionnaire designed to elicit data on such variables as might covary
with the outcome measures, or alternatively exert a confounding effect upon the results
of the study. Examples include nationality, race/ethnicity, age, gender, housing
situation, major, relatives within 200 miles, relatives living in the U. S., native
language, and TOEFL scores. An important variable obtained from the demographic
questionnaire is “relatedness.” This variable is a composite of items 9 and 10 of the
questionnaire (see appendix D).

geese. The Index of Life Stress (ILS; Yang & Clum, 1995) consists of 31
statements which measures six areas of stress relevant to the international student
population: language difficulty ,cultural adjustment, perceived ethnic/racial

discrimination, academic concem, financial concern, and outlook for the future.

32

Subjects rate each statement fi'om never (0) to often (3) according to how often the
individual "feels the way described in each statement.” Item ratings are summed to
produce a total stress score, with higher scores reflecting higher levels of stress.
Sample items include: “My financial situation makes my life very hard,” “My English
embarrasses me when I talk to people,” and “People treat me badly just because I am a
foreigner.”

The test-retest reliability of this instrument over a 1-month interval was .87
(n=20); the internal consistency estimate was .86 (n=101); and the concurrent validity,
measured by the correlation between the ILS and the Life Events Survey Scale
(Sarason, Johnson, & Siegel, 1978 as reported in Yang & Clum, 1995), was r(100)=-
.46, p<.0001.

Hardiness. The Personal Views Survey 11 (PVS II; Hardiness Institute, 1995)
consists of 50 statements which assess the commitment, control, and challenge
components of hardiness. Each statement is answered using a 4-point Likert scale
(0 = Not at all true; 3 = Completely true). Higher scores indicate greater degree of
each component. In contrast to the Revised Hardiness Scale, the PVS contains both
positive and negative indicators of hardiness. Composite scores are calculated by
combining the three component scores.

The internal consistency for the composite score range from 0.87 (Wiebe,
Williams, & Smith, 1991) to 0.90 (Hardiness Institute, 1985). The internal consistency
reliability for commitment, control, and challenge are .72, .62, and .70, respectively

(Wiebe et al., 1991). Similar values are reported by the Hardiness Institute (1985).

33

Test-retest reliabilities of the PVS over time periods of two weeks or more have been
reported to be in the .60's (Hardiness Institute, 1984).

Social Suppgrt. The Index of Social Support (ISS; Yang & Clum, 1995) is a
40-statement scale based on the international students' special social contact patterns.
It assesses the personal meaning of social support, the level of trust and satisfaction
toward the support, and the availability of support when it is needed in relation to
immediate family, extended family, old fiiends in the home country, new fiiends in the
US, churches, school organizations, the international student center on campus, and
community activities. Subjects rate each statement from never (0) to often (3)
according to how often the individual "feels the way described in each statement"
(Yang and Clum, 1995). Item ratings are summed to produce a total score and higher
scores correspond to higher levels of social support. Sample items include: “I have
contact with my old fiiends in my home country,” “I trust the international student
center on campus,” “The international student center on campus means a lot to me,” “I
participate in community activities here,” “My new fiiends in the USA. are available
when I need them,” “I trust my church.”

The test-retest reliability of this instrument over a l-month interval was .81
(n=20); the internal consistency estimate was .81 (n=100); and the concurrent validity,
measured by the correlation between the ISS and the UCLA Loneliness Scale (Russell,
Peplau, and Ferguson, 1978 as reported in Yang and Clum, 1995) was r(100)= -.39,

p<.0001 (Yang and Clum, 1995).

34

Adjustment. The Symptom Check List-90-Revised (SCL-90-R; Derogatis,
1990), is a widely used 90-item Likert-type self-report symptom inventory designed to
reflect the psychological symptom patterns of community, medical, and psychiatric
respondents. Each item is rated on a five-point scale of distress (0-4) ranging from
“Not at All” to “extremely.” The SCL-90-R is scored and interpreted in terms of the
following nine primary symptom dimensions: Somatization, Obsessive-Compulsive,
Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid
Ideation, and Psychoticism.

For the purposes of this study, only the three symptom dimensions of
Depression Anxiety, and Somatization will be used. Sample items for the Depression
scale include: “Loss of sexual pleasure or Interest,” and “Thoughts of ending your
life.” Sample items for the Anxiety scale include: “Nervousness or shakiness inside,”
and “Feeling tense or keyed up.” And sample items for the Somatization scale include:
“Headaches,” and “A lump in your throat.”

The internal consistencies for the Depression, Anxiety, and Somatization scales
have been reported as .90, .87, .82 respectively (Derogatis, Rickels, & Rock, 1976).
The combined test-retest reliabilities of these scales have been reported as .78 and .80
respectively (Derogatis, Rickels, & Rock, 1976). The concurrent validity of the SCL-
90-R was established by calculating correlations between its nine primary symptom
dimensions and the similar scales on the MMPI. The correlation coefficients between
SCL-90-R’s Depression scale and MMPI’s Wiggins Depression, and Tryon Depression

scales has been reported as .75 and .68, respectively. The correlation coefficient

35

between SCL-90-R’s Anxiety scale and MMPI’s Tryon Anxiety scale has been

reported as .57. The correlation coefficients between SCL-90-R’s Somatization scale

and MMPI’s Hypochondriasis and Tryon’s Body Symptom scales has been reported as

.57, .66 (Derogatis, Rickels, & Rock, 1976).

The SUNYA-Psychosomatic Symptoms Checklist-Revised is a self-report scale
which contains 17 common psychosomatic complaints. Subjects are asked to rate. each
item on fi'equency and intensity of occurrence, using a five-point scale. The PSC yields
a composite score, derived by summing the cross-product of each item’s fiequency by
intensity. Sample items include: “Headaches”, “Backaches”, and “Sore throat.”

Test-retest reliability for the PSC has been reported as generally high across 1-
week, 4-week, and 8-week intervals. These were .88, .84, and .80, respectively
(Attanasio, Andrasik, Blanchard, & Arena, 1984).

D n i r ur '

1. Descriptive statistics (i.e., mean, stande deviation, skewness, and range) were
carried out on all the predictor and outcome measures and also on data
obtained through the demographics questionnaire (i.e., age, nationality, major,
gender, housing situation, relatives within 200 miles, relatives living in the US,
native language, and TOEFL scores).

2. Cronbach's alpha, a measure of internal consistency reliability, were computed
for the appropriate measures used in this study.

3. The strength of the relationships between various dependent and independent

variables were evaluated through the generation of a correlation matrix.

36

Additionally, this analysis yielded preliminary information regarding confounds
and covariates.

Chi-square analyses were used to assess mean differences with categorical data.
A set of full-model regression analyses were performed in order to obtain
estimates of the influence of each predictor variable (pre-test symptoms, stress,
hardiness, and social support) and their interactions upon the outcome variable
(Post-test symptoms). Other regression (F -test) analyses, as well as post-hoc
analyses were carried out in order to investigate the roles of such variables as
race, language, nationality, and relatedness.

The prediction of adjustment by the interaction between life stress and social
support and the interaction between life stress and hardiness were tested by

hierarchical regression analysis.

37

Fermet

CHAPTER IV

RESULTS

The results chapter is divided into the following sections:

1.

2.

Data Entry Issues

a. Sample Description.

b. Sample Representativeness.

a. Description of the Variables Used in the Analyses.

b. Internal Consistencies of the Instruments.

c. Comparison of Scores Obtained in the Present Study with Those
Reported in Other Studies.

a. Frequency and Intensity of Symptoms, Syndromes.
l. Psychological Symptoms and Syndromes
2. Psychosomatic Symptoms

a. Descriptive Analysis of the ILS

b. Descriptive Analysis of the ISS.

a. Inferential Statistics and Hypothesis Testing
1. Hypothesis 1
2. Hypothesis 2
3. Hypothesis 3
4. Hypothesis 4

b. The Individual Effects of the Three Components of Hardiness.
1. Challenge
2. Commitment
3. Control

a. Role of Demographic Factors

38

W

The data was first examined for missing values, outliers, and accuracy of data
entry. Eight data entry errors were detected and rectified. The number of missing
values for a scale item ranged fi'om 1 to 6. With regards to the demographic
information, missing values were found for Relatives in the United States (n=2),
Relatives in Lansing/East Lansing (n=2), Distance of Relatives from Lansing/East
Lansing (n=3), Closeness to Relatives (n=2), Friends/Acquaintances in Lansing/East
Lansing (n=2), Mother’s Education (n=2), Father’s Education (n=4), Parents Educated
in the United States (n=1), and Significant Incidents While in the United States (n=8).
Semele Deseription

Table 1 contains descriptive demographic characteristics of the sample which
consisted of 48% males (n=49), and 52% females (n=53). They ranged in age from 17
to 41 years with an average age of 24.43 years. The sample included students of 34
nationalities with students from China forming the largest group (n=21; 20.6%),
followed by students from Thailand (n=11; 10.8%). When participants were assigned
to geographical regions according to their country/culture of origin the Far East was
the largest group (n=54; 52%), followed by the Middle East (n=16; 15%) and Western
Europe (n=12; 12%). For the purposes of hypothesis D, the regions were then
collapsed into European (n=17; 17%) and Non-European (n=85; 83%). The resultant
variable was labeled “Zregion.” On the race/ethnicity dimension, the majority of the
participants identified themselves as Asian (n=68; 67%), followed by White (n=26;

26%), and Black (n=3; 3%). The majority of the participants described their marital

39

status as single (n=71; 70%), followed by married or engaged but currently not

residing with their spouse (n=20; 20%).

4O

Table 1. Sample Demographic Information. (Numbers in parentheses reflect the appropriate MSU

population data)
[ Variable

Sex
Males
Females

Age

17 years or less
18-20 years

21 -23 years
24-26 years
27-29 years
30-32 years
33-35 years
36-38 years
39-41 years

42 years or more

Nationality
China
Thailand
Malaysia
South Korea
Taiwan
Pakistan
India
Germany
Turkey
Philippines
Indonesia
Belgium
France
Lebanon
Sri-Lanka
Russia
Australia
Romania
Japan
United Kingdom
Yugoslavia
Chile
Kenya
Brazil
Netherlands
Peru
Hong Kong
Nigeria
Sweden
Haiti
Colombia
Egypt

41

111

49(1607)
53(1057)

15
31
30

O-‘O-hm

21(269)
1 1(74)
7(83)
6(550)
6(293)
5
4(176)
3

3

3
3(54)

(197)

(54)

u—I—du—I—d—dl—d—nl—Iu—I—I—II—Ir—l—NNNNNNN

I'/-l

48(60)
52(40)

14.7
30.4
29.4
14.7
4.9
3.9

2100)
11(2.8)
7(3.1)
6(21)
6(11)
5

4(6.6)
3

3

3
3(2)

(7.4)

u—nu—o—n—u—n—u—a—ac—n—Ir—n—1—I—INNNNNNN

Table l (cont’d).

Region

Jordan
Tanzania

Far East

(Japan, S.Korea, Taiwan, Thailand, China,
Philippines, Hong Kong, Malaysia)
Middle East

(Pakistan, India, Sri-Lanka, Indonesia)
Near East

(Lebanon, Turkey, Egypt, Jordan)

Africa

(Kenya, Nigeria, Tanzania)

Eastern Europe

(Yugoslavia, Russia, Romania)

Western Europe

(UK, Belgium, Germany, France, Netherlands,
Australia, Sweden)

South America

(Chile, Brazil, Peru, Haiti, Colombia)

Zregion (European Vs. Non-European)

European
Non-European

Language

Arabic
Chinese
Dutch
English
Filipino
French
German
Hindi
Indonesian
Japanese
Korean
Portugese
Romanian
Russian
Serbian
Sinhalese
Spanish
Swahili
Swedish
Thai
Turkish
Urdu

42

54

16

12

17

A

mw———w——NN—O\~umwhwmwuh

53

15

12

17
83

3.9
33.3
2.9
4.9
2.9
3.9
2.9
4.9
2.9

5.9
1.0

2.0
1.0
1.0
2.9
1.0
1.0
10.8
2.9
4.9

Table 1 (cont’d).

Race
Asian 68
Black 3
White 26
Hispanic
Pacific Islander
Mixed Race
Other

Cunt—ow

Marital Status
Married/engaged spouse with me
Married/engaged spouse not with me
Single
Living with Boyfriend/Girlfriend
Separated
Divorced
Widowed

--O

ONONQNQ

Table 2 contains information regarding the social network characteristics of the
participants. It is evident that the sample consisted mostly of individuals with no
established social network upon anival. Of the 102 participants only 33% (n=34) had
relatives in the United States, of whom only 21% (n=7) resided in the Lansing/East
Lansing area. Of these participants, 71% (n=24) indicated that they felt close or very
close to these relatives. Overall, just under half (n=47; 46%) had fiiends or
acquaintance in the Lansing/East Lansing area upon arrival.

The majority (47%) of the participants resided on campus in residence halls

and 75% (n=76) described their new accommodation as different or very different from

their accommodations in their home countries.

43

66.7
2.9
25.5
2.9

6.9
19.6
69.6

ONON

Table 2. Sample Social Network Information

 

 

[ Variable # | 7. ]
Relatives in the United States
Yes 34 33
No 66 65
Missing 2 2
Do you have relatives in the Lansing/East Lansing area
Yes 7 21
No 27 79
How close do you feel to your relatives in the US
Very Close 16 47
Close 8 24
Not Close 10 29
Did you have friends or acquaintances in Lansing/East Lansing area
No 53 52
Yes 47 46
Missing 2 2
Accommodation Type
Live on campus in a residence hall 48 47
Live on campus in the university apartments 39 38
Live in a fraternity or sorority 0 0
Live off-campus in an apt or house 10 10
Live off-campus with parents or relatives 5 5
Live in a cooperative house 0 0
Accommodation Similarity
Very similar 1 1
similar 25 25
Different 31 30
Very Different 45 44

Table 3 contains the education related variables of the sample which consisted

mostly (n= 75; 74%) of graduate students. The largest group (n=28; 28%) was

pursuing degrees in the College of Business followed by the College of Engineering

(n=19; 19%). Although only 5 members (n=5; 4.9%) of the sample identified

themselves as native English speakers, the sample as a whole tended to be relatively

proficient in their English language skills as illustrated by a mean score of 605 in the

TOEFL (mode: 630; median: 610). The usual required score for the purposes of

44

admission to a US. college or university is 550. The participants also tended to be the
children of well-educated parents with roughly 46% of their mothers and 61% of their
fathers holding at least a bachelor’s degree. Only 9 (8%) of their parents were

educated in the United States.

45

Table 3. Sample Education Related Information

 

 

rVarlable I # I v. |
Educational Level
Graduate 75 73.5
Undergraduate 27 26.5
TOEFL scores
490-532 4 3.9
533-575 20 19.6
576-618 24 23.5
619-661 27 26.5
662-704 8 7.8
Mean: 605.52; Mode: 639; Median: 610
SD: 41.88
Major
College of Business 28(448) 28(17)
College of Engineering 19(430) 19(16)
College of Natural sciences 18(408) 18(15)
College of Agriculture 8(27 2) 8(10)
College of Communication Arts & Science 7(242) 7(9)
College of Education 6(136) 6(5)
College of Arts & Letters 5(280) 5(1 1)
ELC 4 4
College of Social Sciences 2(170) 2(6)
College of Human Ecology 2(65) 2(2)
Undecided 1 1
Missing 2 2
Mother’s Educational Level
Less than HS 16 15.7
HS 30 29.4
Some College 6 5.9
Bachelor’s Degree 40 39.2
Graduate, Professional Degree 7 6.9
Deceased l 10
Missing 2 2.0
Father’s Educational Level
Less than HS 6 5.9
HS 21 20.6
Some College 5 4.9
Bachelor’s Degree 45 44.1
Graduate, Professional Degree 18 17.6
Deceased 3 2.9
Missing 4 3.9
Parents Educated in the US
Yes 9 9
No 92 91

46

Sample Represeneetivengs

The representativeness of the sample used in this study was evaluated by
comparing the demographic characteristics of the sample with that of the international
student population at MSU. Values in parentheses in tables 1, 2, and 3 indicate the
demographic characteristics of the international students enrolled at MSU. These data
were collected in a descriptive study carried out by the MSU Oflice of International
Students and Scholars (MSU-OISS) between September 20, 1996 and October 30,
1996, approximately one year prior to the collection of data for the present study.
With regards to the sex of the participants, the international students enrolled at MSU
consist mostly (60%) of males, while the participants in this study were more equally
divided with a slightly higher number of females. Students fi'om South Korea form the
largest single group of international students enrolled at MSU while students from
China formed the largest group in the present study. The top five nationalities
represented in this study, however, include five of the top eight nationalities
represented at MSU as a whole. Finally, the distribution of majors in our sample

closely resembles that of MSU as a whole.

47

Varial s in heP n An

Table 4 contains the full name, abbreviated name, mean, standard deviation,
skewness, and range for each of the variables used in the present analyses. The
distribution of all variables was fairly normal. With the exception of hardiness and its
three components (i.e. challenge, commitment, and control), the distribution of all
other variables was characterized by some positive skewness. The negative skewness
characterizing the distribution of hardiness and its components is indicative of a
generally hardy sample. Furthermore, on certain variables, such as depression, anxiety,
or psychosomatic symptoms, the positive skewness of self-reported symptoms is
typical of most non-clinical samples and indicates that the majority of the sample
reported fewer, and less severe psychological and/or physical symptoms.

Tables 5.1 and 5.2 are correlation matrices composed of the Pearson
correlation coefficients of the strength of the relationships between the various
variables used in the present analyses. Although there are a fairly large number of
significant correlations between the various variables used in this study, no correlation
coeflicient is of a magnitude large enough to raise multicollinearity concerns. Indeed,
as Tabachnik and F idell (1989) have noted multicollinearity becomes a problem when

bivariate correlations are equal to or greater than .90.

48

Table 4. Descriptive Statistics for All Variables.

 

 

[ Variable Name benviation [ M J so ] Sir. | Range
Depression at Time A ADEPR 14.81 1 1.51 1.47 0-58
Anxiety at Time A AANX 4.70 5.24 1.68 0-25
Psychosomatic
Symptoms at Time A PSCPRE 16.36 17.23 2.22 0-98
Somatization at Time A ASOM 4.77 4.88 1.88 0-26
Depression at Time B BDEPR 14.59 12.00 1.28 0-54
Anxiety at TimeB BANX 4.17 5.12 2.07 0-25
Psychosomatic
Symptoms at Time B PSCPST 16.57 15.77 1.54 0-73
Somatization at Time B BSOM 4.68 4.75 1.51 0-22
Hardiness Score HARDINES 89.44 12.76 -().46 55-111
Index of Life Stress STRESS 32.97 13.23 0.26 1-72
Index of Social Support SUPPORT 74.70 16.01 0.36 31-1 18
Components of Hardiness
Commitment ACOMM 31 .77 5.70 -.35 18-41
Challenge ACHAL 25 .79 4.92 -.74 10-36
Control ACONT 31 .57 5.36 -.84 9-4

49

 

~ teen... 3. 5.- wmr act son.. 2.- 8. wot 2... 9895”:
~ :97 :34. samm. 1.3.. aamm. 3:. seem. seen. :94. $0.5de

fl sawmr sstr ssmmr 8.6m... that 3.3.- sst... asbmr maid

a Each. same. :3. :3. same. :8. :8. 20de

g sewn. same. :34. the. same. sewn. HmmUmmH

~ seem. sewn. :34. .38. earn. gmé

_ :34. sewn. sin. same. mmmnmd

fl :8. save. 21.0. 20%;.

m :8. seem. gamma

~ 1.2.. XZ<<N

~ yawn—<4

j“; a: 1 a_ L e_ 1 L 1 1 1 2.3.3:

 

sense... as 5 ea: BEES mass. 8588 3 2.3

50

Table 5.2. Correlations Among the Three Components of Hardiness
[Varlable I 1 H [ 3 J
1. Commitment 1

2. Challenge .43" 1
3. Control .53" .33" l

 

“ Correlation is significant at the 0.01 level (2-tailed).
‘ Correlation is significant at the 0.05 level (2-tailed).

Internal Qeneigtency of the Instruments Used

A set of analyses designed to establish the reliabilities and internal consistencies of the
instruments used in this study was carried out. Table 6 contains the results of these
analyses. The obtained Cronbach Alphas are, by and large, comparable to those
obtained in other studies. In case of the SUNYA-PSC, although no internal
consistency statistics have been reported, the 1-week, 4-week, and 8-week test-retest
reliabilities have been reported to be .88, .84, and .80 respectively. Although this does
not appear to compare particularly well with the 12-week test-retest reliability of .67

obtained in this study, it is measured over a longer period of time than those reported

 

 

in other studies.

Table 6. The Reliabiliy Coefficients (Cronbach Alpha) of the Instruments.

Llnstmment j # of Cases I # of Items | or ] Reported or
SUNYA-PSC (Frequency of Sx-initial) 102 28 .85 See text
SUNYA-PSC (Intensity of Sx-initial) 102 28 .86 See text
SCL-90-R (initial) 102 42 .94 .91
SUNYA-PSC (Frequency of Sx-fmal) 102 28 .76 See text
SUNYA-PSC (Intensity of Sx-final) 102 28 .78 See text
SCL-90-R (final) 102 42 .95 .91

Index of Life Stress 102 31 .86 .86(n=101)
Index of Social support 102 40 .88 .81(n=100)
Personal Views Survey 102 50 .81 Range:.87-.90

51

Qempetge' n 91' me Stores Obteing fly the Peeticipents in the Pment Stegy
With Thege Rewrted in Other Studies

The reported values were based on scores obtained fiom a group of 101
international students (73 males; 28 females) ranging in age from 18 to 40 with a mean
of 23.49 and standard deviation of 4.48. These demographics are descriptive of a

sample that is similar in age and sex to the participants of the present study.

Table 7: [LS and ISS Scores: Present Study Vs. Other Studies.

 

 

 

 

 

 

Instrument [wet Study Remrted
M SD M SD
STRESS 32.97 13.23 36.63 13.49
SUPPORT 74.69 16.01 67.91 18.40

Table 8 contains a comparison of the SUNYA-PSC scores obtained by the
participants in the present study with those reported in other studies. The reported

values were based upon scores obtained from a group of 947 subjects.

Table 8: SUNYA-PSC Scores: Present Study Vs. Other Studies:

 

 

 

 

 

 

Instrument Present Study 82221191!
M SD M SD
SUNYA-PSC(initial) 16.36 17.23 23.7 24.5
SUNYA-PST(final) 16.57 15.77 23.7 24.5

Fmeeney eng Inteneity ef Sympteme, Sendmmg, end Critical Item;

Psychological Smptoms and Smdromes

Each participant’s scores on the Depression, Anxiety, and Somatization scales
of the SCL-90-R was converted into a T-score thus allowing for a comparison with the
normative sample of the instrument. The results of these conversions along with
descriptive statistics appear in Table 9. These scores are based on a mean of 50 and a
standard deviation of 10. Scores above two standard deviations of the mean are

generally considered in the clinical range. It is evident that as a group, the participants

52

did not exhibit clinically significant levels of depression, anxiety, or somatization either

upon arrival or 12 weeks post arrival.

Table 9: Descriptive Statistics and Paired Sample T-Tests for the Depression, Anxiety, and
Somatization scales of the SCL-90-R.

 

 

[ Variable I M [ so | r 1 Sig. ]
Depression (initial) 59.44 8.58 .853 .396
Depression (final) 58.81 9.81

Anxiety (initial) 55.20 10.97 1.654 .101
Anxiety (final) 53.70 1 1.14

Somatization (initial) 51.42 9.55 .558 .578
Somatization (final) 50.99 9.57

The analysis of the scores of individual participants, however, reveals that of
the 102 participant, 57 (55.8%) scored one standard deviation or higher above the
mean on the Depression scale of the SCL-90-R. Of these, 14 (13.7%) obtained
clinically significant scores (2 standard deviations above the mean) on the Depression
scale (refer to Table 10). The number of clinically depressed students (as defined by
scores on the Depression scale of the SCL-90-R) would increase to 19 (18.6%) if
those participants scoring 69 are also included. Although the 12-weeks post-arrival
scores on this scale show an increase in the number of participants scoring 59 or lower
(from 45 to 53), there is also an increase in the number of participants scoring in the

clinical range for depression (14 to 16).

53

Table 10: T-Scores on the Depression Scale of the SCL-90-R.

 

 

[Variable I Range 1 Frequency 1

Depression (initial) 30-39 2
40-49 11
50-59 32
60-69 43
70-79 12
80-89 2

Depression (final) 30-39 4
40-49 16
50-59 33
60-69 33
70-79 12
80-89 4

Of the 102 participants, 33 (32.4%) scored one standard deviation or higher
above the mean on the Anxiety scale of the SCL-90-R upon arrival. Of these, 11
(10.8%) obtained clinically significant scores (refer to Table 11). The 12-weeks post-
arrival scores indicate an overall increase in the number of participants scoring 59 or
lower (from 69 to 74), and a decrease in the number of participants scoring in the

clinical range for anxiety (fi'om 11 to 9).

Table l 1: T-Scores on the Anxiety Scale of the SCL-90-R

 

 

Warlable I Range 1 Frequency 1

Anxiety (initial) 30-39 2
40-49 27
50-59 40
60-69 22
70-79 9
80-89 2

Anxiety (final) 30-39 7
40-49 26
50-59 41
60-69 19
70-79 7
80-89 2

Finally, on the Somatization scale of the SCL-90-R, 19 (18.6%) participants

scored one standard deviation or higher above the mean upon arrival. Of these, 4

54

(3.9%) obtained clinically significant scores (refer to Table 12). The 12-weeks post-
arrival scores on this scale, revealed an overall decrease in the number of participants
scoring 59 or lower (from 83 to 79), a decrease in the number of participants scoring
in the clinical range for anxiety (from 4 to 2), and an increase in the number of

participants who scored one standard deviation above the mean (fiom 15 to 21).

Table 12: T-Scores on the Somatization Scale of the SCL-90-R

 

 

I Variable T Range A F requencyj

Somatization (initial) 30-39 13
40-49 36
50-59 34
60-69 15
70-79 3
80-89 1

Somatization (final) 30-39 12
40-49 37
50-59 30
60-69 21
70-79 2
80-89 0

A set of items from the SCL-90-R were selected for individualized evaluation.
The criteria for the selection of these items consisted of a) significance for college
counseling center attention (e. g. suicidal ideations); and b) relevance to an aspect of
the study (e. g. feeling lonely, feeling no interest in things, and feeling hopeless). The
results of these analyses are reflected in Table 13. Immediately upon arrival, 3 (2.94%)
of the participants endorsed having experienced suicidal ideations, while 5 (4.9%)

endorsed these thoughts 12 weeks post anival.

55

Table 13: Frequency of Critical Items.

 

 

rCritical Item I Response 1 Fr (time 1) J Fr (time 2) I
Thoughts of ending your life Moderately 2 3
Quite a bit 0 0
Extremely l 1
Thoughts of death and dying Moderately 1 4
Quite a bit 1 0
Extremely l 1
Feeling lonely Moderately 22 24
Quite a bit 14 10
Extremely 9 8
Feeling no interest in things Moderately 7 10
Quite a bit 4 7
Extremely l 2
Feeling hopeless about the future Moderately 9 7
Quite a bit 2 5
Extremely 2 1

 

Psychosomatic Symptoms

Descriptive analysis of the individual items comprising the SUNYA-PSC
yielded the information reflected in Table 14. This analysis consisted of determining
the number of participants who endorsed experiencing an ailment at least once a week
or more often (i.e. several times a week or daily), with an intensity at least in the
moderate range or worse (i.e. severe or extreme). The most common and bothersome
psychosomatic symptom among the participants upon arrival (T1) was fatigue followed
by headaches, backaches, and insomnia. Although at 12 weeks post arrival, fatigue,
headaches, and backaches remained, and with the exception of headaches which
remained constant, increased as the most common and bothersome psychosomatic
symptoms, there were some noteworthy changes in this clinical picture. Common
cold, which was not endorsed with the frequency and intensity described above upon

arrival, was endorsed by roughly 10% of the sample as a frequent and moderately (or

56

worse) bothersome ailment 12 weeks post arrival. Also, cold sores, a Viral disease

with frequent flare ups at times of high stress, was endorsed by roughly 6% of the

 

 

sample.
Table 14: Frequently and IntenselLExperienced Psychosomatic Symptoms.
IS’ymptom film” Endorsed (T l) I Symptom I Times Endorsed (T2) I
1. Fatigue 32 1. Fatigue 36
2. Headaches 17 2. Headaches l7
3. Backaches 12 3. Backaches 14
4. Insomnia ll 3. Weakness l4
4. Stomach pain 1 l 5. Insomnia 13
6. Weakness 9 6. Stomach pain 1 1
7. Dininess 7 7. Common cold 9
8. Migraine 6 8. Sore throat 7
9. Diarrhea/Constipation 5 9. Cold sores 6
9. Sore throat S 10. Migraine 5
9. Upset stomach 5 10. Upset stomach 5
9. Nausea 5 12. Diarrhea/Constipation 4
13. Common cold 4 12. Dizziness 4
l4. Infected eye 3 12. Nausea 4
l4. Irregular heart beat 3 15. Heartburn 3
14. Heartburn 3 15. Heart palpitations 3
l7. Tonsilitis 2 l7. Tonsilitis 2
l7. Sinus infection 2 l7. Infected eye 2
17. Heart palpitations 2 17. Irregular HB 2
20. Bronchitis 1 l7. Sinus infect 2
20. Cold sores l 21. Laryngitis l
20. High blood 1 21. Chest pain 1
21 . Bronchitis 1

meeriptive Analysis of the Index of Life Stress

The Index of Life Stress (ILS), in addition to yielding a composite score,
provides information on five stress domains specific to the international student’s
experience. These domains consist of a) concern about finances and desires to stay in
the United States; b) language difficulties; c) interpersonal stress; d) stress from
cultural adjustment and desire to return to one’s own country; and e) academic
pressure. Table 15 reflects the results of an item analysis of the ILS along each of

these five dimensions. This analysis consisted of first rank ordering the items in

57

descending order according to the frequency with which the item was endorsed as
“often true.” Then calculating an arithmetic mean for each dimension by dividing the
sum of the position ranks of the items comprising each dimension by the number of
items in each dimension. The results describe, in descending order, the stress factors
most salient to the participants in the present study. Academic pressure is identified as
the most salient source of stress for the participants followed by financial concerns and

thoughts about staying in the United States.

Table 15. Salient Sources of Life Stress.

 

 

[immersion I M I so I
1. academic pressure 5.60 4.67
2. concern about finances and desires to stay in the United States 13.00 6.93
3. stress from cultural adjustment and desire to return to one’s own country 15.75 8.84
4. language difficulties 15.80 4.97
5. interpersonal stress 23.33 5.86

The items comprising the ILS were also analyzed for frequency of being
endorsed as “often true.” The top ten most frequently endorsed items are listed in
Table 15.1. The prevalence of academic pressures as a stressfill theme is also evident

from this analysis.

Table 15.]: Frequency of Occurrence of Items Comprising the ILS.

 

 

I [m I Frequency I
1 It would be the biggest shame for me if I fail in school (e) 44
2 I don’t like American food (d) 35
3 I study very hard in order not to disappoint my family (e) 33
4. I worry about my academic performance (e) 31
5. I wony about my filture career in my home country (d) 28
6 I worry about my financial situation (a) 27
7 I’m not doing as good as I want to in school (e) 25
8. I worry about my future: will 1 return to my home country or stay in the USA? (a) 22
9. Iworry about whetherlwrll have my future careerintheU. SA (a) 18
10.1 can t express myself well 1n English (b) 15

The letters in the parentheses correspond to the five stress domains described above.

58

Dgerintg' e Angtyete' ef the Index 91' fieig Support

The Index of Social Support (ISS), in addition to yielding a composite score,
provides information on four general sources of social support specific to the
international student’s experience. These sources consist of a) contact with old friends
in home countries, secondary family, and the international student organizations on
campus (contact with one’s own culture in general); b) contact with community
activities and student organizations; c) contact with new fiiends in the US. and
immediate family ; and (1) contact with religious places (e. g. churches). Table 16
reflects the results of an item analysis of the ISS along each of the four dimensions.
This analysis consisted of first rank ordering the items in descending order according to
the frequency with which the item was endorsed as “often true.” Then calculating an
arithmetic mean for each dimension by dividing the sum of the position ranks of the
items comprising each dimension by the number of items in each dimension. The
results describe, in descending order, the most salient sources of social support to the
participants in the present study. Contact with new fiiends and members of one’s
immediate family is identified as the most salient source of social support for the

participants followed by contact with sources representing one’s own culture.

Table 16. Salient Sources of Social Simport.

 

 

I Dimension I M I SD I

1 contact with new friends in the US. and direct family 11.71 6.7

2. contact with one’s own culture in general 15.86 8.88
3. contact with religious places 26.20 2.77

_4_ contact with community activities and student organizations 35.50 3.03

59

The items comprising the ISS were also analyzed for frequency of being
endorsed as “often true.” The top ten most frequently endorsed items are reflected in
Table 16.1. Perception of one’s family and cultural resources as reliable and important

sources of social support is evident.

Table 16.1: Frequency of Occurrence of Items Comprising the ISS.

 

 

[mm IFrequcncy 1
l. I trust my family 91
2. My family is available when I need it 86
3. I trust my old friends in my home country 82
4. My old fiiends in my home country mean a lot to me 73
5. I am satisfied with my old fiiends in my home country 72
Q. I have contacts with my old fiiends in my home country 59
7. My old friends in my home country are available when I need them 56
8. I trust my extended family 51
1 My friends in the USA. are available when I need them 47
ILL I have contact with my new fiiends in the USA. 47

6O

Inferential Statistics and Hypethesis testing

In the present study, the outcome variable, adjustment, is operationally defined
in terms of scores on the Depression, Anxiety, and Somatization scales of the SCL-90-
R (SCL-90-R; Derogatis, 1990), as well as the total scores on the SUNYA-PSC
(Attanasio, Andrasik, Blanchard, & Arena, 1984). Accordingly, the first group of
analyses consist of the evaluation of two sets of proposed structural equations. Each
set is composed of four equations each of which is a hypothetical firnction of the
relationship between an outcome variable, (i.e. Depression, Anxiety, Psychosomatic
Symptoms, and Somatization) and a group of predictors. The two sets of equations
fully test the first two hypotheses proposed in this study.

Hypothesis 1

This hypothesis proposes that the hardiness level of new international
students will significantly moderate the relationship between stress and adjustment.
Specifically, among new international students with higher levels of hardiness, stress
will be unrelated to adjustment; whereas among new international students with lower
levels of hardiness, stress will be significantly related in a negative direction with
adjustment. Structurally, and given the operational definitions of adjustment in this
study, this hypothesis may be represented as follows:

1. BDEPR = BIADEPR + BZHARDINES + B3STRESS + yHARDINESS*STRESS
2. BANX = BIAANX + fizHARDINES + 83STRESS + yHARDINESS*STRESS
3. PSCPST = BIPSCPRE + BZHARDINES + BgsTRESS + yHARDINESS‘STRESS

4. BSOM = BIASOM + BzHARDlNES + B3STRESS + yHARDINESS‘STRESS

61

Elm

Multiple Regression (MR) technique using the Windows Mcrosoft Corp.,
1995) version of the Statistical Package for the Social Sciences (SPSS-7 .5; 1995) was
used to evaluate the proposed models.

Tables 17.1 and 17.2 contain the results of the evaluation of equation 1.

Table 17.1: The Amount of Variation in BDEPR Explained by Equation 1.

 

 

 

 

Dependent Variable: BDEPR (Depression at time B)

I Predictorts) I 11’ I AR’ I Sig. AF I

a. ADEPR .40 .40 .000

b. ADEPR .51 .11 .000
HARDINESS
STRESS

c. ADEPR .5 1 .00 .364
HARDINESS
STRESS
HARDINESS‘STRESS

Table 17.2: The Evaluation of Coefficients Associated with the Predictor Variables.

I Predictor(s) I r I Sig. I a I t I Sig. I

a. ADEPR 66.70 .000 .63 3. 18 .000

b. ADEPR 33.98 .000 .48 6.22 .000
HARDINESS -.15 -l .9 .060
STRESS .28 3.35 .001

c. ADEPR 25.65 .000 .47 5.99 .000
HARDINESS -. l 5 -1 .9 .056
STRESS .26 3.18 .002
HARDINESS‘STRESS -.07 -.91 .364

Alpha = .05

As expected, the most significant amount of variation in Depression at time 2
(40%) is explained by the baseline level of Depression established at time 1. The
addition of HARDINESS and STRESS into the analysis yields a significant additional

amount of explained variation (11%). The interaction of HARDINESS and STRESS,

62

however, does not yield any additional amount of explained variation. The F statistics
of 25.65 associated with the full model denotes that it explains a significant amount of
the total variation in Depression at time 2. Finally, although the t statistic associated
with the interaction of HARDINESS and STRESS does not warrant the rejection of
the null hypothesis associated with the first hypothesis, as defined in terms of
Depression as outcome, the t statistics associated with ADEPR and STRESS indicate
that these variables are significant predictors of Depression while HARDINESS is
found to be a marginally significant predictor of this outcome.

Tables 18.1 and 18.2 contain the results of the evaluation of equation 2.

Table 18.1: The Amount of Variation in BANX Explained by Equation 2.

 

 

 

 

Dependent Variable: BANX (Anxiety at time B)

I Predictor(s) I 11‘ IAR’ I Sig. AF I

a. AANX .42 .42 .000

b. AANX .46 .04 .027
HARDINESS
STRESS

c. AANX .47 .01 .121
HARDINESS
STRESS
HARDINESS‘STRESS

Table 18.2: The Evaluation of the Coefficients Associated with the Predictor Variables.

I Predictor(s) I r I Sig. I p I r I Sig. I

a AANX 71.55 .000 .65 8.46 .000

b. AANX 27.67 .000 .55 6.76 .000
HARDINESS -.11 -l.28 .205
STRESS .16 1.86 .067

c. AANX 21.67 .000 .54 6.59 .000
HARDINESS -.1 l -1.35 .180
STRESS .14 1.59 .116
HARDINESS‘STRESS -. 12 -1.57 .121

Alpha = .05

63

As expected the most significant amount of variation in Anxiety at time 2
(42%) is explained by the baseline level of Anxiety established at time 1. The addition
of HARDINESS and STRESS into the analysis yields a significant additional amount
of explained variation (4%). The interaction of HARDINESS and STRESS, however,
does not yield any Significant additional amount of explained variation. The F statistic
of 21 .67 associated with the full model denotes that it explains a significant amount of
the total variation in Anxiety at time 2. The t statistic associated with the interaction of
HARDINESS and STRESS does not warrant the rejection of the null hypothesis
associated with the first hypothesis as defined in terms of Anxiety as outcome. The t
statistics associated with the other predictors indicate that other than AANX no other
variables are significant predictors of Anxiety at time 2.

Tables 19.1 and 19.2 contain the results of the evaluation of equation 3.

Table 19.1: The Amount of Variation in PSCPST Explained by Equation 3.

 

 

Dependent Variable: PSCPST (Psychosmatic Symptoms at time B)
LPredictor(s) I R’ I 1111' I Sig. AF 7
a PSCPRE .44 .44 .000
b. PSCPRE .56 . 1 2 .000
HARDINESS
STRESS
c. PSCPRE .59 .03 .014
HARDINESS
STRESS
HARDINESS*STRESS

64

Table 19.2: The Evaluation of the Coefficients Associated with the Predictor Variables.

 

 

I Predictor(s) I r ISig. I B I t ISig.J
a. PSCPRE 79.14 .000 .67 8.90 000
b. PSCPRE 41.41 .000 .55 7.68 .000
HARDINESS -.12 -1.64 .104
STRESS .29 3.80 .000
c. PSCPRE 34.31 .000 .52 7.38 .000
HARDINESS -.13 -1.74 .085
STRESS .27 3.49 .001
HARDINESS*STRESS -.17 -251 .014

Alpha = .05

As expected the most significant amount of variation in Psychosomatic
Symptoms at time 2 (44%) is explained by the baseline level of psychosomatic
symptoms established at time 1. In this case, however, the addition of HARDINESS,
STRESS, and the interaction of HARDINESS and STRESS also yield significant
additional amounts of explained variation (12% and 2.8%, respectively). The F
statistic of 34.31 associated with the full model denotes that it explains a significant
amount of the total variation in psychosomatic symptoms at time 2. The t statistic
associated with the interaction of HARDINESS and STRESS warrants the rejection of
the null hypothesis associated with the first hypothesis as defined in terms of
psychosomatic symptoms as outcome. Finally, the t statistics associated with PSCPRE

and STRESS identify them as a significant predictors of psychosomatic symptoms.

65

Tables 20.1 and 20.2 contain the results of the evaluation of equation 4.

Table 20.1: The Amount of Variation in BSOM Explained by Equation 4.

 

 

 

 

Dependent Variable: BSOM (Somatization at time B)
I Predictor(s) I R’ I 1111’ I Sig. AF I
a. ASOM .40 .40 .000
b. ASOM .49 .08 .001
HARDINESS
STRESS
b. ASOM .49 .00 .255
HARDINESS
STRESS
HARDINESS*STRESS
Table 20.2: The Evaluation of Coefficients Associated with Predictor Variables.
eredicrom) I r I Sig. I p I t I Sigj
a. ASOM 67.62 .000 .64 8.22 000
b. ASOM 30.92 .000 .53 6.87 .000
HARDINESS -.14 -l .77 .081
STRESS .22 2.68 .009
c. ASOM 23.59 .000 .51 6.52 .000
HARDINESS -.15 -1.83 .071
STRESS .20 2.47 .015
HARDINESS*STRESS .09 -1 . 15 .255
Alpha = .05

As expected the most significant amount of variation in Somatization at time 2
(40%) is explained by the baseline level of Somatization established at time 1. The
addition of HARDINESS and STRESS into the analysis yields a significant additional
amount of explained variation (8.3%). The interaction of HARDINESS and STRESS,
however, does not yield any additional amount of explained variation. The F statistic
of 23.59 associated with the full model denotes that it explains a significant amount of
the total variation in Somatization at time 2. Finally, although the t statistic associated
with the interaction of HARDINESS and STRESS does not warrant the rejection of

the null hypothesis associated with the first hypothesis, as defined in terms of

66

Somatization as outcome, the t statistics associated with ASOM and STRESS indicate
that these variables are significant predictors of Somatization, while HARDINESS is

found to be a marginally significant predictor of this outcome.

67

Hypothesis 2

This hypothesis proposes that the new international students’
perceptions of social support will significantly moderate the relationship between stress
and adjustment. Specifically, among new international students with more positive
perceptions of social support, stress will be unrelated to adjustment, whereas among
new international students with less positive perceptions of social support, stress will
be significantly related in a negative direction with adjustment. Structurally, and given
the operational definitions of adjustment in this study, this hypothesis may be
represented as follows:

5. BDEPR = BIADEPR + BZSUPPORT + B3STRESS + ySUPPORT‘STRESS

6. BANX = BIAANX + pzsurroar + p3STRESS + ySUPPORT‘STRESS

7. PSCPST = BIPSCPRE + BZSUPPORT + I33STRESS + ySUPPORT‘STRESS
8. BSOM = BIASOM + I328UPPORT + [33STRESS + ySUPPORT‘S'I‘RESS
Evaluation

Tables 21.1 and 21.2 contain the results of the evaluation of equation 5.

Table 21.1: The Amormt of Variation in BDEPR Explained by Equation 5.
Dependent Variable: BDEPR

LPredictor(s) I 11’ Ian2 I Sig. AF J
a ADEPR .40 .40 .000

 

b. ADEPR .49 .09 .000
SUPPORT
STRESS

c. ADEPR .50 .01 .279
SUPPORT
STRESS
SUPPORT‘STRESS

68

Table 21.2: The Evaluation of Coefficients Associated with Predictor Variables.

 

 

I Predictor(s) I r I Sig. I p I T I Sig. I
a ADEPR 66.70 .000 .63 8.17 .000
b. ADEPR 31.93 .000 .50 6.39 .000
SUPPORT -05 -.70 .487
STRESS .32 3.94 .000
c. ADEPR 24.29 .000 .49 6.20 .000
SUPPORT -.06 -.79 .433
STRESS .32 4.02 .000
SUPPORT‘STRESS -.08 4.09 .279

Alpha = .05
As expected the most significant amount of variation in Depression at time 2
(40%) is explained by the baseline level of Depression established at time 1. The
addition of SUPPORT and STRESS into the analysis yields a significant additional
amount of explained variation (9.4%). The interaction of SUPPORT and STRESS,
however, does not yield any additional amount of explained variation. The F statistic
of 24.29 associated with the full model denotes that it explains a significant amount of
the total variation in Depression at time 2. Finally, although the t statistic associated
with the interaction of SUPPORT and STRESS does not warrant the rejection of the
null hypothesis associated with the second hypothesis, as defined in terms of
Depression as outcome, the t statistics associated with ADEPR and STRESS indicate

that these variables are significant predictors of Depression.

69

Tables 22.1 and 22.2 contain the results of the evaluation of equation 6.

Table 22.1: The Amount of Variation in BANX Explained by Equation 6.

 

 

 

 

Dependent Variable: BANX
findictofls) I R1 I an2 I Sig. AF I
a. AANX .42 .42 .000
b. AANX .45 .03 .060
SUPPORT
STRESS
d. AANX .45 .00 .503
SUPPORT
STRESS
SUPPORT‘STRESS
Table 22.2: The Evaluation of Coefl'lcients Associated with Predictor Variables.
I Predictor(s) I F I Sig. I 11 I t I Sig. I
a. AANX 71.55 .000 .65 8.46 .000
b. AANX 26.68 .000 .57 7.10 .000
SUPPORT -.01 -.09 .928
STRESS .19 2.30 .024
c. AANX 20.01 .000 .57 6.87 .000
SUPPORT -.01 -.14 .889
STRESS .20 2.35 .021
SUPPORT‘STRESS -.05 -.67 .503
Alpha = .05

As expected the most significant amount of variation in Anxiety at time 2

(42%) is explained by the baseline level of Anxiety established at time 1. The addition
of SUPPORT and STRESS into the analysis yields a marginally significant additional
amount of explained variation (3.2%). The interaction of SUPPORT and STRESS,
however, does not yield any additional amount of explained variation. The F statistic
of 20.01 associated with the full model denotes that it explains a significant amount of
the total variation in Anxiety at time 2. Finally, although the t statistic associated with
the interaction of SUPPORT and STRESS does not warrant the rejection of the null

hypothesis associated with the second hypothesis, as defined in terms of Anxiety as

70

outcome, the t statistics associated with AANX and STRESS indicate that these
variables are significant predictors of Anxiety.

Tables 23.1 and 23.2 contain the results of the evaluation of equation 7.

Table 23.1: The Amormt of Variation in PSCPST Explained by Equation 7.

 

 

 

 

Dependent Variable: PSCPST
[irradiating I R’ I AR’ I Sig. AF I
a. PSCPRE .44 .44 .000
b. PSCPRE .56 . 12 .000
SUPPORT
STRESS
e. PSCPRE .56 .00 .210
SUPPORT
STRESS
SUPPORT*STRESS
Table 23.2: The Evaluation of Coefficients Associated with Predictor Variables.
LPredictor(s) I F I Sig. I I1 I t I Sig. I
a. PSCPRE 79.14 .000 .67 8.90 .000
b. PSCPRE 40.96 .000 .56 7.77 .000
SUPPORT -. 10 -l .45 .152
STRESS .31 4.21 .000
c. PSCPRE 31.31 .000 .56 7.72 .000
SUPPORT -.1 l -l .54 .127
STRESS .32 4.29 .000
SUPPORT*STRESS -.09 -1.26 .210
Alpha = .05

As expected the most significant amount of variation in psychosomatic
symptoms at time 2 (44%) is explained by the baseline level of psychosomatic
symptoms established at time 1. The addition of SUPPORT and STRESS into the
analysis yields a significant additional amount of explained variation (12%). The
interaction of SUPPORT and STRESS, however, does not yield any additional amount
of explained variation. The F statistics of 3 1 .31 associated with the full model denotes

that it explains a significant amount of the total variation in psychosomatic symptoms

71

at time 2. Finally, although the t statistic associated with the interaction of SUPPORT
and STRESS does not warrant the rejection of the null hypothesis associated with the
second hypothesis, as defined in terms of psychosomatic symptoms as outcome, the t
statistics associated with PSCPRE and STRESS indicate that these variables are
significant predictors of psychosomatic symptoms.

Tables 24.] and 24.2 contain the results of the evaluation of equation 8.

Table 24.1: The Amount of Variation in BSOM Explained by Equation 8.

 

 

 

 

Dependent Variable: BSOM
BMIctofls) I R’ I AR’ I Sig. AF I
a. ASOM .40 .40 .000
b. ASOM .47 .07 .003
SUPPORT
STRESS
f. ASOM .48 .01 .234
SUPPORT
STRESS
SUPPORT‘STRESS
Table 24.2: The Evaluation of Coefficients Associated with Predictor Variables.
I Predictor(s) I F I Sig. I p I t I Sig. I
a. ASOM 67.62 000 .64 8.22 .000
b. ASOM 29.1 1 .000 .55 7.13 .000
SUPPORT -.04 -.48 .631
STRESS .26 3.29 .001
c. ASOM 22.29 .000 .55 7.08 .000
SUPPORT -.04 -.58 .565
STRESS .27 3.36 .001
SUPPPRT‘STRESS -.09 -1 . 18 .234
Alpha = .05

As expected the most significant amount of variation in Somatization at time 2
(40%) is explained by the baseline level of Somatization established at time 1. The
addition of SUPPORT and STRESS into the analysis yields a significant additional

amount of explained variation (6.8%). The interaction of SUPPORT and STRESS,

72

however, does not yield any additional amount of explained variation. The F statistic
of 22.29 associated with the full model denotes that it explains a significant amount of
the total variation in Somatization at time 2. Finally, although the t statistic associated
with the interaction of SUPPORT and STRESS does not warrant the rejection of the
null hypothesis associated with the second hypothesis, as defined in terms of
Somatization as outcome, the t statistics associated with ASOM and STRESS indicate
that these variables are significant predictors of Somatization.

The second group of analyses consists of the evaluation of two sets of
proposed structural equations. Each set of structural equations is composed of four
equations each of which is a hypothetical firnction of the relationship between an
outcome variable (i.e. Depression, Anxiety, Psychosomatic Symptoms, and
Somatization), and a categorical predictor (e. g. Race), and a covariate (e. g. ADEPR).
The two sets of equations fully test the second two hypotheses proposed in this study.

mamas;

This hypothesis proposes that the adjustment of new international students will
be significantly negatively correlated with the level of discrepancy between US. culture
and the new international students’ native culture. Specifically, the mean of the
adjustment scores of the new international students for whom English is the native
language will be significantly lower than the mean of the adjustment scores of the new
international students for whom English is not the native language. Although, as
stated, this hypothesis can be evaluated using a bi-variate correlation analysis, such a

procedure would not allow for the control of baseline levels of Depression, for

73

example.

Accordingly, this hypothesis will be tested via the Analysis of Covariance
technique and using the Windows version of the Statistical Package for the Social
Sciences (SPSS-7.5, 1995).

The first set of structural equations proposed for hypothesis 3 are:

9. BDEPR = BINATLANGZ + pzADEPR

10. BANX = p.NATLAN<32 + I32AANX

11. PSCPST = BINATLANGZ + EZPSCPRE

12. BSOM = BINATLANGZ + pzASOM
Evaluation

The descriptive analysis of native languages spoken by the participants
indicated that only 5 individuals were native English speakers. Consequently, in the
following analysis a Levene’s Test of Equality of Error Variances was conducted to
evaluate the feasibility of performing an ANCOVA. This test evaluates the null
hypothesis that the error variance of the dependent variable is equal across the two
groups.

Tables 25.1 and 25.2 contain the results of the evaluation of Equation 9.

Table 25.1. The Results of the Levene’s Test
I Dependent Variable I F I Sig I

 

 

 

 

BDEPR .47 .493
Table 25.2. The Results of the ANCOVA.
[Predictor I r I Sifi
ADEPR 65.45 .000
NATLANG2 .54 .464

The results of the Levene’s test indicate that the error variances of the

74

dependent variable, BDEPR, is not equal across the two groups of native and non-
native speakers of English. An ANCOVA of equation 9 was, nevertheless, carried out
yielding no Significant results.

A post-hoc analysis using the variable NATLANGI which divides the
participants into European (1) and Non-European (2) Language Speakers was
performed.

Tables 26.1 and 26.2 contain the results of this analysis.

Table 26.1. The Results of the Levene’s Test.
rDependent Variable I F I Sig I

 

 

 

 

BDEPR 6.32 .014
Table 26.2. The Results of the ANCOVA.
Predictor I F rSig I
ADEPR 63.59 .000
NATLANGI 4.74 .032

These results indicate that controlling for the effects of baseline levels of
Depression, there exists a significant difference across groups in adjustment as defined
by scores on the Depression scale of the SCL-90-R. The performance of an
independent samples t-test (Table 26.3) revealed that the mean Depression score of

speakers of a Non-European language was significantly higher than that of speakers of

a European language.

Table 26.3. The Results of the Independent samples T-Test comparing the means of BDEPR across
NATLANGI.

 

 

FNATLANGl IGroup IN IMean [SD I
BDEPR 1 16 7.81 5.66

2 86 15.85 12.46
1:252

Sig. (2-tailed) = .013

75

Tables 27 .1 and 27.2 contain the results of the evaluation of Equation 10.

Table 27.1. The Results of the Levene’s Test.
I Dependent Variable I F I Sig I

 

 

 

 

BANX 2.38 .126
Table 27.2. The Results of the ANCOVA.
I Predictor I 1‘ Big I
AANX 70.65 .000
NATLANGZ 1.2 .276

The results of the Levene’s test indicate that the error variances of the
dependent variable, BAN X, is not equal across the two groups of native and non-
native speakers of English. A post-hoe analysis using the variable NATLANGI which
divides the participants into European and Non-European Language speakers was

performed. Tables 28.1 and 28.2 contain the results of these analyses.

Table 28.1. The Results of the Levene’s Test.
[Dependent Variable I r I Sig I

BANX 4.25 .042

 

 

Table 28.2. The Results of the ANCOVA.
I Predictor I r I SigJ

AANX 70.37 .000
NATLANGI 3.371 .069

 

 

These results indicate that controlling for the effects of baseline levels of
Anxiety measured at time 1, there exists marginally significant difl‘erences across
groups in adjustment as defined by scores on the Anxiety scale of the SCL-90-R. The
performance of an independent samples t-test (Table 28.3) revealed that the mean
Anxiety score of speakers of a Non-European language was marginally significantly

higher than that of speakers of a European language.

76

Table 28.3. The Results of the Independent Samples T-Test Comparing the Means of BANX Across
NATLANGI.

I NATLANGI I Group I N I Mean I srfl
BANX 1 16 1.94 2.29
2 86 4.58 5.40

 

 

t = 1.92
Sig. (2-tailed) = .058

Tables 29.1 and 29.2 contain the results of the evaluation of Equation 11.

Table 29.1. The Results of the Levene’s Test.

 

 

 

 

[Dependent Variable I r I Sng
PSCPST .039 .844
Table 29.2. The Results of the ANCOVA.
ITredictor I F I Sig I
PSCPRE 80.12 .000
NATLANGZ 1.51 .222

The results of the Levene’s test indicate that the error variances of the
dependent variable, PSCPST, is not equal across the two groups of native and non-
native speakers of English. A post-hoc analysis using the variable NATLANGI which
divides the participants into European and Non-European Language speakers was
performed.

Tables 30.1 and 30.2 contain the results of these analyses.

Table 30.1. The Results of the Levene’s Test.
I Dependent Variable I r I Sig I

 

 

 

 

PSCPST 1.42 .236
Table 30.2. The Results of the ANCOVA.
I Predictor I r I Sifl
PSCPRE 77.64 .000
NATLANGI 1.41 .238

These results also indicate that the error variances of the dependent variable,
PSCPST, is not equal across the two groups of native and non-native speakers of

English; rendering the performance of an ANCOVA unfeasible.

77

Tables 31.1 and 31.2 contain the results of the evaluation of Equation 12.

Table 31.1. The Results of the Levene’s Test

 

 

 

 

[Dependent Variable J r I Sig I
BSOM 3.71 .057
Table 31.2. The Results of the ANCOVA.
I Predictor I r I Sig I
ASOM 65.83 .000
NATLANGZ 1.54 .218

The results of the Levene’s test indicate that the error variances of the
dependent variable, BSOM, is very nearly equal across the two groups of native and
non-native speakers of English. The ANCOVA, however, does not indicate the
existence of a significant difference in Somatization between the English and the non-
English speaker participants.

A post-hoc analysis using the variable NATLANGI which divides the
participants into European and Non-European Language speakers was performed.

Tables 32.1 and 32.2 contain the results of these analyses.

Table 32.1. The Results of the Levene’s Test.
F Dependent Variable I F I Sig I

BSOM 2.65 . 107

 

 

Table 32.2. The Results of the ANCOVA.

 

 

[Predictor I r I Sig I
ASOM 67.91 .000
NATLANGl 4.92 .029

These results indicate that although the error variances of the dependent
variable, BSOM, is not equal across the groups, controlling for the effects of baseline
levels of Somatization measured at time 1, there exists significant differences across

groups in adjustment as defined by scores on the Somatization scale of the SCL-90-R.

78

The performance of an Independent Samples t-test (Table 32.3) revealed that the mean
Somatization score of speakers of a Non-European language was significantly higher

than that of speakers of a European language.

Table 32.3. The Results of the Independent Samples T-Test Comparing the Means of BSOM Across

 

 

NATLANGI.
[NATLANGI IGroup I N IMean I SDJ
BSOM 1 16 2.44 2.48
2 86 5.09 4.96 PI
t=-2.09 5':

Sig. (2-tailed) = .039

 

Eryn—01353. .
This hypothesis proposes that the adjustment of new international

students will be significantly negatively correlated with the level of discrepancy
between US. culture and the new international student’s native culture. Specifically,
the mean of the adjustment scores of the new European (or from European extraction)
international students will be significantly lower than the mean of the adjustment
scores of the new non-European (or from non-European extraction) international
students. Although, as stated, this hypothesis can be evaluated using a bi-variate
correlation analysis, such a procedure would not allow for the control of baseline levels
of adjustment. Accordingly, this hypothesis will be tested via the Analysis of
Covariance technique and using the Windows version of the Statistical Package for the
Social Sciences (SPSS-7.5, 1995).

The set of structural equations proposed for hypothesis 4 are:

13. BDEPR = BIZREGION + fizADEPR

14. BANX = [3.ZREGION + BZAANX

79

15. PSCPST = BIZREGION + 82PSCPRE

16. BSOM = B.ZREGION + 82ASOM
Evaluation

The descriptive analysis of regions of the world represented by the participants

indicated the presence of 17 European (or European extraction) and 8S non-European
(or non-European extraction) individuals. In the following analyses a Levene’s Test of
Equality of Error Variances was conducted to evaluate the feasibility of performing an
ANCOVA. This test evaluates the null hypothesis that the error variance of the
dependent variable is equal across the two groups.

Tables 33.1 and 33.2 contain the results of the evaluation of Equation 13.

Table 33.1. The Results of the Levene’s Test.

 

 

 

 

[Dependent Variable I r I Sig I
BDEPR 4.45 .037
Table 33.2. The Results of the ANCOVA.
Endictor I r I Sig I
ADEPR 66.92 .000
ZREGION 3.88 .052

The results of the Levene’s test indicate that the error variances of the
dependent variable, BDEPR, is equal across the two groups of European and non-
European participants. The results of the ANCOVA indicates the existence of a
marginally significant difference in Depression between the European and the non-
European participants. An independent samples t-test (Table 33.3) was performed to
fiirther investigate this difference. This analysis revealed that the mean Depression
score of Non-European participants was marginally significantly higher than that of

European participants.

80

Table 33.3. The Results of the Independent Samples T-Test Comparing the Means of BDEPR Across
ZREGION.

[Outcome Izruicrorv IN IMean [so I

 

 

BDEPR 1 9.76 7.44
2 85 15.55 12.53
= -l .84
Sig. (2-tailed) = .069
1: European
2: Non-European

Tables 34.1 and 34.2 contain the results of the evaluation Of Equation 14.

Table 34.1. The Results of the Levene’s Test.
I Dependent Variable I F I SFI
BANX 2.10 .150

 

 

Table 34.2. The Results of the ANCOVA.
LPredlctor I F I Sig I
AANX 70.70 .000
ZREGION .50 .479

 

 

The results of the Levene’s test indicate that the error variances of the
dependent variable, BANX, is not equal across the two groups of European and non-
European participants. The results of the ANCOVA does not indicate the existence of
a significant difference in Anxiety between the European and the non-European
participants.

Tables 35.1 and 35.2 contain the results of the evaluation of Equation 15.

Table 35.1. The Results of the Levene’s Test

IDependentVariabie IF ISigI
PSCPST .068 .795

 

 

Table 35.2. The Results of the ANCOVA.
LPredictor I r I Sig I

PSCPRE 84.17 .000
ZREGION 3.49 .065

 

 

The results of the Levene’s test indicate that the error variances of the

dependent variable, PSCPST, is not equal across the two groups of European and non-

81

European participants. The results of the ANCOVA does not indicate the existence of
a significant difference in psychosomatic symptoms between the European and the non-
European participants.

Tables 36.1 and 36.2 contain the results of the evaluation of Equation 16.

Table 36.1. The Results of the Levene’s Test

 

 

 

 

[Dependent Variable I r I Sig I
BSOM 6.90 .010
Table 36.2. The Results of the ANCOVA.

I Predictor I r [ SigJ

ASOM 67.78 .000
ZREGION 2.08 .152

The results of the Levene’s test indicate that the error variances of the
- dependent variable, BSOM, is equal across the two groups of European and non-
European participants. The results of the ANCOVA, however, do not indicate the
existence of a significant difference in Somatiztion between the European and the non-
European participants.
The Effect: ef the Three Component; pf Hardiness Upon Adi ustment

These analyses were undertaken in order to address research question h: are
there any differential effects in adjustment due to the three components of hardiness.
Hardiness, as stated earlier, is comprised of Challenge, Commitment, and Control
(these components have been described in detail elsewhere in this manuscript). The
evaluation of the individual effects of these components was carried out through the
employment of ANCOVA with a set of structural equations. The Challenge
component of hardiness was evaluated using the following structural equations:

17. BDEPR = BIACHALL + pzADEPR

82

18. BANX = BIACHALL + BZAANX
19. PSCPST = BIACHALL + EZPSCPRE
20. BSOM = p.AC1iALL + BZASOM
The term ACHALL was produced by dividing the sample into two groups
according to their scores on the Challenge scale of the PVS-II. More specifically,
those participants with scores in the top one-third were placed in the “High Challenge”
cell and those participants with scores in the lowest one-third were placed in the “Low
Challenge” cell. This division yielded a total of 86 scores (42 High; 44 Low).
Equation 17 was evaluated first. A Levene’s test was carried out in order to
assess the equality of variances between the two cells. Table 37.1 contains the results

Of this test.

Table 37.1. The Results of the Levene’s Test.
f Dependent Variable I F I Sig I

BDEPR . 168 .683

 

 

Although the results of the Levene’s test did not indicate equality of variances
between the two cells, an ANCOVA was, nevertheless, carried out. Table 37.2

contains the results of this analysis.

Table 37.2. The Results of the ANCOVA.
I Predictor I F I Sig I
ADEPR 61.65 .000
ACHALL .828 .366
The results of the ANCOVA do not indicate the existence of a significant

 

 

difference in Depression between the High Challenge and the Low Challenge
participants.
Tables 38.1 and 38.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 18.

83

Table 38.1. The Results of the Levene’s Test
[Dependent VariableJ 1i [ Sig I

 

 

 

 

BANX .558 .457
Table 38.2. The Results of the ANCOVA.
[ Predictor IF [ SigJ
AANX 95.51 .000
ACHALL .249 .619

Although the Levene’s test did not indicate equality of variances across cells,
The ANCOVA procedure was, nevertheless, carried out. The results of the ANCOVA
do not indicate the existence of a significant difference in Anxiety between the High
Challenge and the Low Challenge participants.

Tables 39.1 and 39.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 19.

Table 39.1. The Results of the Levene’s Test
I Dependent Variable [ r I Sig]

PCSPST .3.688 .058

 

 

Table 39.2. The Results of the ANCOVA.
rPredictor I r I Sig I

PSCPRE 61.00 .000
ACHALL .150 .700

 

 

The Levene’s test did indicate marginal equality of variances across cells. The
results of the ANCOVA, however, do not indicate the existence of a significant
difference in psychosomatic symptoms between the High Challenge and the Low
Challenge participants.

Tables 40.1 and 40.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 20.

Table 40.1. The Results of the Levene’s Test
I Dependent Variable I r I Sig I

BSOM .548 .461

 

 

84

Table 40.2. The Results of the ANCOVA.

rPredictor I r I Sifl
ASOM 66.93 .000
ACHALL .538 .466

 

Although the Levene’s test did not indicate equality of variances across cells,
the ANCOVA procedure was, nevertheless, carried out. The results of the ANCOVA
do not indicate the existence of a significant difference in Somatization between the
High Challenge and the Low Challenge participants.

It would seem, therefore, that Challenge is not a significant predictor of
adjustment among international students.

The Commitment component of hardiness was evaluated using the following
structural equations:

21. BDEPR = 13.ACOMM + BZADEPR
22. BANX = B.ACOMM + pzAANX
23. PSCPST = BIACOMM + szSCPRE
24. BSOM = BIACOMM + [32ASOM

The term ACOMM was produced by dividing the sample into two groups
according to their scores on the Commitment scale of the PVS-II. More specifically,
those participants with scores in the top one-third were placed in the “High
Commitment”cell and those participants with scores in the lowest one-third were
placed in the “low Commitment” cell. This division yielded a total of 78 scores (37
High; 41 Low).

Equation 21 was evaluated first. A Levene’s test was carried out in order to

ascertain the equality of variances between the two cells. Table 41.1 contains the

85

results of this test.

Table 41.1. The Results of the Levene’s Test
[Dependent Variable I r I Sig I

BDEPR 14.40 .000

 

 

The Levene’s test did indicate equality of variances between the two cells.

Table 41.2 contains the results of the AN COVA applied to equation 21.

Table 41.2. The Results of the ANCOVA.

 

 

Endictor IF I Sig I
ADEPR 42.11 .000
ACOMM 19.14 .000

The results of the ANCOVA indicate the existence of a significant difference in
Depression between the High Commitment and the Low Commitment participants. In
order to determine the direction of the difference between the two groups, a t test for
independent samples was carried out using the means of the Depression scores of the

two groups. The results of this analysis are reflected in Table 41.3.

41.3. T-Test for Independent Samples: Mean Depression Scores of the High Commitment Vs. the Low
Commitment Groups.

 

 

I Outcome I ACOMM [ N T M [ SD I t I SigJ
BDEPR Low 41 20.68 13.85 4.74 000
High 37 8.89 6.41

These results indicate that the Low Commitment participants, on average,
obtained significantly higher depression scores than the High Commitment participants.
Tables 42.1 and 42.2 contain the results of the Levene’s test and the ANCOVA
as applied to equation 22.
Table 42.1. The Results of the Levene’s Test

LDependent Variable Ir ISigI
BANX 13.67 .000

 

 

86

Table 42.2. The Results of the ANCOVA.

 

 

I Predictor I r I Sig I
AANX 49.68 .000
ACOMM 6.47 .013

The Levene’s test did indicate equality of variances across cells. The results of
the ANCOVA do indicate the existence of a significant difference in Anxiety between
the High Commitment and the Low Commitment participants. In order to determine
the direction of the difference between the two groups, a t test for independent samples
was carried out using the means of the Anxiety scores of the two groups. The results

of this analysis are reflected in Table 42.3.

42.3. T-Test for Independent Samples: Mean Anxiety Scores of the High Commitment VS. The Low
Commitment Groups.

 

 

[Outcome I ACOMM I N I M J SD I t I Sig I
BANX Low 41 6.27 6.58 3.49 .001
High 37 2.24 2.56

These results indicate that the Low Commitment participants , on average,
Obtained Significantly higher anxiety scores than the High Commitment participants.
Tables 43.1 and 43.2 contain the results of the Levene’s test and the ANCOVA
as applied to equation 23.
Table 43.1. The Results of the Levene’s Test

[Dependent Variable [r ISigI
PCSPST 5.92 .017

 

 

Table 43.2. The Results of the ANCOVA.
[Predictor I F I Sig I
PSCPRE 60.51 .000
ACOMM 9.64 .003

 

 

The Levene’s test indicates equality of variances across cells. The results of the
AN COVA indicate the existence of a significant difference in psychosomatic

symptoms between the High Commitment and the Low Commitment participants. In

87

order to determine the direction of the difference between the two groups, at test for
independent samples was carried out using the means of the psychosomatic symptoms

scores of the two groups. The results of this analysis are reflected in Table 43.3.

43.3. T-Test for Independent Samples: Mean Psychosomatic Symptoms Scores of the High Commitment
Vs. the Low Commitment Groups.

 

 

[Outcome I ACOMM I N I M [SD I t [ Slg
PSCPST Low 41 21.90 18.67 3.31 .001
High 37 10.84 8.40

These results indicate that the Low Commitment participants, on average,
obtained significantly higher psychosomatic symptoms scores than the High
Commitment participants.

Tables 44.1 and 44.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 24.

Table 44.1. The Results of the Levene’s Test
I Dependent Variable I F I Sig I

BSOM 46.88 .000

 

 

Table 44.2. The Results of the ANCOVA.
[Predictor [r [Sig I
ASOM 44.91 .000
ACOMM 12.51 .001

 

 

The Levene’s test indicates equality of variances across cells. The results of the
AN COVA indicate the existence of a significant difference in Somatization between
the High Commitment and the Low Commitment participants. In order to determine
the direction of the difference between the two groups, at test for independent samples
was carried out using the means of the Somatization scores of the two groups. The

results of this analysis are reflected in Table 44.3.

88

44.3. T-Test for Independent Samples: Mean Somatization Scores of the High Commitment Vs. The Low
Commitment Groups.

 

 

Fiatcome [ACOMM I N [ M I SD Lt I SigJ
BSOM Low 41 7.00 5.80 3.82 000
High 37 3.08 2.43

These results indicate that the Low Commitment participants, on average,
Obtained significantly higher Somatization scores than the High Commitment
participants.

It would seem, therefore, that Commitment is a significant predictor of
adjustment among international students. In order to evaluate the role of Commitment
in predicting adjustment within the context of stress and in interaction with stress, a
series of MR analyses were applied to the following equations:

25. BDEPR = BIADEPR + BgCOMM + B;STRESS + yCOMM‘STRESS

26. BANX = BrAANX + BZCOMM + 83STRESS + yCOMM*STRESS

27. PSCPST = BIPSCPRE + 82COMM + 83STRESS + yCOMM‘STRESS

28. BSOM = BIASOM + BZCOMM + BgSTRESS + yCOMM‘STRESS
Evaluation

Tables 45.1 and 45.2 contain the results of the evaluation of equation 25.

Table 45.1. The Amount of Variation in BDEPR Explained by Equation 25.

 

 

Dependent Variable: BDEPR
IPredictor(s) I R’ I AR” I Sig. AF I
a. ADEPR .39 .39 .000
b. ADEPR .534 .145 .000
COMM
STRESS
c. ADEPR .540 .007 .243
COMM
STRESS
COMM*STRESS

89

Table 45.2. The Evaluation Of Coefficients Associated with Predictor Variables.

 

 

[Predictor(r) [ r I Sig. I a I t I Sig.J
a. ADEPR 63.43 .000 .62 7.97 .000
b. ADEPR 37.37 .000 .48 6.40 .000
COMM -25 -332 .001
STRESS .25 3.15 .002
c. ADEPR 28.48 .000 .49 6.37 .000
COMM -.06 -3.46 .001
STRESS .32 2.78 .007
COMM‘STRESS -.08 -117 .243
Alpha= .05

As expected the most significant amount of variation in Depression at time 2
(39%) is explained by the baseline level of Depression established at time 1. The
addition of Commitment and Stress into the analysis yields a significant additional
amount of explained variation (15%). The interaction of Commitment and Stress,
however, does not yield any additional amount of explained variation. The F statistic
of 28.48 associated with the firll model denotes that it explains a significant amount of
the total variation in Depression at time 2. Finally, although the t statistic associated
with the interaction of Commitment and Stress is not significant, the t statistics
associated with ADEPR, Commitment, and Stress indicate that these variables are

significant predictors of Depression.

90

Tables 46.1 and 46.2 contain the results of the evaluation of equation 26.

Table 46.1. The Amount of Variation in BANX Explained by Equation 26.

 

 

 

 

Dependent Variable: BANX
IPredictor(s) [ R2 I AR’ [Sig AF I
a. AANX .42 .42 .000
b. AANX .47 .05 .007
COMM
STRESS
c. AANX .48 .01 .228
COMM
STRESS
COMM‘STRESS
Table 46.2. The Evaluation of Coefficients Associated with Predictor Variables.
I Predictor(s) I F I Sig. I8 I t I Sig. I
a. AANX 71.55 .000 .65 8.46 .000
b. AANX 29.41 .000 .56 7.1 l .000
COMM -.17 -2.13 .036
STRESS .14 1.66 .101
c. AANX 22.53 .000 .56 7.15 .000
COMM -.18 -2.28 .025
STRESS .11 1.33 .188
COMM‘STRESS -.09 -1.21 .228
Alpha= .05

As expected the most significant amount of variation in Anxiety at time 2
(42%) is explained by the baseline level of Anxiety established at time 1. The addition
of Commitment and Stress into the analysis yields a significant additional amount of
explained variation (5.7%). The interaction of Commitment and Stress, however, does
not yield any additional amount of explained variation. The F statistic of 22.53
associated with the full model denotes that it explains a significant amount of the total
variation in Anxiety at time 2. Finally, although the 1 statistic associated with the
interaction of Commitment and Stress is not Significant, the t statistics associated with

AANX and Commitment indicate that these variables are significant predictors of

91

Anxiety.
Tables 47.1 and 47.2 contain the results of the evaluation of equation 27.

Table 47.1. Tire Amount of Variation in PSCPST Explained by Equation 27.
Dependent Variable: PSCPST

 

 

 

 

[ Predictor(s) [ R2 I an’ I Sig. AF I
a. PSCPRE .44 .44 .000
b. PSCPRE .56 .12 .000
COMM
STRESS
c. PSCPRE .58 .02 .027
COMM
STRESS
COMM*STRESS
Table 47.2. The Evaluation of Coefficients Associated with Predictor Variables.
[Predictorm I r I Sig. I a I t I Sig]
a PSCPRE 79.14 .000 .67 8.90 000
b. PSCPRE 41.00 .000 .55 7.72 .000
COMM -.1 l -1.46 .147
STRESS .31 3.99 .000
c. PSCPRE 33.26 .000 .54 7.66 .000
COMM -.13 -1.78 .079
STRESS .27 3.51 .001
COMM*STRESS -.15 -2.24 .027
Alpha = .05

As expected the most Significant amount of variation in psychosomatic
symptoms at time 2 (44%) is explained by the baseline level of psychosomatic
symptoms established at time 1. The addition of Commitment and Stress into the
analysis yields a significant additional amount of explained variation (12%). The
interaction of Commitment and Stress also yields an additional significant amount of
explained variation (2.2%). The F statistic of 33.26 associated with the 1111] model
denotes that it explains a significant amount of the total variation in psychosomatic

symptoms at time 2. The t statistic associated with the interaction of Commitment and

92

Stress is significant indicating that the effects of this interaction upon the development

of psychosomatic symptoms are not due to chance alone. The t statistics associated

with PSCPRE and Stress also indicate that these variables are significant predictors of

psychosomatic symptoms.

Tables 48.1 and 48.2 contain the results of the evaluation of equation 28.

Table 48.1. The Amount of Variation in BSOM Explained by Equation 28.
Dependent Variable: BSOM

 

I Predictor(s) [11’

[ an2

I Sig. AF I

 

b.

ASOM .40

ASOM .49
COMM
STRESS

ASOM .50
COMM

STRESS

COMM*STRESS

.40

.09

.01

.000

.000

.241

Table 48.2. The Evaluation of Coefficients Associated with Predictor Variables.

 

 

[Predictorm I r I Sigj p I t I Sig. I
a ASOM 67.62 . .64 8.22 .000
b. ASOM 31.64 .55 7.24 .000
COMM .16 -2.06 .042
STRESS .22 2.68 .009
c. ASOM 24.17 .55 7.22 .000
COMM -.17 -221 .030
STRESS .19 2.34 .021
COMM*STRESS -09 -1.18 .241
Alpha= .05

As expected the most significant amount of Variation in Somatization at time 2

(40%) is explained by the baseline level of Somatization established at time 1. The

addition of Commitment and Stress into the analysis yields a significant additional

amount of explained variation (9%). The interaction of Commitment and Stress,

however, does not yield any additional amount of explained variation. The F statistics

Of24.17 associated with the firll model indicates that it explains a significant amount of
the total variation in Somatization at time 2. Although the t statistic associated with
the interaction of Commitment and Stress is not significant, the t statistics associated
with ASOM and Commitment, and Stress indicate that these variables are significant
predictors of Somatization.
The Control component of hardiness was examined using the following
structural equations:
29. BDEPR = BIACONT + pzADEPR
30. BANX = BIACONT + BZAANX
31. PSCPST = BIACONT + 82PSCPRE
32. BSOM = ptACONT + [32ASOM
The term ACONT was produced by dividing the sample into two groups
according to their scores on the Control scale of the PVS-H. More specifically, those
participants with scores in the top one-third were placed in the “High Control” cell and
those participants with scores in the lowest one-third were placed in the “Low
Control” cell. This division yielded a total of 71 scores (35 High; 36 Low).
Equation 29 was evaluated first. A Levene’s test was carried out in order to
assess the equality of variances between the two cells. Table 49.1 contains the results

of this test.

Table 49.1. The Results of the Levene’S Test.
[Dependent Variable I r I Sig I

BDEPR 2.74 . 102

 

 

Although the Levene’s test did not indicate equality of variances between the

two cells, an ANCOVA was, nevertheless, carried out. Table 49.2 contains the results

94

of the ANCOVA applied to equation 29.

Table 49.2. The Results of the ANCOVA.

 

 

rPredictor I F I Sig I
ADEPR 45.45 .000
ACONT 2.52 .117

The results of the ANCOVA do not indicate the existence of a Significant
difference in Depression between the High Control and the Low Control participants.
Tables 50.1 and 50.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 30.

Table 50.1. The Results of the Levene’s Test.

Dependent Variable Ir [Sign
BANX 4.55 .036

 

 

Table 50.2. The Results of the ANCOVA.
[ Predictor I r I Sig I
AANX 51.23 .000

ACONT 1.73 .193

 

 

The Levene’s test indicates equality of variances across cells. The results of the
ANCOVA, however, do not indicate the existence of a significant difference in Anxiety
between the High Control and the Low Control participants.

Tables 51.1 and 51.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 31.

Table 51.1. The Results of the Levene’s Test
I Dependent Variable I F I Sig I

PCSPST 2.01 .161

 

 

Table 51.2. The Results of the ANCOVA.
[Predictor I r [ Sig I
PSCPRE 57.04 .000

ACONT 6.00 .017

 

 

Although the Levene’s test did not indicate equality of variances across cells,

the ANCOVA was, nevertheless, carried out. The results of the ANCOVA indicate

95

the existence of a significant difference in psychosomatic symptoms between the High
Control and the Low Control participants. In order to determine the direction of the
difl‘erence between the two groups, a t-test for independent samples was carried out
using the means of the psychosomatic symptoms scores of the two groups. The results

of this analysis are reflected in Table 51.3.

51.3. T-Test for Independent Samples: Mean Psychosomatic Symptoms Scores of the High Control Vs.
The Low Control Groups.

 

 

I Outcome I ACONT IN I M I so I t I Sig
PSCPST Low 36 22.83 20.47 1.80 .076
High 35 15.40 13.42

Although these results indicate that the Low Control participants , on average,
obtained higher psychosomatic symptoms scores than the High Control participants;
this difference was only marginally significant. In order to evaluate the role of Control
in predicting psychosomatic symptoms within the context of stress and in interaction
with stress, a MR analysis was carried out using the following equation:

31a. PSCPST = BrPSCPRE + B;CONT + B3STRESS + yCONT*STRESS
Evaluation

Tables 51.4 and 51.5 contain the results of the evaluation of equation 3 la.

Table 51.4. The Amount of Variation in PSCPST Explained by Equation 31a.

 

 

Dependent Variable: PSCPST
[Predictorm I R’ I AR” [Sig AF I
a. PSCPRE .44 .44 .000
b. PSCPRE .56 .12 .000
CONT
STRESS
c. PSCPRE .59 .03 .006
CONT
STRESS
CONT*STRESS

96

 

Table 51.5. The Evaluation of Coefiicients Associated with Predictor Variables.

 

 

I Predicior(e) I F ISig. I B I t I Sig. I

a. PSCPRE 79.14 .000 .67 8.90 .000

b. PSCPRE 41.48 .000 .56 7.81 .000
CONT -.12 -l.67 .098
STRESS .30 4.04 .000

c. PSCPRE 35.34 .000 .55 7.86 .000
com -.11 -157 .119
STRESS .27 3.64 .000
CONT‘STRESS -19 -2.83 .006

Alpha= .05

As expected the most significant amount of variation in psychosomatic
symptoms at time 2 (44%) is explained by the baseline level of psychosomatic
symptoms established at time 1. The addition of Control and Stress into the analysis
yields a significant additional amount of explained variation (12%). The interaction of
Commitment and Stress also adds an additional significant amount of explained
variation (3.4%). The F statistic of 35.34 associated with the fiill model indicates that
it explains a significant amount of the total variation in psychosomatic symptoms at
time 2. The t statistic associated with the interaction of Control and Stress is
significant, denoting that Control moderates the relationship between stress and
psychosomatic symptoms such that the higher the scores on the Control scale of the
PVS-H, the lower the scores on the SUNYA-PSC. Finally, the t statistics associated
with PSCPRE and Stress indicates that these variables are significant predictors of
psychosomatic symptoms.

Tables 52.1 and 52.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 32.

97

Table 52.1. The Results of the Levene’s Test

 

 

 

 

I Dependent Variable I F I sf]
BSOM 11.92 .001
Table 52.2. The Results of the ANCOVA.
[Predictor I F I Sig I

ASOM 45.38 .000

ACONT 2.64 .109

The Levene’s test indicates equality of variances across cells. The results of the
ANCOVA, however, do not indicate the existence of a significant difference in
Somatization between the High Control and the Low Control participants.

R91; 91' Demgmphig Factors in the Adi ustment 9f Internatignal Students

The following analyses were carried out in order to address research question
g: are there any demographic characteristics that help predict the adjustment of new
international students. In addition to the two demographic characteristics tested in the
evaluation of hypotheses 3 and 4 (Native Language and Region of Origin); the effects
of the following demographic variables upon adjustment of international students were
also evaluated: Race , Age, Sex, Relatedness, Marital Status, Father’s Education,
Mother’s Education, Parents Educated in the United States, Acquaintance in the
Lansing/East Lansing area, Proficiency in the English language, and Similarity of
Current Accommodations to Those in the Country of Origin. Of these, only
Acquaintance in the Lansing/East Lansing area was found to be a Significant predictor
of psychosomatic symptoms controlling for the effects of baseline levels of
psychosomatic symptoms.

Table 53.1 and 53.2 contain the results of the Levene’s test and the ANCOVA

as applied to equation 33.

98

33. PSCPST = BrACQELAN + BZPSCPRE

Table 53.1. The Results of the Levene’s Test
[Tiependerrt Variable I F I Sig I

 

 

 

PSCPST 6.35 .013
Table 53.2. The Results of the ANCOVA.
[Predictor I F I Sig I
PSCPRE 63.45 .000

ACQELAN 6.10 .015

The Levene’s test did indicate equality of variances across cells. The results of
the ANCOVA also indicate the existence of a significant difference in psychosomatic
symptoms scores between those participants with acquaintance in the Lansing/East
Lansing area versus those with no acquaintance in the Lansing/East Lansing area. In
order to determine the direction of the difference between the two groups, a t test for
independent samples was carried out using the means of the psychosomatic symptoms

scores of the two groups. The results of this analysis are reflected in Table 53.3.

53.3. T-Test for Independent Samples: Mean Psychosomatic Symptoms Swres of the With Acquaintance
in the LansH/East Lansing area Vs. No A uaintance in the Lansing/East Lansin area Participants.
firmware I ACQELAN I N M I so f t Fifi
PSCPST No 53 19.08 16.32 2.28 .025

Yes 47 12.43 12.27

 

These results indicate that those participants with no acquaintances in the
Lansing/East Lansing area, on average, obtained significantly higher psychosomatic

symptoms scores than participants with acquaintances in the Lansing/East Lansing

area.

99

CHAPTER V

DISCUSSION

“The great tragedy of science: the slaying of an original, beautifiil hypothesis by

an ugly fact.”
Thomas Huxley

The present study was motivated by three general objectives. First, the limited
and contradictory nature of current literature on the health status of international
students enrolled in colleges and universities in the United States and other countries
represents a gap in empirical knowledge regarding the adjustment of this group to their
new environment. This study was designed to assess the prevalence of depression,
anxiety, psychosomatic symptoms and somatization tendencies among this population.
Second, the present study was designed to elicit data on the major sources of stress in
the lives of new international students as well as salient sources of social support.
Third, the dearth of empirical attention to factors explaining the variation in adjustment
of international students limits the ability of those involved in the governance and
welfare of this group to select individuals likely to adjust comfortably to studies
aborad; prepare those likely to exhibit adverse reactions to the stress of such an
undertaking; and counsel international students presenting to college counseling centers
with various types of adjustment disorders.

The first two objectives formed the bases of the following three research
questions: a) in terms of physical and psychological symptoms, how do international

students adjust to their new environment; b) what are the main sources of stress and

100

significant sources of social support among international students; and c) is there a
relationship between immediate post-arrival health and 12-week post-arrival health of
international students.

The third objective was operationalized as an extension of a well established
line of research examining the roles of hardiness and perceptions of social support in
the relationship between stress and illness. This objective formed the basis of the
following research questions: (1) does hardiness play a role in the adjustment of new
international students; e) do perceptions of social support play a role in the adjustment
of new international students; f) does the interaction of perceptions of hardiness and
stress play a role in the adjustment of new international students; g) does the
interaction of perceptions of social support and stress play a role in the adjustment of
new international students; h) are there any demographic characteristics that help
predict the adjustment of new international students; and i) are there any differential
efi‘ects due to the three components of hardiness on the adjustment of new
international students. As well as addressing these questions, the present study was
designed to assess the empirical status of four hypotheses. The exact nature of these
predictions as well as their empirical status will be discussed in this section.

The state of physical and psychological health of international students has not
been the subject of many empirical studies. The available data can be described as
contradictory in that some studies have described this population as a physically and
psychologically resilient group with few adjustment related disorders (e. g. Allen &

Cole, 1987). Other studies have painted a sharply contrasting picture of a group that is

101

afflicted with a wide range of adjustment problems manifested by mental and physical
symptoms (Gunn, 1985; Ward, 1961). Foreign Student Syndrome (F SS) was a label
applied by Lyle Ward to a set of depression-like symptoms that he observed in the
course of an archival study of the health status of international students. Other research
into the effects of geographical displacement on physical and psychological health, lead
to the development of concepts such as Lysgaard’s (1955) U-Curve hypothesis and
Oberg’s (1960) Culture Shock. These, mostly overlapping, concepts suggest that the
process of a sojoumer’s adjustment to his or her new environment is characterized by
an initial period of euphoria and excitement followed by a period of disillusionment and
maladjustment which is then replaced by a normal sense of well-being typical of others
in the environment.

A large body of empirical research has consistently demonstrated a significant
link between stress and a range of physical and psychological symptoms. This
relationship, though significant, is not universal. This variation in adjustment of those
exposed to stress has motivated the exploration and examination of factors which may
moderate, mediate, or otherwise influence this relationship. A set of individual
differences have been conceptualized and tested in the past two decades as possible
mitigating factors in the stress-illness link. Two such factors are hardiness proposed
by Kobasa and her colleagues in 1979 and social support (Anotonsovsky, 1979).
According to Kobasa’s theory hardy individuals are significantly less likely to develop
stress related symptoms when faced with major changes in their circumstances. The

personality of an hardy individual, according to this theory, is characterized by a) a

102

sense commitment to personally meaningful goals and values; b) a deeply held belief
that he/she can influence the outcome of personally relevant events; and c) a belief that
change is normal.

The role of social support as a bufi‘er against the deleterious effects of stress
was first explored within the sociological literature in the early 1970's. This literature
is mostly based upon the premise that individuals with access, actual or perceived, to
such social structures as family/fiiends, counseling centers, employment/placement
offices, financial institutions, etc. are less likely to exhibit stress related symptoms.

The Health Status of International Students

The results of the present study are generally consistent with that portion of the
literature which describes a healthy and resilient population. Overall, the depression,
anxiety, and somatization scores of the group are well within the normal range. This
state of affairs was found to be true both during the period immediately after arrival
and twelve weeks post-arrival. While the results of the psychosomatic symptoms
surveys were less encouraging, there was still reason to believe that as a group the
international students fared better than their domestic counterparts. For example, in a
survey of 27 colleges and universities, Andrasik, Holroyd, and Abell (1979), found that
52 percent of the domestic students surveyed reported headaches at least once or twice
a week. This compares favorably with the frequency of headaches among the
participants of the present study. A number of other symptoms and morbidities with
demonstrated association with stress such as common cold, upper respiratory

infections, cold sores, backaches, and insomnia exhibited elevated rates.

103

Although the overall results are hardly a cause for alarm, a more detailed
analysis of the results revealed the existence of moderate to severe levels of depression
and anxiety among roughly a fifth of the participants. A small number of the
participants reported having experienced suicidal ideations after their arrival in the
United States and these thoughts had persisted twelve weeks into their stay.
Loneliness and hopelessness about the future were also relatively common experiences
among the participants.

Although these results are consistent with Ward’s observations referred to as
the Foreign Student Syndrome, the fact that the majority of the participants did not
report clinical levels of depression, anxiety, or physical symptoms is also consistent
with the overarching observation that the correlation between exposure to stress and
development of symptoms, though significant, is not universal.

Mam Sources of Stress_ and Social Support

The results of the present study are consistent with those of other studies
(Yang & Clum, 1984; for example) in that they describe a population deeply dedicated
to the pursuit of academic excellence. It is not surprising, therefore, to find that
academic pressures is ranked as the most significant source of stress for this group.
Contrary to what might be expected, cultural adjustment and interpersonal difficulties
were ranked third and fifth respectively as major sources of stress. This apparent
paradox is resolved when one takes into account that the majority of the sample
consisted of Asian students. One can speculate that the presence of relatively large

numbers of Asians both as students enrolled at MSU, and in the local community

104

facilitate the transition of these students into the American society by reducing the
need for immediate immersion into the host culture. This speculation is fortified by the
finding that contact with sources representing one’s own culture in the US. is ranked
as the second most salient source of social support by the participants.

The Role of Hardiness in the Adjustment of New International Students

The role of hardiness in the adjustment of new international students was
examined through the evaluation of the hypothesis predicting that the hardiness level of
new international students will significantly moderate the relationship between stress
and adjustment. Specifically, it was predicted that among new international students
with higher levels of hardiness, stress will be unrelated to adjustment; whereas among
new international students with lower levels of hardiness, stress will be significantly
related in a negative direction with adjustment.

Although Kobasa’s hardiness theory forms the empirical basis of this
hypothesis, an accurate comparison of the results of this study with those reported by
Kobasa and her colleagues is rendered diflicult because of differences in the
operationalization of the outcome variable. More specifically, Kobasa defined
adjustment in terms of composite scores on the Seriousness of Illness Survey (Wyler,
Masuda, & Holmes, 1968). This instrument does not yield scores on specific domains
such as depression or anxiety. Furthermore, because Kobasa’s samples consisted
exclusively of males, she excluded all typically feminine symptoms and syndromes.
The results of the present study were consistent with Kobasa’s findings only when

adjustment was defined in terms of psychosomatic symptoms. When adjustment was

105

defined in these terms, hardiness was demonstrated to significantly moderate the
hanriful effects of stress. This finding is consistent with those reported by other
investigators (Bernard & Belinsky, 1993; Williams, Wiebe, & Smith, 1992; and Banks
& Gannon, 1988, for example) who defined adjustment in terms of self-reports of
physical symptoms. The results of the present study also revealed that the baseline
levels of depression, anxiety, and somatization were all significant predictors of
depression, anxiety, and somatization. Although this finding is consistent with those
frequently cited in the literature, the insignificant amount of change in participants’ self
reports of depression, anxiety, psychosomatic symptoms, and somatization between
the time of arrival and twelve weeks post-arrival is inconsistent with Lysgaard’s U-
Curve hypothesis. It is possible, however, that the deterioration in the health of
international students, as predicted by Lysgaard, either occurs some time after the
twelfth week of stay in the new country ( as was found by Parr, Bradley, and Bingi,
1992) or, alternatively, that the entire U-Curve process begins and ends during the
twelve weeks so that by the time the second phase of data collection was completed
the participants had re-adjusted. The latter possibility is less likely, however, given
that the survey instruments instructed the participants to describe their symptoms since
the first survey.

Other noteworthy results obtained in this study were the finding that although
stress consistently and significantly predicted adjustment (except when adjustment was
defined in terms of anxiety symptoms), with the exception of adjustment defined in

terms of psychosomatic symptoms, hardiness did not appear to exert either a main

106

effect on adjustment or a moderating effect on the relationship between stress and
adjustment. The latter finding is inconsistent with both the early general consensus
prevalent in the stress-illness literature centered around the belief that hardiness
exerted a direct, rather than a stress—buffering, effect on the harmful effects of stress,
and more recent studies that have reported findings supportive of the conceptualization
of hardiness as a moderator of the relationship between stress and illness.

The finding that hardiness moderates the experience of psychosomatic
symptoms but not somatization among new international students is paradoxical
considering the high correlation between these two variables. The literature on the
psychological symptoms of Asian students indicates that cultural factors increase the
likelihood that, among this population, emotional conflicts will be experienced
physically. It is thus reasonable to expect that the self-report of somatization would
mirror that of psychosomatic symptoms. A possible explanation for this finding is that
in the present study, psychosomatic symptoms are defined in terms of clearly defined
conditions such as cold sores, common colds, laryngitis, etc. whereas somatization is
defined in terms of more ambiguous experiences such as heavy feelings in arms and
legs, and feeling weak in parts of your body.

As with previous studies which have examined the hardiness components
separately (Wiebe, Williams, & Smith, 1992; Hull, VanTreuren, & Vimelli, 1987), the
predicted relationship between hardiness and adjustment, was more consistently
demonstrated for commitment and, to a lesser degree, for control than for challenge.

More specifically, it was found that participants with higher scores on the commitment

107

component of hardiness consistently and significantly reported lower levels of
depression, anxiety, psychosomatic symptoms, and somatization; while participants
with higher scores on the control component reported significantly lower levels of
psychosomatic symptoms. These findings were obtained even while controlling for the
effects of baseline levels of these symptoms. In contrast with the results of multiple
regression carried out to evaluate the role of hardiness in the relationship between
stress and adjustment, the assessment of the role of commitment in this relationship
revealed that with the exception of psychosomatic symptoms, commitment exerted a
main effect on adjustment; while consistent with those findings, it was demonstrated
that commitment moderated this relationship. Control was also found to be a
moderator of the relationship between stress and self-reports of psychosomatic
symptoms.

There are at least three potential explanations for these findings. First, given
that the participants in this study consist of a group of academically committed
individuals, the majority of whom are from cultures which ascribe high value to
educational success, it seems reasonable to suggest that for this group commitment to
academic success would result in elevated levels of scholarly behaviors. In this way
commitment can be viewed as a trait which exerts its stress-buffering effects by
stimulating adaptive coping behaviors. This indirect effect of commitment in the
relationship between stress and adjustment cannot be evaluated in this study since no
data regarding the scholarly behavior or academic achievement of the participants were

collected. This might form the basis of a future study. Second, as was suggested by

108

Williams et al. (1992) and supported by Contrada’s (1989) findings, the insignificant
role of the challenge component of hardiness may be due to the possibility that unlike
commitment and control, challenge exerts its health-enhancing effects through means
other than adaptive coping processes. More specifically, challenge may impact
appraisal of stress and thus have a beneficial effect on physiological reactivity. Ifthis is
in fact the case, then the self-report survey of health would not yield the kind of data
necessary to assess physiological reactivity. Rather, it is possible that while the
variation in physiological reactivity is accounted for significantly by the challenge
component, the variation in self-reports of depression, anxiety, and psychosomatic
symptoms is not explained adequately by this trait among a group whom despite low
levels of aversion to change does not, nevertheless, engage in adaptive coping
behaviors such as those elicited by commitment. Third, as suggested by Hull et al.
(1987), it is possible that the relative lack of significant findings using the challenge
component is due to psychometric inadequacies of the scale.

The Role of Social Support in the Adjustment of New International Students

It was hypothesized that new international students’ perceptions of social

support will Significantly moderate the relationship between stress and adjustment.
Specifically, it was predicted that among new international students with more positive
perceptions of social support, stress will be unrelated to adjustment, whereas among
new international students with less positive perceptions of social support, stress will
be significantly related in a negative direction with adjustment. Unlike hardiness, social

support was demonstrated to be an entirely unproductive predictor of adjustment in

109

the participants of this study. The results of the study failed to demonstrate the utility
of social support either as a unique predictor of adjustment, (regardless of the manner
in which adjustment was defined) or as a moderator of the relationship between stress
and illness. There are at least two potential explanations for this finding. First, all the
participants in the present study have exhibited both the willingness and the ability to
separate from their significant sources of social support in pursuit of achievement of
their individual academic goals. Second, the majority of the participants in this study
were Asian. The descriptive analysis of the Index of Social Support indicates that this
group does not perceive of community based and student organizations as entities with
essential roles in their adjustment to their new environments. Comparative cultural
studies have also consistently described this population as one that is reluctant to seek
help from sources outside of the immediate family (Fumham & Trezise, 1983, for
example). It seems intuitively plausible to suggest that, on aggregate, this group is
characterized by mostly self-reliant individuals who do not rely upon external sources
of support for the attainment of adaptive fiinctioning.
The Roles of Native Language and Geographical Region of Origm

It was hypothesized that the adjustment of new international students will be
significantly negatively correlated with the level of discrepancy between U. S. culture
and the new international students’ native culture. Specifically, it was predicted that
the mean of the adjustment scores of the new international students for whom English
is the native language will be significantly lower than the mean of the adjustment

scores of the new international students for whom English is not the native language.

110

It was also hypothesized that the adjustment of new international students will be
significantly negatively correlated with the level of discrepancy between U. S. culture
and the new international student’s native culture. Specifically, it was predicted that
the mean of the adjustment scores of the new European (or fi'om European extraction)
international students will be significantly lower than the mean of the adjustment
scores of the new non-European (or fi'om non-European extraction) international
students. The results of this study were consistent with others reporting a greater
degree of difficulty in adjustment among non-native speakers of English and among
students from countries outside of Europe ( Akka, 1967; Brislin, 1981; and Church,
1982, for example). When adjustment was defined in terms of depression it was found
that participants whose native language belonged to the European family of languages
exhibited significantly fewer depressive symptoms than non-European language
participants. This finding was also true when participants were divided into two
regions based upon European and non-European countries of origin. This state of
affairs also approached significance when adjustment was defined in terms of anxiety
symptoms. It must be noted, however, that the discovery of this discrepancy is in
reality not as significant as it might appear given the generally satisfactory state of
adjustment of non-native speakers of English and students form non-European
countries.
Strengths and Limitations

The present study is characterized by at least the following eight strengths.

First, efforts were made to select instruments with highly established statistical

111

properties and/or instruments specifically designed for use with international students.
Unlike a fairly large number of previous studies examining the relationship between
stress and adjustment, the present study employed a broad definition of this vague
construct thereby increasing the construct validity of the study. Third, in response to a
fiequently cited weakness of the hardiness literature as well as the literature on the
health status of international students, the measurement of the baseline levels of
adjustment variables allowed for the controlling of the effects of these factors when
evaluating the roles of other predictor variables of interest. The results of the present
study were derived from a sample that is demographically Similar to the population as a
whole thus enhancing the generalizability of the results. Fifth, the inclusion of every
member of the population of interest in the pool of potential participants also enhanced
the external validity of the findings. The collection of data on a fairly large set of
variables with potentially confounding effects allowed for the assessment of the roles
of these factors as well as for the controlling of any such effects. Seventh, in contrast
with a large number of previous hardiness studies, the present study examined the roles
of the components of hardiness. Finally, various statistical techniques and procedures
were applied to appropriate data (i.e. continuous versus categorical).

This study contained several limitations one of which was the employment of a
survey design. Along with the problem of survey bias associated with this design, the
measurement of symptomology is sometimes contaminated by a set of factors that have
typically influenced the experiencing and endorsement of such experiences. Two such

factors are neuroticism (Benishek, 1994) and cultural background (Chen, 1993). In

112

addition, no information about utilization of prophylactic and/or ameliorative
interventions was collected. This possible confounding factor would have allowed the
investigator to control for the effects of such extraneous variables as ingestion of anti-
depressants or the receiving of counseling and psychotherapy from a mental health
professional. Finally, the problem of unequal variances between groups may have been
eliminated if the sample size was larger.

mph—cations for Future Research and Practice

The implications of the present study can be categorized in terms of
implications for college counseling centers and administrators of international student
programs; and also in terms of implications for those interested in understanding the
factors contributing to the maladjustment and academic difliculties of domestic
students.

Although the results of the present study are consistent with that portion of the
literature describing a rather robust group, there is a small group of international
students, perhaps about twenty percent, who experience moderate to severe
adjustment difficulties. Given the differences in help-seeking perceptions and
behaviors among the various sub-groups of international students, it behooves the
office of international students and scholars to implement a brief screening procedure
in order to identify those new international students with adjustment problems during
the initial phases of their stay in the host country. International students from
European countries and those fi'om European extraction may have the potential to

become effective peer counselors. With basic training in counseling skills, they could

113

serve as group leaders for support groups. Furthermore, they could serve as role
models for helping others understand how to meet basic needs.

The finding that having fiiends and acquaintances in the Lansing/East Lansing
area prior to arrival is significantly associated with the reporting of fewer
psychosomatic symptoms suggests that the development of a mentoring program could
be of value in the facilitation of adjustment of new international students to life and
study in the United States. Such a program could be staffed with volunteers from a
group of international students with at least one year of residence in the United States.

It is possible, of course, that new international students are provided access to
mentors and sponsors during the orientation week organized by the office of
international students. It seems, however, that providing the new international student
with a name, phone number, and/or e-mail address of a sponsor prior to their departure
fiom their country might have the desired adjustment-inducing effects sought by these
programs. The development of such a program can also help ameliorate the
experience of loneliness prevalent among roughly a quarter of the participants.

The finding that academic pressures comprise the most pervasive source of
stress for this group (along with the finding that “it would be the biggest Shame for me
if I fail in school” is the most stressful thought reported by roughly half the
participants) would suggest that some members of this group may be more at risk for
suicidal ideations and attempts. It would seem, therefore, that close monitoring of the
grades of these students would enable administrators to identify those at risk and allow

the counseling center staff to provide such services as may be necessary.

114

Roughly four-fifth of the participants in this study indicated problems in
adjusting to American cuisine. Given that the majority of the participants reside in
university residence halls, it seems reasonable to recommend that those responsible for
the development of menus receive training in the nuances of various ethnic cuisines. It
may also be practical to provide these students with access to kitchen facilities.

Finally, the understanding of the factors contributing to the robustness and
academic success of international students despite the repeated observations that this
group has the ability to persevere, sometimes despite what may seem like
insurmountable odds, may enable fiiture investigators to generate suggestions not only
for the facilitation of adjustment of domestic students to the rigors and demands of
college level work, but also to assist administrators in selecting those candidates with
greater potential for successful adjustment to university life.

Future research in the area of individual differences in the adjustment of
international students can address the somewhat surprising finding that perceptions of
social support did not play a role in the adjustment of this group. Research can also
examine the role of other individual differences such as openness to experience,
neuroticism, and extraversion in the stress-illness link. Another potentially fi'uitful
project would be to further investigate the finding that hardiness appears to play a
mitigating role in the development of psychosomatic symptoms but not in the
development of such psychological symptoms as depression, anxiety, and somatization.

In order to avoid the problems associated with self-report of symptoms, future

researchers can chose more objective means of assessing adjustment. These can

115

 

include the use of laboratory procedures to assess the levels of such physiological

factors as Immunoglobulin-A and stress hormones such as Adrenocorticotropine.

116

 

ll 1.! 49.3.1 Ii. '-

APPENDICES

117

APPENDIX A

Initial Contact Letter
Dear New International Student,

I feel privileged to be among the first Americans to welcome you to our
country. It is my sincere wish that your stay here will be filled with success and
memorable experiences. As a former international student (in Great Britain and the
United States) I feel a special sense of camaraderie with you. By virtue of having left
your family and country in pursuit of knowledge and education, you have joined a
segment of the US. college student population that has experienced continuous
grth both in numbers, and more importantly in its contributions to the intellectual
and cultural atmosphere of this country’s institutes of higher education.

It is safe to assume that in the coming days, weeks, and months, you will
encounter new circumstances and situations. You may, in fact, have already
experienced such changes in your living arrangements, language, or means of
transportation. These new situations can be filled with opportunities and be rather
exciting. They can also require some degree of adaptation and adjustment.

You are one of a small group of newly arrived international students at
Michigan State University who is being asked to provide information on the
relationship between changes in life circumstances and physical and psychological
symptoms. Participation in this study is voluntary and is not, in any way, related to
your enrollment or education at MSU. There are no penalties or any adverse
consequences for not participating.

You may be assured of complete confidentiality. The questionnaires have an
identification number for matching purposes only; thus, I can group together your
initial package with your final responses. Although Your name will never be placed on
any questionnaire, I will, separately, ask you for your name and address so that I may

send you the final survey package in mid-November, 1997.

118

The results of this study are important to people concerned about the effect
that change in life circumstances has on physical and psychological health. This may
include, fixture international students, counseling center staff, the medical community,
and interested citizens. You may receive a summary of the results by contacting me in
March, 1998.

Ifafier reading this letter you wish to participate in this study, please mail in
the enclosed sign-up sheet so that I can contact you and have you complete the first
survey package.

I would be most happy to answer any questions you might have. My name,
phone number, and electronic mail address appear on this letter. Please note that this
is my own personal research project. It is not associated with an interest of MSU.

Thank you in advance for your assistance.

Sincerely,

Saied Ali Mirzadeh M.A. Robbie Steward Ph.D.
Doctoral Program in Counseling Psychology Erickson Hall
Michigan State University Rm. 436

347-4910 355-8503
mirzadeh@pilot.msu.edu devine@msu.edu

P.S. As an added incentive for completing this survey, your identification number
will be entered into a drawing for one of three cash prizes of $150, $100, and
$7 5.

119

 

APPENDIX B

Informed Consent Form

1, , agree to participate in this study and

 

understand the following conditions:

1. I have freely consented to participate in this study and understand that this
means that I agree to complete the initial survey package as well as the final
survey package.

 

2. The purpose of the study has been explained to me, and I both understand the
explanation and what my participation involves.

3. I understand that the risks involved with this survey study are minimal. It is
possible that some of the items in the survey package may elicit an emotional
response. I understand that at all times during data collection qualified
counselors will be available to me should I need them. The names and '
credentials of these counselors appear on this form.

4. The benefits of participating in this study to me include the possibility of a
financial reward as well as the opportunity to learn more about the health and
views of international students. I also agree that the results of this study are
important to people concerned about the effect that change in life
circumstances has on physical and psychological health. This may include,
future international students, counseling center staff, the medical community,
and interested citizens. I understand that I can receive a summary of the results
by contacting the investigators in March of 1998.

5. I understand that my responses will be kept ABSOLUTELY
CONFIDENTIAL. I fidrther understand that any information regarding my
name, address, phone number, and electronic mail address along with any other
indicator of my participation in this study will be destroyed by December 15,
1997.

6. I understand that the initial survey packet will take approximately 10 to 20
minutes to complete. I also understand that the final survey packet will take
approximately 30 to 45 minutes to compete.

120

7. I understand that I may be randomly selected to receive either a $150 first
prize, or a $100 second prize, or a $75 third prize, as an added incentive for
participating in and completing this study. If I am one of the three persons
randomly selected, this payment will be made to me on or before December 15,
1997.

 

Your name (printed)

 

Your signature

 

Date signed

mm

Susan Doyle M.A. (Clinical psychology); L.L.P. Ingham County Community Mental
Health Center

Ellen Berger M.S. (Counseling psychology); post-comps doctoral student in
Counseling psychology (MSU).

Ali Mirzadeh M.A. (Clinical psychology); L.L.P. post-comps doctoral student in
Counseling psychology (MSU).

121

 

Name:

e-mail:

Address:

Phone:

APPENDIX C

HEALTH DY PR E T
participant sign-up sheet

please print
(family name)

 

(first name)

 

 

 

 

 

 

 

122

m (all. sat». nus-rum

APPENDIX D

Dempgpgphig Infprmgtipn

Sex: please circle Male Female

Country of Origin (where are you from?)

 

Date of Birth: Month
Day
Year

 

 

 

Native Language (What Language do you speak at home?)

 

How would you describe your Race and/or Ethnicity? Please circle
Asian
Black
Hispanic, Chicano, or Latino
Pacific Islander
White (European or Middle Eastern)
Mixed race
Other, please specify

 

What is your marital status?
_married/engaged and spouse with me
_mariied/engaged but spouse not with me
_____single

__living with Boyfi'iend/Girlfiiend
_separated

_divorced

_widowed

When did you enter the United States?
Month

Day
Year

 

 

 

123

 

10.

11.

12.

13.

Is this your first trip to the United States?

 

circle one Yes No
If no,
when was the last time you visited the United States?
How long did you stay?

 

How many times have you traveled to the United States?

 

Do you have any relatives living in the United States ?
circle one Yes No
If yes, then please answer the following:

A- Do they live in the Lansing/East Lansing area? Yes No
If not, how far away do they live? circle one
1- Less than 250 miles
2- More than 250 miles

B- How would you characterize your relationship with this (these)
relative(s)? circle one
1-Very close
2-close
3-Not close

Did you know anyone in the Lansing/East Lansing area before your arrival?
circle one Yes No

What was your total TOEFL score?

What is your current or intended major?

 

Are you a graduate or undergraduate student?
Please write which

 

124

 

APPENDIX E

Instructions:

For each of the symptoms listed below, please indicate how

frequently it occurs and how bothersome (or how bad) it is.

quengy Scores (how often)
4-occurs daily

3-oocurs several times a week
2-occurs about once a week
l-occurs about once a month
0-not a problem

queng

(1) Headaches

(2) Backaches

(3) Stomach pain

(4) Asthma

(5) Insomnia

(6) High blood pressure
(7) Fatigue

(8) Nausea

(9) Heart palpitation
(10) Diarrhea or constipation
(1 l) Dizziness

(12) Weakness

(13) Fainting

(l4) Laryngitis

(15) Sore throat

(16) Chest pain

(17) Tonsillitis

(18) Infected eye

(19) Irregular heart beat
(20) Heartburn

(21) Bronchitis

(22) Pneumonia

(23) Sinus infection
(24) Migraine

(25) Common cold
(26) Ear infections

(27) Cold sores

(28) Upset stomach

125

Intensity Scores (how bad)
4-extremely bothersome when occurs
3-severely bothersome when occurs
2-moderately bothersome when occurs
l-slightly bothersome when occurs
O-not a problem

Intensig

 

APPENDIX F

Instructions: Below is a list of problems people sometimes have. Please read each
one carefully, and circle the number that best describes HOW MUCH THAT
PROBLEM HAS DISTRESSED OR BOTHERED YOU SINCE YOUR ARRIVAL
IN THE US. Please do not Skip any items.

0 Not at all

1 A little bit F
2 Moderately -'
3 Quite a bit

4 Extremely

l. Headaches. S l
2. Nervousness or shakiness inside. A
3. Faintness or dizziness. S
4. Loss of sexual interest or pleasure. D I
5. Pains in heart or chest. 8

6. Feeling low in energy or slowed down. D

7. Thoughts of ending your life. D

8. Trembling. A

9. Crying easily. D

10. Feelings of being trapped or eaught. D

11. Suddenly scared for no reason. A

12. Blaming yourself for things. D

13. Pains in lower back. S

14. Feeling lonely. D

15. Feeling sad. D

16. Worrying too much about things. D

17. Feeling no interest in things. D

18. Feeling fearful. A

19. Heart pounding or racing. A

20. Nausea or upset stomach. S

21. Soreness of your muscles. S

22. Trouble getting your breath. S

23. Hot or cold spells. S

24. Numbness or tingling in parts of your body. 8

25. A lump in your throat. S

26. Feeling hopeless about the future. D

27. Feeling weak in parts of your body. S

28. Feeling tense or keyed up. A

29. Heavy feelings in your arms or legs. S

30. Thoughts of death or dying. D

31. Overeating. D

32. Awakening in the early morning. D

33. Sleep that is restless or disturbed. D

34. Feeling everything is an effort. D

35. Spells of terror or panic. A

36. Feeling so restless you couldn't sit still. A

 

126

37. Feelings of worthlessness. D38. Feelings of guilt. D

39. The feeling that something bad is going to happen to you. A
40. Thoughts and images of a frightening nature. A

41. Feeling lonely even when you are with people. D

42. Trouble falling asleep. D

 

127

APPENDIX G

ISS

Directions: Please indicate how much you feel or how often you act the way

described in each of the following statements. Circle one number for each statement.

Please indicate "N/A" if the item does not apply to you.

\OWQO‘M'bUNi—o

Never True 0
Rarely True 1
Sometimes True 2
Often True 3
Not Applieable N/A

. I have contact with my family.

. My new friends in the U.S.A. are available when I need them.

. I have contact with my old friends in my home country.

. Community activities here mean a lot to me.

. I am satisfied with student organizations on campus.

. I trust my family.

. I have contact with my extended family (uncles, aunts, cousins, etc).
. I trust my new friends in the U.S.A

. I trust my extended family (uncles, aunts, cousins, etc).

. I trust the international student center on eampus.

. My family means a lot to me.

. I trust my church (or any religious place) here.

. My extended family (uncles, aunts, etc.) is available when I need them.
. I am satisfied with my old friends in my home country.

. I am satisfied with my family.

. I have contact with the international student center on eampus.

. My old friends in my home country are available when I need them.
. I have contact with student organizations on eampus.

. My family is available when I need it.

. I participate in community activities here.

. I am satisfied with my new friends in the U.S.A.

. I trust my old friends in my home country.

. I have contact with my church (or any religious place) here.

. My extended family means a lot to me.

. I trust the people I meet in community activities.

. My new friends in the U.S.A. mean a lot to me.

. My church (or any religious place) here means a lot to me.

. I am satisfied with the international student center on campus.

. I am satisfied with my extended family (uncles, aunt, etc.).

. I am satisfied with my church (or any religious place) here.

. I have contact with my new friends in the U.S.A.

. The student organizations on campus are available when I need them.
. My church (or any religious place) here is available when I need it.

128

“39152.. “7&akq

 

34. People I meet in community activities are available when I need them.
35. Student organizations on campus mean a lot to me.

36. The international canter on campus is available when I need it.

37. My old friends in my home country mean a lot to me.

38. The international student center on campus means a lot to me.

39. I am satisfied with community activities here

40. I trust student organizations on campus.

129

 

Directions:

APPENDIX H

ILS

following statement. Circle one number for each statement.

®Q¢M§WN~

9.
10
11

12.
13.
14.
15.
16.
17.
l8.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.

29
30
31
32
32

0 Never

1 Rarely

2 Sometimes
3 Often

My English embarrasses me when I talk to people.

I don't like the religions in the U.S.A.

I worry about my academic performance.

I worry about my future career in my home country.

I can feel racial discrimination toward me from other students.
I'm not doing as good as I want to in school.

My English makes it hard for me to read articles, books, etc.

It's hard for me to develop opposite-sex relationships here.

I don't like the ways people treat each other here.

. I don't like American food.

. People treat me badly just because 1 am a foreigner.

I owe money to others.

1 think that people are very selfish here.

I don't like the things people do for their entertainment here.

I ean feel racial discrimination towards me in stores.

I worry about whether 1 will have my future eareer in the U.S.A.
Americans' way of being too direct is uncomfortable to me.

I study very hard in order not to disappoint my family.

I can feel racial discrimination toward me from professors.

I can't express myself well in English.

It would be the biggest shame for me if I fail in school.

I worry about my financial situation.

I don't like American music.

I can feel racial discrimination toward me in restaurants.

My financial situation influences my academic study.

I worry about my filture: will I return to my home country or stay in the U.S.A.?
I haven't become used to enjoying the American holidays.

I don't want to return to my home country, but I may have to do so.
. My English makes it hard for me to understand lectures.

. I want to go back to my home country in the future, but I may not be able to do so.

. My financial situation makes my life here very hard.
. My housing situation is not comfortable
. My housing situation is not comfortable

130

Please indicate how often you feel the way described in each of the

m -r _n‘-— —-a

 

APPENDIX I

Personal Views Survey 11

This questionnaire concerns attitudes toward oneself and the world that may
influence your experience and actions. There are no right or worn answers . What is
important is that you record your current opinions accurately.

Please indicate how you feel about each item by circling a number from 0 to
3 in the space provided. A 0 indicates that you feel the statement is not at all true;
circling a 3 means that you feel the item is completely true. As you will see, the
items are worded very strongly; this is to help you decide the extent to which you
agree or disagree.

0 = Not at all true

I = A little true

2 = Quite a bit true
3 = Completely true

. I often wake up eager to take up my life where it left off the day before

. 1 like a lot of variety in my work

. Most of the time, people listen carefully to what I have to say

. Planning ahead can help avoid most future problems

. What happens to me tomorrow depends on what I do today

. I feel uncomfortable if I have to make any changes in my everyday schedule

. No matter how I hard I try, my efforts will accomplish nothing

. Its hard to imagine anyone getting excited about working

. The ”tried and true" ways are always the best

10. I feel that it's almost impossible to change my family's mind about something
11. Most people who work for a living are just manipulated by their bosses

12. New laws shouldn't be made if they hurt a person's income

13. When you marry and have children you have lost your freedom of choice

14. Trying your best at work really pays off in the end

15. People who never change their minds usually have good judgments

16. Most of what happens in life is just meant to happen

17. It doesn't matter how hard you work at your job, since only the bosses profit by it
18. I don't like conversations when others are confused about what they mean to say
19. Trying hard doesn't pay, since things still don't turn out right

20. Daydreams are more exciting than reality for me

21. I won't answer a question until I'm really sure I understand it

22. When I make plans I'm certain I can make them work

23. I really look forward to my work

24. It doesn't bother me to shift to another task before I have finished the first

25. When performing a difficult task, I know when to seek help

26. It's exciting to learn something about myself

27. Changes in routine are interesting to me

28. It's very hard for me to change a friend's mind about something

\OQQO‘U-bMNI—I

131

 

29

35

36.
37.
38.

39

40.
41.
42.
43.
44.
45.
46.
47.
48.
49.

50

. Thinking of yourself as a free person just leads to frustration
30.
31.
32.
33.
34.

It bothers me when my daily routine gets interrupted

When I make a mistake, there's very little I can do to make things right again
By working hard you can always achieve your goals

I respect rules because they guide me

It's best to handle most problems by just not thinking about them
. Most good athletes and leaders are born, not made

I enjoy it when things are uncertain or unpredictable

People who do their best should get full support from society
Most of my life gets spent doing things that are worthwhile

. Lots of times 1 don't really know my own mind

I have no use for theories that are not closely tied to the facts
Ordinary work is just too boring to be worth doing

If other people get angry at me, it's usually no fault of mine
Changes in routine bother me

It's hard to believe people who say their work helps society

I can't do much to prevent it if someone wants to hurt me

Most days, life is really interesting and exciting to me

People who believe in individuality are only kidding themselves
It's usually impossible for me to change things in my life

I want to be sure someone will take care of me when I get old

. Politicians run our lives

132

41‘... .

 

APPENDIX J

Demographic Information 11

How would you describe your Race and/or Ethnicity? Circle all that apply
Asian
Black
Hispanic, Chicano, or Latino
Pacific Islander
European or of European heritage (i.e. Australian, Canadian, New
Zealander, South African, etc.)
Middle Eastern
Mixed race
Other, please specify

 

Where do you live? circle one

1- live on campus in a residence hall

2- live on campus in the university apartments
3- live in a fraternity or sorority

4- live off-campus in an apartment or house
5- live off-campus with parents or relatives

6- live in a cooperative house

My current housing circumstances are

1- very similar
2- similar
3- different

4- very different
from my housing circumstances in my country.

I am
Very satisfied
Satisfied
Neutral
Not satisfied
Very unsatisfied

with my housing situation.

133

 

5. Since your arrival in the United States have you experienced any unusual events
such as death in the family, marriage, divorce, birth of a child, severe protracted
illness, severe auto accident, etc.?

‘ I. ”a...

 

134

BIBLIOGRAPHY

Adler, N., & Matthews, K. (1994). Health psychology: Why do some
people get sick and some stay well. Annual Review of Psychology, 45, 229-259.

Akka, R. I. (1967). The Middle Eastern Student on the American College
Campus. Journal of the American College Health Association, 15, 251-254.

Allen, F. C. L., & Cole, J. B. (1987). Foreign student syndrome: Fact or
fable? Journal of American College Health, 35. 182-186.

Andrasik, , F., Holroyd, K. A., & Abell, T. (1979). Prevalence of headache
within a college student population: A preliminary analysis. Headache, 20, 384-387.

 

Antonovsky, A. Health, stress, and coping. San Francisco: Jossey-Bass,
1979.

Bandura, A. (1977). A social learning theory. Englewood Cliffs, NJ .:
Prentice-Hall.

Banks, J. K., & Gannon, L. R., (1988). The influence of hardiness on the
relationship between stressors and psychosomatic symptomology. American Journal
of Community Psychology, 16 (I), 25-37.

Benishek, L. A. (1993). Hardiness, life stress, and neuroticism: A structural
equation model of self-reported illness. Unpublished doctoral dissertation, Michigan
State University, East Lansing.

Bernard, L. C., & Belinsky, D. (1993). Hardiness, stress, and
maladjustment: Effects on self-reported retrospective health problems and
prospective health center visits. Journal of Social Behavior and Personality, 8, 97-
110.

Bodnar, J. C., & Kiecolt-Glaser, J. K. (1994). Caregiver depression after
bereavement: Chronic stress isn't over when it's over. Psychology and Aging, 9,
372- 380.

Brett, J. M. (1980). The effects of job transfer on employees and their
families. In C. L. Cooper & R. Payne (Eds), Current concerns in occupational
stress (pp. 175-210). New York: Wiley.

Brislin, R. (1981). Cross-cultural encounters. New York: Pergamon.

135

Brown, J. D., & McGill, K. L. (1989). The cost of good fortune: When
positive life events produce negative health consequences. Journal of Personality
and Social Psychology, 57, 1103-1110.

Cannon, W. E. (1942). "Voodoo" death. American Anthropologist, 44,
169- 182.

Cassel, J. (1976). The contribution of the social environment to host
resistance. American Journal of Epidemiology. 104, 107-123.

Cemovsky, Z. Z. (1989). Life stress measures and reported frequency of I
sleep disorders. In T. W. Miller (Ed), Stressful life events. Madison, CT:
International Universities Press.

 

Chen, G-M., (1993). Self-disclosure and Asian students' abilities to cope 4'
with social difficulties in the United States. The Journal of Psychology, 127, 603- E
610. .

Church, A. T. (1982). Sojourner adjustment. Psychological Bulletin. 91,
540-572.

Cohen, J. (1977). Statistical power analysis for the behavioral science. NY:
Academic Press.

Cohen, S., & Edwards, G. M. (1989). Personality characteristics as
moderators of the relationship between stress and disorder. In R. W. J. Neufeld
(Ed.),Advances in the investigation of psychological stress (pp. 235-283). New
York: Wiley.

Cohen, S., & Hoberman, H. M. (1983). Positive events and social supports
as buffers of life change stress. Journal of Applied Social Psychology, 13, 99-125.

Contrada, R. J. (1989). Type A behavior, personality hardiness, and
cardiovascular responses to stress. Journal of Personality and Social Psychology,
57,895-903.

Contrada, R. J., Leventhal, H., & O'Leary, A. (1991). Personality and
health. In L. A. Pervin (Ed), Handbook of personality: Theory and research (pp.
63 8-669). New York: The Guilford Press.

Dean, A., & Lin, N. (1977). The stress-buffering role of social support?

Problems and prospects for systematic investigation. Journal of Nervous and Mental
Disease, 165, 403-417.

136

DeArmond, M. M. (1983). Mental health and international students.
National Association of Foreign Student Affairs Newsletter, April/May, 137-150.

Derogatis, L. R., Rickels, K., & Rock, A. (1976). The SCL-9O and the
MMPI: A step in the validation of a new self-report scale. British Journal of
Psychiatry, 128, 280-289.

Dohrenwend, S. D., & Dohrenwend, B. P. (Eds). (1974). Stressfill life
events: Their nature and effects. New York: Wiley.

Domingues, P. M. (1970). Student personeel services for international
students. Journal of the National Association of Women Deans, Administrators and I
Counselors, 33, 83-89.

Ebbin, A. J ., & Blankenship, E. S. (1988). Stress-related diagnosis and 1
barriers to health care among foreign students: Results of a survey. Journal of
American College Health, 36, 311-312. is

 

Fiske S. T., & Taylor, S. E. (1991). Social Cognition. New York:
McGraw- Hill.

Funk, S. C. (1992). Hardiness: A review of theory and research. Health
Psychology, 11, 335-345.

Funk, S. C., & Houston, B. K. (1987). A critical analysis of the hardiness
scale's validity and utility. Journal of Personality and Social Psychology, 53(3), 572-
578.

Fumham, A., & Trezise, L. (1983). The mental health of foreign students.
Social Science and Medicine, 17, 365-370.

Ganellen, R. J ., & Blaney, P. H. (1984). Hardiness and social support as
moderators of the effects of life stress. Journal of Personality and Social
Psychology, 47, 156-163.

Gentry, W. D., & Kobasa, S. C. (1984). Social and psychological resources
mediating stress-illness relationships in humans. In W. D. Gentry (Ed), Handbook
of behavioral medicine (pp. 87-116). New York: Guilford Press.

Gore, S. (1978). The effect of social support in moderating the health

consequences of unemployment. Journal of Health and Social Behavior, 19, 157-
165.

Gunn, A. (1988). Health care and foreign students. Journal of American
College Health, 36, 345-348.

137

Hartmann, E. L. (1985). Sleep disorders. In H. 1. Kaplan & B. J. Sadock
(Eds), Comprehensive textbook of psychiatry (4th ed). Baltimore: Williams &
Wilkins.

Heller, K., & Monahan, J. (1977). Psychology and community change.
Homewood, IL: Dorsey.

Hendricks, G. L., & Skinner, K. A. (1977). Adaptive social patterns of
foreign students. Journal of College Student Personnel, 18, 124-127.

 

ls
Hinkle, L. E., Jr. (1974). The effects of exposure to culture change, social '
change, and changes in interpersonal relationships on health. In B. S. Dohrenwend
& B. P. Dohrenwend (Eds), Stressful life events: Their nature and effects (pp. 9-44).
New York: Wiley.
Holmes, T. H., & Masuda, M. (1974). Life change and illness susceptibility. L
In B. S. Dohrenwend & B. P. Dohrenwend (Eds), Stressful life events: Their nature It.

and effects. New York: Wiley.

Holohan, C. J .,& Moos, R. H. (1985). Life stress and health: Personality,
coping, and family support in stress resistance. Journal of Personality and Social
Psychology, 49, 739-747.

Holroyd, K. A., & Coyne, J. (1987). Personality and health in the 1980's:
Psychosomatic medicine revisited. Journal of Personality, 55, 359-375.

House, J. S. (1981). Work stress and social support. Reading, MA:
Addison- Wesley.

Hull, J. G., Van Treuren, R. R, & Vimelli, S. (1987). Hardiness and health:
A critique and alternative approach. Joumal of Personality and Social Psychology,
53(3),518-530.

Hull, W. F. (1978). Foreign students in the United States of America:
Coping behavior within the educational environment. New York: Praeger.

Jemmott, J. B., 1H, & Magloire, K. (1988). Academic stress, social support,
and secretory immunoglobulin A. Journal of Personality and Social Psychology,
55, 803-8 10.

Johnson, J. H., & Sarason, I. G. (1979). Moderator variables in life stress

research. In I. G. Sarason and C. D. Spielberger (Eds), Stress and anxiety (V016).
Washington, DC: Hemisphere Publishing.

138

Kasl, S. V., & Cobb, S. (1970). Blood pressure changes in men undergoing
job loss: A preliminary report. Psychosomatic Medicine, 32, 19-38.

Klineburg, O., & Hull, W. (1979). At a foreign university: An international
study of adaptation and coping. New York: Praeger.

Ko, Y.-H. (1979). The mental health of the overseas chinese students in the
new environment. Acta Psychologica Taiwanica, 21, 1-7.

Kobasa, S. C. (1979). Personality and resistance to illness. American
Journal of Community Psychology, 7, 413-423.

Kobasa, S. C. (1982b). The hardy personality: Toward a social psychology
of stress and health. In G. S. Sanders & J. Suls (Eds), Social psychology of health
and illness (pp. 3-32). Hillsdale, NJ: Lawrence Erlbaum Associates.

Kobasa, S.C., & Maddi, S. R. (1977). Existential personality theory. In R.
Corsini (Ed), Current personality theories. Itasca, 111.: Peacock.

Kobasa, S. C., & Puccetti, M. C. (1983). Personality and social resources in
stress resistance. Journal of Personality and Social Psychology, 45, 839-850.

Kobasa, S.C., Maddi, S. R., & Courington, S. (1981). Personality and
constitution as mediators in the stress-illness relationship. Journal of Health and
Social Behavior, 22, 368-378.

Kobasa, S. C., Maddi, S. R., & Kahn, S. (1982). Hardiness and health: A
prospective study. Journal of Personality and Social Psychology, 42, 168-177 .

Kobasa, S. C., Maddi, S. R., & Puccetti, M. C. (1982b). Personality and
exercise as buffers in the stress-illness relationship. Journal of Behavioral Medicine.
5,391-404.

Krueger, D. W. (1981). Stressful life events and the return to heroin use.
Journal of Human Stress, 7(2), 3-8.

Lazarus, R. S. (1966). Psychological stress and the coping process. New
York: McGraw-Hill.

Leavy, R. L. (1983). Social support and psychological disorder: A review.
Journal of Community Psychology, 11, 3-21.

Lloyd, G, Alexander, A. A., Rice, D. G., & Greenfield, N. S. (1980) Life
events as predictors of academic performance. Jorunal of Human Stress, 6(3), 15-
26.

139

 

Lysgaard, S. (1955). Adjustment in a foreign society: Norwegian Fulbright
Scholars visiting the United States. International Social Science Bulletin, 7, 45-51.

Maddi, S. R., & Kobasa, S. C. (1984). The hardy executive: Health under
stress. Homewood, 11: Dow Jones-Irwin.

Maddi, S. R., Kobasa, S. C., & Hoover, M. (1979). An alienation test.
Journal of Humanistic Psychology, 19, 7 3-76.

Malatesa, V. J., & Adams, H. E. (1984). The sexual dysfunctions. In H. E.
Adams & P. B. Sutker (eds), Comprehensive handbook of psychopathology. New
York: Plenum.

Marion, PB. (1986). Research on foreign students at college and
universities in the United States. In K. R. Pyle (Ed), Guiding the development of
foreign students. San Francisco: Jossey-Bass.

May, R. (1939). The art of counseling. Nashville: Bingdon-Cokesbury.
Paperback reprint: Nashville: Apex Books, 1967a.

May, R. (1969c). Love and will. New York: Norton.

Mechanic, D. (1974). Social structure and personal adaptation: Some
neglected dimensions. In G. V. Coelho, D. A. Hamburg, and J. E. Adams (Eds),
Coping and adaptation. New York: Basic Books.

Miller, D. F ., & Harwell, D. J. (1983). International students at an American
University: Health problems and status. The Journal of School Health, 53, 45-49.

Miller, B. C., & Sollie, D. L. (1986). Normal stresses during transition to
parenthood. In R. H. Moos (Ed), Coping with life crises: An integrated approach.
New York: Plenum.

Nash, D. (1991). The Course of sojourner adaptation: A new test of the U-
curve hypothesis. Human Organization, 50, 283-286.

Oberg, K. (1960). Culture shock: An adjustment in new cultural
environments. Practical Anthropology, 7, 177-182.

Pagel, M. D., Smilkstein, G., Regen, H., & Montano, D. (1990).
Psychological influences on new-bom outcomes: A controlled prospective study.
Social Science & Medicine, 30, 597-604.

Parr, 6., Bradley, L., Bingi, R. (1992). Concerns and feelings of
international students. Journal of College Student Development, 33, 20-25.

140

 

Pruitt, F. J. (1978). The adaptation of foreign students on American
campuses. Journal of the National Association of Women Deans, Administrators
and Counselors, 41, 144-147.

Rabkin, J. G., & Streuning, E. L. (1976a). Life events, stress, and illness.
Science, 194, 1013-1020.

Rabkin, J. G., & Streuning, E. L. (1976b). Social change, stress, and illness:
A selective literature review. Psychoanalysis and Contemporary Science, 5, 573-624.

Rahe, R. H., & Arthur, R. H. (1978). Life change and illness studies.
Journal of Human Stress, 4, 3-15.

Ray, M. M. (1967). Health and disease among international students in the
United States. Journal of American College Health Association, 15, 361-364.

Rhodewalt, F., & Zone, J. B. (1989). Appraisal of life change, depression,
and illness in hardy and nonhardy women. Journal of Personality and Social
Psychology, 56, 81-88.

Rosengren, A., Tibblin, G., & Wilhelmsen, L. (1991). Self-perceived
psychological stress and incidence of coronary artery disease in middle-aged men.
American Journal of Cardiology, 68, 1171-1175.

Sam, D. L., & Eide, R. (1991). Survey of mental health of foreign students.
Scandinavian Journal of Psychology, 32, 22-30

Selye, H. (1956). The stress of life (lst ed). New York: McGraw-I-Iill.

Seligman, M. E. P. (1974). In depression and learned helplessness, R. J.
Friedman & M. M. Katz (Eds), The Psychology of depression: Contemporary
theory and research. New York: Wiley.

Sklar, L. A., & Anisman, H. (1979). Stress and coping factors influence
tumor growth. Science, 205, 513-515.

Stone, A. A., Reed, B. R., & Neale, J. M. (1987). Changes in daily event
frequency precede episodes of physical symptoms. Journal of Human Stress, 13, 70-
74.

Suls, J ., & Rittenhouse, J. D. (1987). Personality and physical health: An
introduction . Journal of Personality, 55, 157-167.

Tabachnik, B. G., & Fidell, L. S. (1989). Using multivariate statistics.
Philadelphia, Pa: Harper & Row.

141

 

Torrey, E. F., Van Rheenan, F. J ., & Katchadourian, H. A. (1970).
Problems of foreign students: An overview. Journal of American College Health
Association, 19, 83-86.

Totman, R., Kiff, J ., Reed, S. E., & Craig, J. W. (1980). Predicting
experimental colds in volunteers from different measures of recent life stress.
Journal of Psychosomatic Research, 24, 155-163.

Ward, L. E. (1962). Some observations of the underlying dynamics of
conflict in foreign students. Journal of American College Health Association, 10,
430-440.

Wehrly, B (186). Counseling international students, issues, concerns, and
programs. International Journal for the Advancement of Counseling, 9, 11-22

Weiten, W. (1988). Pressure as a form of stress and its relationship to
psychological symptomolgy. Journal of Social and Clinical psychology, 6(1), 127-
139.

Weiten, W., & Lloyd, M. L. (1994). Psychology applied to modern life.
Pacific Grove, Ca: Brooks/Cole.

Wiebe, D.J., Williams, P.G., & Smith, T.W. (1991). Convergent and
discriminant validity of the hardiness scales: Another look at the potential
neuroticism confound. Unpublished manuscript. University of Utah Department of
Psychology, Salt Lake City, Utah.

Williams, N. A., & Deffenbacher, J. L. (1983). Life stress and chronic yeast
infection. Journal of Human Stress, 9, 26-31.

Williams, P. G., Wiebe, D. J ., & Smith, T. W. (1992). Coping processes as
mediators of the relationship between hardiness and health. Journal of Behavioral
Medicine, 15, 237-255.

Wyler, A. R., Masuda, M., & Holmes, T. H. (1968). Seriousness of Illness
rating scale. Journal of Psychosomatic Research, 1 1, 363-3 75.

Yang, H, & Clum, G. A. (1994). Life stress, social support, and problem-
solving skills predictive of depressive symptoms, hopelessness, and suicide ideation

in an Asian student population: A test of a model. Suicide and Life-Threatening
Behavior, 24, 127-139.

Yang, B., & Clum, G. A. (1995). Measures of life stress and social support
specific to an Asian student population. Journal of Psychopathology and Behavioral
Assessment, 17, 51-67.

142

Zeiss, A. M. (1980). Aversiveness versus change in the assessment of life
stress. Journal of Psychosomatic Research, 24, 15-19.

Zurin, L. M. & Rubin, R. T. (1967). Paranoid psychotic reactions in foreign

students from non-Western countries. The Journal of the American Health
Association, 15, 220-226.

143

WIT—‘— —‘ _-

 

"71111111111till“