mm.
lfiifitflfi‘m ‘
H .

wv N’l kw ‘
..

‘1, 1 Au
4d J r:

 

n u
m.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

M 11111111111111111111111111111"111111111411 L

31293 01015

This is to certify that the

dissertation entitled

THE OH NO! SYNDROME: A LANGUAGE EXPECTATION
MODEL OF UNDERGRADUATES' NEGATIVE
REACTIONS TOWARD FOREIGN
TEACHING ASSISTANTS
presented by

NAGESH RAO

has been accepted towards fulfillment
of the requirements for

the Doctor of PhilosogggflfinCommunication

 

 

47” WHO/Z

1 Major professor

June 27, 1994
Date

 

MS U is an Aflirmativc Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY
Michigan State
Unlverslty

 

 

 

PLACE ll RETURN BOXto mallow. Mutton yum.
TO AVOID FINES Mum on or More dd. duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU loAn Mun-live Action/Equal Oppommlty 1m
Walla-9.1

THE OH NO! SYNDROME: A LANGUAGE EXPECTATION MODEL
OF UNDERGRADUATES’ NEGATIVE REACTIONS TOWARD
FOREIGN TEACHING ASSISTANTS

BY

Nagesh Rao

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Communication

1994

ABSTRACT
THE OH NO! SYNDROME: A LANGUAGE EXPECTATION MODEL

OF UNDERGRADUATES’ NEGATIVE REACTIONS TOWARD
FOREIGN TEACHING ASSISTANTS

BY
Nagesh Rao

The "Oh No! Syndrome" represents the reactions of U.S.
undergraduates towards a foreign teaching assistant (TA)
when s/he walks into class on the first day of a term.

Since 1982, the Oh No! Syndrome has been studied as a
"foreign TA language problem." It became apparent, however,
that training the foreign TA alone would not get rid of the
Oh No! phenomena. Researchers have argued recently that
U.S. undergraduates play an equally important role for the
existence of the Oh No! Syndrome.

To better understand the role of undergraduates in the
Oh No! Syndrome, this study offered the Language Expectation
Model (LEM) to explore undergraduates' affective, cognitive,
and behavioral responses when they interact with a foreign
TA on the first day of class. The LEM predicted that
undergraduates have a strong negative expectation of their
foreign TAs’ language skills, that is, they will speak with
an accent that is difficult to follow. Specifically, when

undergraduates' expectations are confirmed, they should feel

angry and anxious, evaluate the foreign TA negatively, and
should drop a class taught by a foreign TA. Conversely,
when undergraduates' expectations are violated, they should
be relieved and happy, evaluate the foreign TA favorably,
and should stay in a class taught by a foreign TA.

Students were nested in a 2 x 2 x 2 design (confirmed
versus violated expectations, strong versus weak
expectations, chemistry lab lecture versus interpersonal
communication lecture). Undergraduate students began the
experiment by answering a questionnaire measuring their
language expectations of foreign TAs. After watching a
foreign TA lecture on a video tape, students evaluated the
foreign TA on task, relational, and communication
competence.

Results indicated that the data were consistent with
most of the hypotheses. When expectations were confirmed,
students felt more angry and anxious, evaluated the foreign
TA less favorably, and were more likely to drop a class
taught by a foreign TA, than when expectations were
violated. Interestingly, when expectations were violated,
students with strong expectations had more positive
evaluations of foreign TAs than students with weak
expectations, in certain cases.

The implications of these findings for: (1) the LEM,
(2) training undergraduates to communicate more effectively

with foreign TAs, and (3) future research were discussed.

To my parents Sarasa and Raghavendra Rao

iv

ACKNOWLEDGMENTS

First and foremost, I would like to acknowledge the
efforts of my parents Sarasa and Raghavendra Rao, my sister
Pammi and my brother Sudhakar. For the past 32 years, they
have loved, cared, and supported me through my ups and
downs. In their infinite wisdom, they urged me to study in
the U.S. by explaining, "Apart from learning about
communication, you will also learn about life in a way that
only travel teaches." Without their concern for quality
education and many phone calls with the constant inquiry,
"When are we going to meet Dr. Nagesh Rao?," my Ph.d. would
never have become a reality.

Several people have made significant contributions to
my Ph.D. education. Most of all, I thank Steve Wilson, my
advisor and chair, for providing a challenging learning
environment. His propensity in asking tough questions,
giving lots of valuable feedback, and competing with me to
grow the longest pony-tail, provided the foundation for a
truly liberal education. Frank Boster guided me through the
initial difficult years, taught me to study a topic from a
variety of perspectives, and never to ignore rival
explanations. Mary Bresnahan introduced me to my

V

dissertation topic, and wholeheartedly supported my effort
to take the issue to new heights. James Dearing showed me,
by demonstration, a mentoring process that is a wonderful
combination of challenge and support. Galen Bodenhausen, in
his quiet and inimitable manner, provided the education to
look at the role of emotions and stereotyping in my
dissertation. I thank Nancy Mark for providing me the
opportunity to apply my research interests in intercultural
communication on a day-to-day basis.

I thank Ann Wooten for taking care of the
administrative details of my scholarship, and Deb Tigner for
making copies, sending faxes, and always coming through with
a smile. A special thanks goes to Marge Barkman for being
my philosopher, friend, guide, astrologer, and graduate
secretary all rolled into one! She always said the right
thing to make me feel better. My friends Sathya, Jennie
Grau, Money, Ravi, Nishant, Vijay and Karin, Mohan and
Nikki, offered the kind of solace, affection, and support
that only true friends give. I would also like to thank the
Wilson, Boster, and Dearing families for providing me with
hot meals, having me stay in their houses to complete my
dissertation, and most of all, making me feel a part of
their "homes." Last but not the least, the warmth,
affection, patience, support, and proofreading skills of
Jennifer Smith played an invaluable role in completing my
dissertation.

vi

TABLE OF CONTENTS

QHAEIEB Page
LIST OF TABLES O O O O O O O O O ..... O O O O I O O O O O O O O O O O O O O O O O O O O 0 ix
LIST OF FIGURES O O O O O O ..... O O O O O O O O O O O O O O O O O O O O O O O O O O O O x

I. INTRODUCTION .0.0.0.0.0000...OOOOOOOOOOOOOOOOOOOO 1

II. LITERATURE REVIEW................................ 6
Foreign TAs’ Language Problem.................... 7
Perception of Some Government 0fficials..... 7
Perception of University Officials.......... 9
Perception of Parents....................... 10
Perception and Role of Undergraduates....... 11

Summary of Literature Review..................... 15

III. LANGUAGE EXPECTATION MODEL......... ...... ........ 19
Introduction................ ..... ................ 19
Language Expectations............................ 23
Language Expectation Model....................... 27

Identified as Foreign TA ........ ............ 27
No Prior Language Expectations for
Foreign TAs. ..... ............... ............ 30

Strong versus Weak Prior Expectations for
Foreign TAs................................. 34
Hypotheses .................................... 43
Hypotheses about Affect/Emotion............. 47
Hypotheses about Evaluation of the
Foreign TA.................................. 50
Hypotheses about Subtyping the Foreign TA... 52
Hypotheses about the Dimensions of
Evaluation.................................. 53
Hypotheses about Behavioral Intentions...... 54

IV 0 METHODS O O O O I O O O I O O O O O O O O O O O O I O O O O O O O O O O O O O ....... 57
overView O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 57
Phase 1: Pretest for Stimulus Materials..... ..... 58

Phase 2: Experimental Measures and Procedures.... 62

vii

QHAEIEB Page

V. RESULTS......... ...... ...... .................... . 67
Overview......................................... 67
Profile of the participants...................... 67
Manipulation check............................... 69
Hypotheses about Affect/Emotion ........ ..... ..... 71
Hypotheses about Evaluation of the Foreign TA.... 76
Hypotheses about Subtyping the Foreign TA........ 84
Hypotheses about the Dimensions of Evaluation.... 86
Hypotheses about Behavioral Intentions........... 88

VI. DISCUSSION ...... .... ..... .... ...... . ............. 94
Overview......................................... 94
General patterns of findings..................... 94

Confirmed hypotheses........................ 94
Disconfirmed hypotheses......... ............ 96
Implications for the Language Expectation Model.. 100
Suggestions for training students................ 101
Implications for future research................. 105
How would I conduct this study differently
now? ........................ .. .............. 105
What should we do next?.... ................. 107
How can we extend the scope of the LEM?..... 110
Going beyond the LEM........................ 111

COhClUding Remarks 0 O O O O O O O O O O O O O O O O ...... O O O O O O O O 114
APPENDICES. O O O O O O O O O O I O 000000000000000000000000000 O O O O 115
LIST OF REFERENCES 0 O O O O O O ...... O O O O O O O O O O 000000000000 O 13 3

viii

2.
3.
4.

10.

11.

12.

LIST OF TABLES

An expectation by situational informational
interactional model of social judgment .........
Profile of the participants ....................
Descriptive data on experimental variables......
Cell means for negative emotions by
confirmation of expectations and strength of
expectations....................................
Cell means for positive emotions by
confirmation of expectations and strength of
expectations....................................
Cell means for foreign TA evaluation by
confirmation of expectations and strength of
expectations....... ..... . ..... .................
Cell means for foreign TA evaluation on task
competence by confirmation of expectations and
strength of expectations............ ...... ......
Cell means for foreign TA evaluation on
communication competence by confirmation of
expectations and strength of expectations.......
Cell means for subtyping by confirmation of
expectations and strength of expectations.......
Cell means for task, relational, and
communication competence by confirmation of
expectations....... ...... ........ ........ .......
Cell means for dropping the class by
confirmation of expectations and strength of
expectations............. .................... ...
Cell means for taking a class with a foreign TA
in the future by confirmation of expectations
and strength of expectations..... ....... ........

ix

39
68
7O

74

76

80

82

82

85

87

9O

92

LIST OF FIGURES

Page
Language Expectation Model -- Meeting the
foreign TA on the first day of class ...... ...... 28
Language Expectation Model -- No prior language
expectations for foreign TAs..................... 31
Language Expectation Model -- Expectations of
foreign TAs’ language skills .................... 36

Revised Language Expectation Model --
Expectations of foreign TAs' language skills .... 102

CHAPTER ONE

INTRODUCTION

Sugmin is a fairy typical foreign teaching assistant
(TA). He is from Taiwan working on his doctoral degree in
chemistry in a fairly large U.S. university, and his only
source of income is a teaching assistantship from his
department. Sugmin is in charge of conducting lab sessions
for a 200 level undergraduate chemistry class. 0n the first
day of class, when Sugmin walks into class, there is a
strong murmur of disapproval. Students exchange glances and
fidget uneasily in their chairs. A few walk out of the
class.

The reactions of these undergraduates can be summed up
as, "Oh No! not another foreign teaching assistant!" This
reaction is labelled as the "Oh No! S ndrome," and defined
as the shared perception by undergraduate students that
their teacher is unlike other teachers, and may have
significant problems in speaking English. Foreign TAs,
however, constitute an important part of the instructional
faculty at many American universities (e.g., see Bailey,
1984, Bresnahan and Kim, 1991b). Several universities have

a substantial number of foreign TAs teaching a majority of

2
math and sciences courses (Rittenberg, 1992). For example,
Michigan State University (MSU) is ranked 18th nationally in
overall foreign student enrollment. In 1984, the number of
foreign TAs constituted 25 percent of all TAs on campus. In
1991, they constituted a little more than 50 percent of all
TAs. In addition, these foreign TAs were predominantly from
four countries - 117 from the People's Republic of China
(22.85%), 58 from South Korea (11.33%), 49 from India
(9.57%), and 48 from Taiwan (9.38%). It is clear that there
is a strong representation from mainland and Southeast Asia.
It is also interesting to note that the remaining foreign
TAs (240 in number, little less than 50%) are from 62
different countries from all over the world (all these
figures are based on a Michigan State University Report of
the Office of Planning and Budgets, Fall, 1991).

The Oh No! Syndrome is a significant and growing
problem. While the number of foreign TAs appear to be
increasing year after year (Rittenberg, 1992), there is
a slow but increasing body of research indicating the
hardened lack of receptivity to foreign TAs by U.S.
students (e.g., see Bresnahan & Kim, 1991b, Rubin &

Smith, 1990, Wol-Young, 1989). This indicates a need
to conduct more systematic research which helps us
understand these negative perceptions of foreign TAs,
and aids in designing programs for both undergraduates

and foreign TAs to make their interactions within and

3

outside the classroom more fruitful.

Research in the 1980’s focused primarily on the
foreign TAs' difficulties in speaking English, and on issues
such as adjusting to the U.S. and to its class room culture
(Bailey et al., 1984). Most of the research to date
continues in the same vein with the assumption that the lack
of language proficiency of foreign TAs is the primary cause
of the Oh No! Syndrome. Consistent with this assumption,
training programs were initiated in a number of universities
foreign TAs' language skills.

It soon became apparent, however, that training foreign
TAs alone would not suffice. A few researchers explained
that undergraduates may have an equally important role in
creating and maintaining the Oh No! Syndrome. For example,
Bresnahan and Kim (1991a, 1991b) have investigated how
personality traits (e.g., dogmatism, authoritarianism) play
a significant role in influencing students' receptivity to
foreign TAs. In addition, Rubin and Smith (1990) have shown
that undergraduates perceive the foreign TAs' accents to be
stronger than they are (based on standardized tests), and
hence evaluate these TAs negatively. Thus, although the
primary complaint of undergraduates is that they don't
understand their foreign TAs undergraduate students are
turned off and tend to drop classes taught by foreign TAs
(even when the accent is low to moderate). Aligned with

this logic, there now are a few programs (e.g., Michigan

4
State University's "Oh No! to 0.k." program) assisting
undergraduates in communicating more effectively with their
foreign TAs (Rao, 1993).

Although there are many training programs for foreign
TAs, and a few for undergraduates, theoretical foundations
for these programs are still sparse. There is no research
that directly explicates the Oh No! Syndrome, that is,
research which explains the cognitive, affective, and
behavioral processes of foreign TAs and undergraduates as
they interact with each other in the classroom. Research to
date often has been variable analytic in nature, and has
offered limited conceptual understanding of this phenomenon.
A model that explains the cognitive, affective, and
behavioral processes of both undergraduates and foreign TAs
will offer both theoretical focus and clearer direction for
effective training programs.

The purpose of this thesis, therefore, is to offer a
Language Expectation Modal that explores the cognitive,
affective, and behavioral processes of undergraduates when
they encounter their foreign TA in class for the first time.
It is called a Language Expectation Model as it is believed
that the language (vocabulary, grammar, etc,.) and para-
language (accent, style, etc,.) used by the foreign TA
triggers off the processes described in the model. This
thesis focuses only on the undergraduates' perspective as

there are a number of researchers studying this issue as a

5
"foreign TA" problem (Briggs et al., 1990). Further, as a
part of programmatic research in the future, this researcher
eventually hopes to study the cognitive, affective, and
behavioral processes of foreign TAs, as well as the
interaction between undergraduates and foreign TAs within
the class room.

To accomplish its purpose, this thesis is divided into
six chapters. In this chapter (Chapter One), I provided an
overview of the Oh No! Syndrome. In Chapter Two, I review
literature relevant to the Oh No! Syndrome, and argue the
need for defining the construct by offering a model. In
Chapter Three, I map out the Language Expectation Model by
drawing on a variety of theories which suggest links between
the cognitive, affective, and behavioral processes of
undergraduates when they meet their foreign TA for the first
time in class. Specific hypotheses will be proposed from
the explication of the model. In Chapters Four and Five, I
present the methods and the results of a study which tests a
part of the model. Finally, in Chapter Six, I suggest
theoretical and training implications arising from the

study.

CHAPTER TWO

LITERATURE REVIEW

Eleven years ago Bailey (1982) warned researchers that
training programs for foreign TAs were being created without
a systematic investigation of the Oh No! Syndrome. This
appears to be true even today. Research relevant to the Oh
No! Syndrome will be reviewed in three parts. First, I will
demonstrate how different participants in this drama -
undergraduates, their parents, university officials, and
government officials all see the Oh No! Syndrome as
primarily a "language proficiency" issue. In other words,
these members believe that the onus for the problem is
firmly on the shoulders of the foreign TAs, and that they
should improve their language skills. Second, I will show
that while there is some recent (and very useful) research
discussing the role of undergraduates in this equation, it
is limited in scope. Third, I will document that there is
almost no research illustrating how the foreign TAs perceive
the Oh No! Syndrome.

A focus primarily on the foreign TAs’ language skills,
the lack of inclusion of the undergraduates' role, and the
lack of theoretical grounding in creating programs provides

support for a model to explicate the Oh No! Syndrome - the

Language Expectation Model.

Foreign 155' Language Problem
There is no doubt that many of the foreign TAs in U.S.

universities have difficulties in speaking English.
However, it is surprising to see the number of training
programs for improving the foreign TAs’ language skills
based on the assumption that the Oh No! Syndrome is nothing
but a "foreign TA language problem." A recent annotated
critical bibliography on this subject (Briggs et al., 1990)
is comprehensive with 137 citations covering papers,
dissertations, manuals, textbooks, videos, reports, and
presentations (which include surveys, administration and
policy, research, testing, and training). At least 125 of
these citations are oriented towards explaining how and why
the foreign TAs have a language problem, and the majority of
them propose ad hoc solutions for improving their language

skills.

Perception of Some Government Officials

State lawmakers often assume that the Oh No! Syndrome
arises primarily because of language deficiencies of foreign
TAs. Bresnahan and Kim (1991b) report the following
testimony offered to the House Education Committee of the
Ohio State Legislature by a State representative in May,

1985:

8
I first became aware of problems faced by students who
take classes from foreign-born teaching assistants when
my daughter told me about her difficulty understanding
her college math instructor. Although the instructor
has a thorough knowledge of the subject matter, his
limited experience in speaking English made it
difficult for him to convey that knowledge to the
students. Conversations with other college students
confirmed that this is not an isolated problem. In
fact, the problem is not restricted to Ohio. The
University of Pittsburgh granted tuition refunds
recently to students who complained they could not
understand the English spoken by two foreign-born
instructors.
[Proceedings of the House Education "Committee of the
Ohio State Legislature," May 21, 1985. Sponsor-
testimony by Barbara C. Pringle, State Representative

of the 11th House District]

In January, 1988, a bill was passed by the Michigan
State Senate that required a minimum proficiency level in
English for all foreign instructors (see Appendix A).
Although this bill was defeated in the Michigan House of
Representatives, it argued that anyone who taught in
publicly supported higher education should be able to

communicate in a manner that could be comprehended. In

1‘

9
addition, Thomas and Monoson (1989) report that at least
twelve states have language proficiency legislation or
policy aimed mostly at public universities, and nonnative
English speakers in these universities.

Although the bills in the Ohio and Michigan
Legislatures see the Oh No! Syndrome problem primarily as
the language deficiency of foreign TAs, Bresnahan (1990)
argues there are reasons for doubting this view. For
example, she stresses the lack of formal complaints to the
university by undergraduates indicating that the language

proficiency issue may be exaggerated.

Perceptions of University Officials

Most university officials concur with the view of
government officials in focusing on the language
deficiencies of foreign TAs. Without any systematic
research in the area (Bailey, 1982), training programs are
being created in many universities. These programs have
taken a variety of approaches - making the entrance tests in
English for foreign TAs more stringent (Johncock, 1991),
improving the language skills of foreign TAs (Anderson-
Hsieh, 1990; Byrd & Constantinides, 1988; Sequeira &
Constantino, 1989), teaching U.S. classroom culture (Boyd,
1989), and using theater and drama to improve the
pronunciation skills of foreign TAs (Stevens, 1989). Of the

137 citations in the annotated bibliography offered by

10
Briggs et al. (1990), almost a half of them deal with
training programs for foreign TAs. While these programs are
valuable, they are not theoretically based and in most cases
there is little done by way of evaluating them.

Further, university officials note that they are doing
their best to maintain the quality of foreign TAs. In most
universities, foreign TAs have to pass stringent tests at
both the national level (like TOEFL - Test of English as a
Foreign Language), and at the university level (conducted by
English language centers in many universities) before being
admitted into the university. However, most students are
required to have some form of departmental monitoring to
maintain the teaching standards. University officials argue
that individual departments fail to keep up this monitoring
process as it involves significant amounts of time and
effort. In the State News (1989), a large midwestern
university’s student newspaper, university officials
explained that if there is a problem, it has been in
continued departmental monitoring (or the lack of) rather

than in initial screening of candidates.

Pereeption of Parents

Parents play an important role in the interaction
between undergraduates and foreign TAs. Parents provide
tuition and boarding costs for most undergraduates. In

addition, their tax dollars (in part) pay for the salaries

11
of both professors and politicians. Therefore, it is not
surprising when undergraduates and/or their parents offer an
argument as an irate consumer: "I pay for this education and
I have a right to expect teaching assistants who can speak
English fluently. You cannot expect me to work harder to
understand foreign accents." However, as Bresnahan and Kim
(1991b) note, "other than parental anecdotes about the
misery of their kid's math class, systematic evidence about
the attitudes of parents is rarely available" (p. 4). This
clearly stresses the need for programmatic research in this

area .

Perception and Role of Undergraduepee

Several studies (Bailey, 1982, 1983, 1984; Inglis,
1988) suggest that undergraduates’ attribute their
frustration in a foreign TAs class mainly to the foreign
TAs' poor speaking skills. Bailey (1982), in her
dissertation, interviewed 392 students at the University of
California, Los Angeles, who felt that their poor
understanding of the subject matter was due to the foreign
TAs English skills. Two more studies (Bailey, 1983, 1984)
reinforced this view.1 Like Bailey (1982), Inglis (1988)
also discovered that there was a strong positive correlation
between foreign TAs’ poor performance on standardized
speaking tests and undergraduates' negative evaluations of

them (;:.61).1 However, the lack of formal complaints by

12
undergraduates, and Bailey’s (1984) call for systematic
research stresses the need to also understand
undergraduates’ role in this issue.

Many researchers have done just that -- identified
other factors responsible for the Oh No! Syndrome.

Bresnahan (1990), and Bresnahan and Kim (1991a, 1991b) have
conducted several systematic studies to explain the
undergraduates' negative reaction to foreign TAs.

Bresnahan (1990) surveyed attitudes towards foreigners among
350 undergraduates in a large midwestern university. This
survey focused on three areas of student concern -
interpersonal relationships, receptivity to foreign
immigrants, and level of acceptance of foreign TAs. Results
confirmed students' concern about the inability of foreign
TAs to communicate effectively, and also suggested that
these feelings are equally strong regardless of whether they
originate from actual encounters or hearsay complaints of
friends. It is likely, however, that attitudes created by
indirect experience are easier to change than those created
by direct experience (Fazio, 1990).

Further; Bresnahan (1990) indicates that language
attitudes tendwto be enduring and resistant to change:i She
cites Day’s (1982) work who argues that children as young as
3.6 years of age in this study were able to differentiate
high and low prestige dialects of English. Acknowledging

the importance of language proficiency, Bresnahan and Kim

11

13
turned to other areas to provide a more comprehensive view
of the Oh No! Syndrome.
Bresnahan and Kim (1991a, 1991b) attempted to identify
personality characteristics of undergraduates to explain the

’\

Oh No! Syndrome. In their first study, Bresnahan and Kim
(1991a) found a significant relationship between
undergraduates' level of authoritarianism (Altmeyer, 1988),
and their receptivity to foreign TAs (; =-.43). {Altmeyer
(1988) defined authoritarianism as the level of an
individual’s willingness to submit to established authority.
Results suggested that more authoritarian students were less
receptive to foreign TAs from China, Korea, and Taiwan who
generally used an authoritarian style of teaching.\ These
students also indicated that dealing with foreign TAs was
unpleasant and saw the discussion of foreign culture as
wasting their time in class. {

Bresnahan and Kim (1991b) also studied the relationship
between three more individual traits (dogmatism,
individualism, and communal orientation) and receptivity to
foreign TAs. Dogmatism is the extent to which an individual
is open-minded across situations (Troidahl & Powell, 1972).
ngividualism (Triandis et a1. 1988) is the degree to which
an individual is oriented towards his/her own goals
(individualistic) compared to his/her group goals

(collectivistic). Communal orientation (Thompson &

DeHarpport, 1990) is the extent to which individuals attempt

14
to resolve conflict to the mutual benefit of both parties.
Results indicated that undergraduates who were highly
dogmatic (close-minded) and focused on only their own goals
(individualistic and low communal orientation) were less
receptive to foreign TAs. From this Bresnahan and Kim
(1991b) argue that undergraduates also have a role to play

w—--—

in the Oh No! Syndrome.1

.2

‘7‘1Rubin and Smith (1990) offer further support for this
argument. They discovered that undergraduates’ negative
perceptions of foreign TAs were due more to the
undergraduates’ stereotypical attitudes towards foreign TAs
rather than the actual accentedness of the TAs;:)Each
subject (N=92) listened to 4 minutes of a speech presented
as a classroom lecture by an university instructor. A
photograph of the instructor (Caucasian or Oriental) was
displayed while the subjects listened to their speech.
Students perceived the accentedness to be more acute than it
actually was based on standardized accent tests. Also, the
study showed that factors extraneous to the accent, like
course content and instructor ethnicity played a greater
role in creating the negative perceptions of the
undergraduates. From this, Rubin and Smith (1990) argue
that both undergraduates and foreign TAs own a part of the
Oh No! Syndrome, and training programs should be directed at

both groups.

Several other researchers also suggest that the

15
language proficiency of foreign TAs alone does not explain
the Oh No! Syndrome. Instructors’ ethnicity and educational
status (Brown, 1990), perceived low status of teaching in
research institutions, and ethnocentrism of undergraduates
(Constantinides, 1987) also play a significant role in this
issue.

From these studies, it can be concluded that most
undergraduates perceive the Oh No! Syndrome as a problem
with the language proficiency of foreign TAs. However, it
is clear that undergraduates who are highly dogmatic, highly
authoritarian, individualistic, and have low communal
orientation have a low receptivity to foreign TAs. In
addition, it is evident that factors other than accent
(instructor ethnicity, course content) play a significant
role in shaping the negative perceptions of undergraduates

towards foreign TAs.

Summapy of Literature Review

The analysis of viewpoints offered by some state
governments, university officials, undergraduates and their
parents, and the foreign TAs make it apparent that the Oh
No! Syndrome is a problem at many levels. For some of the
state governments, undergraduates, and parents, it is purely
a problem of rectifying the language deficiencies of foreign
TAs. However, research indicates that undergraduates’

traits and expectations based on the foreign TAs

16
accentedness play a significant role in defining their
attitudes towards foreign TAs. Beep foreign TAs and
undergraduates "own" this problem. Hence, undergraduates
also need to be trained cross culturally to tackle the Oh
No! issue. Finally, except for two studies (Jenkins, 1989:
Wol-Young, 1989), there is no research on how foreign TAs
perceive the Oh No! Syndrome and their own role in it.

This overview stresses the need for systematic research
to identify the status of the Oh No! Syndrome at many
levels. The most useful approach would be to involve all
the concerned parties (foreign TAs, undergraduates and their
parents, university officials, and the state government),
identifying their perception of the problem from a variety
of perspectives, and then suggesting changes oriented
towards specific individuals and groups. However, before
reaching this stage, it is important to study the individual
components in a more systematic manner.

This thesis suggests one such approach. The fact that
the foreign TAs’ language skills needs further improvement
is not questioned. There is a dire need, however, to
understand the role of undergraduates’ cognitive, affective,
and behavioral processes during their interaction with
foreign TAs. A Language Expectation Model (LEM) is proposed
to explicate these processes, and offer a clearer picture of
the Oh No! Syndrome. Such an approach has several

advantages.

17

First, it will offer theoretical insights into the
interaction between foreign instructors and their native
students. Most studies so far have focused on only one or
two of the processes; the cognitive or behavioral (e.g.,
Bailey, 1982, Byrd & Constantinides, 1992, Rubin & Smith,
1990). Affective responses of both foreign TAs and
undergraduates largely have been ignored. The strength of
the LEM is in its inclusion of all three processes -
cognitive, affective, and behavioral. Second, the model
also could be used to explain a number of interactions
outside the classroom. With increasing global
interdependence, there are more and more situations where a
"foreign" person provides some sort of service (like the
TAs) in a "native" country -- doctor-patient, business
executives and clients (automobile industry, software
industry, etc,.), diplomatic liaisons, etc,. Research
findings, therefore, from the Oh No! Syndrome may be
generalizable to other situations.

Third, these theoretical insights could be used to
create new training programs for TAs and undergraduates and
modify existing ones. There are a few programs that focus
on improving the language skills and crosscultural awareness
of foreign TAs, and enhancing the crosscultural sensitivity
of undergraduates (e.g., Michigan State University's "Oh No!
to O.K. program). Further, these training programs could be

used to design programs in other areas listed above.

18
Diversity training and multiculturalism are buzz words
today, but they do suggest important ways to improve global
harmony. The LEM offers one way to assist this process.
The LEM, however, is not the only way to explain the Oh No!
Syndrome. Alternate explanations are suggested in the
discussion chapter. It is necessary to test LEM's approach

against rival explanations in the future.

CHAPTER THREE

LANGUAGE EXPECTATION MODEL

Ipppoduction
The Language Expectation Model (LEM) focuses on three

important factors to explicate the Oh No! Syndrome: 1) the
language, and more specifically, the accent of the foreign
TAs, 2) the expectations of the undergraduates, and 3)

process by which the undergraduates’ expectations influence

-their evaluation of foreign TAs' in the class room. The
importance of looking at these issues is now discussed.

Let us recall Sugmin's situation from the introductory
chapter in this thesis. When Sugmin walked into the lab,
students notice that he is from a different country. It is
likely that students’ expectations of foreign TAs' language
skills are evoked at this stage. Further, when Sugmin
starts talking in an accent dissimilar to theirs, it may
confirm their expectations. It is possible that the foreign
TA may not look foreign (TAs from England, Australia, and
other parts of Europe). In these cases, the expectations
are sparked only when the TA starts speaking. Either way,
it appears the foreiginASL.éggent_significantly influences

undergraduates’ perception and evaluation of their foreign

l9

20
TA.

I first realized the important role expectations play
in our lives when Prof. Frank Boster, one of my teachers,
talked about Robert Rosenthal’s work with the so-called
disadvantaged children (Afro-American, Mexican-American, or
any children who live in conditions of poverty). It was
generally believed that these children did poorly in school
because they are members of the disadvantaged group.
Rosenthal and Jacobson (1968) argued that the poor
performance of these students had more to do with the

teachers’ expectations of these students than the students’

”......-

background.;mWhen teachers expect disadvantaged children to
do poorly, Rosenthal and Jacobson (1968) explained, they
communicate this expectation to the children in subtle ways
to influence the actual behavior of these studentgfj In
other words, if my teacher expects me to perform poorly in
class and behaves accordingly, I may perform poorly simply
because of my teachers expectations. This is one example of
the "self-fulfilling prophecy" concept.

Rosenthal and Jacobson’s (1968) study stresses the
significant role expectations play in our evaluation of and
behaviors towards others. Researchers in several other
areas reinforce this idea. For example, the nonverbal
expectancy violations model (Burgoon, 1978; Burgoon & Jones,

1976) suggests that we hold expectations about the nonverbal

behavior of others. Specifically, Burgoon and Aho (1982)

21
treat conversational distance as a message and explain how
confirming or violating conversational distance influences
judgments of source credibility, attraction, etc,. The
nonverbal expectancy violations model begins by assuming
that people have established expectations about the
distances to be maintained during conversations. The model
proposes, contrary to popular opinion, that there may be
positive evaluations in certain situations when expectations
are violated. Burgoon and Aho (1982) explain that positive
evaluations are likely depending on the reward power the
initiator of the violation has, and how extreme the
violation is. For example, if the initiator has higher
status, and deviates by moving closer than the expected
distance, it is likely that s/he will be evaluated more
positively than if s/he had maintained the normal distance
or deviated farther away from the norm. There is a
threshold, however. The initiator of the violation cannot
move too close to the person s/he is speaking to. This
would be a negative violation, and the initiator would be
evaluated negatively. If the initiator of the violation has
low reward value, s/he is likely to be evaluated negatively
if they do not maintain the expected distance. While this
model initially explained only the consequences of violating
conversational distance (Burgoon, 1983, 1985; Burgoon &
Jones, 1976), it was later extended to include a variety of

nonverbal behaviors (for example, immediacy behaviors; see

22
Burgoon & Hale, 1988).

There is considerable research directly related to the
LEM in the area of language attitudes. The term language
attitudes represents a number of empirical studies looking
at the social evaluation of speakers based on various
aspects of language (Giles & Powesland, 1975). Several
studies (see Edwards, 1982; Ryan & Giles, 1982 for reviews)
over the last three decades have shown that accentedness
influences a variety of behaviors including recall of
information about outgroup members (Gill & Badizinski,
1992), negative evaluation of outgroup members in their
ability to succeed, intelligence, and social awareness
(Arthur, Farrar, & Bradford, 1974), categorizing outgroup
members to a lower status (Callan, Gallois, & Forbes, 1983),
and possibility of integration with the majority (Lyczak,
Fu, & Ho, 1976).

Overall, research in this area suggests that the
language we speak and the way we speak it (pronunciation,
accent, grammar, etc,.) influences the way we are evaluated.
Lambert et al. (1960) has shown that eubjective evaluations
of speakers from minority groups are systematically
influenced by stereotypes held about such groups. More
directly, Rubin and Smith (1990) show that the Oh No!
Syndrome is created more by undergraduates’ stereotypical
attitudes towards foreign TAs than the TAs’ perceived

language deficiency.

23

This discussion on the role of expectations and
language attitudes reinforces the need to systematically
study the undergraduates’ cognitive, affective, and
behavioral processes during their interaction with a foreign
TA. Before offering the LEM, langpage expectation is
defined drawing on Burgoon and Walther’s (1990) definition
of expectation, and Giles and Powesland’s (1975) explication
of class-related language standard and context-related

language standard.

Lepgpage Expectations
Burgoon and Walther (1990) explain:

... communication expectancies are cognitions about the
anticipated behavior of specific others, as embedded
within and shaped by the social norms for the
contemporaneous roles, relationships, and context.
While expectancies have associated evaluations and
conative implications, we prefer to reserve the term
"expectancy" for what is predicted to occur rather than
what is desired (p. 236).
For the LEM, we are looking at the undergraduates’
cognitions about the anticipated behaviors of the foreign TA
in the classroom. Further, within the social norm of a
teacher-student relationship, students expect to be taught
by a teacher whose English is easy to understand. These

expectations are not focused on specific foreign TAs, but

24
are stereotype-based expectations with a common reaction to
all foreign TAs (Rubin & Smith, 1990). Hewstone and Brown’s
(1986) definition of a stereotype with three essential
aspects is particularly relevant here:

1. ... individuals are categorized, usually on the
basis of easily identifiable characteristics such
as sex or ethnicity.

2. A set of attributes is ascribed to all (or most)
members of that category. Individuals belonging
to the stereotyped group are assumed to be similar
to each other, and different from other groups, on
this set of attributes.

3. The set of attributes is ascribed to any
individual member of that category. (p.29)

From this definition, we are stereotypical only when we
ascribe attributes to certain groups, and epply it to eyery
member from that group irrespective of individual
differences among these group members. With the Oh No!
Syndrome, the foreign TAs are categorized generally on the
basis of their nationality. Most often, they are from
countries who form the majority of TAs in most campuses in
the U.S. - People’s Republic of China, Korea, Taiwan, and
India (Rittenberg, 1992). In these countries (except
India), students start learning English in their high
school, and rarely find situations to practice their

English. The attributes most likely ascribed to the

25
majority of the foreign TAs are: 1) they speak with a thick
and unintelligible accent, 2) their language skills are
poor, and 3) therefore, they are inept teachers. These
foreign TAs are differentiated from native TAs who are seen
to have proficient skills in English. While there are
foreign TAs from countries where English is widely spoken
(e.g., England, Australia, Jamaica, etc,.), these foreign
TAs constitute the minority and are generally not seen as a
"foreign TA" (Rittenberg, 1992). TAs from Korea, China, and
Taiwan (who tend to be the majority) are often seen as the
"foreign" TA.

There are three aspects of these stereotype-based
language expectations one can study - how they are created,
their manifestation in specific communication behaviors, and
how to change expectancies. The LEM will focus only on how
these expectancies are manifested in undergraduates’
cognitive, affective, and behavioral processes. Future
research will endeavor to study the other two aspects which
are equally important. There is substantial research
analyzing why undergraduates perceive foreign TAs
negatively, and many programs to train foreign TAs to
improve their English skills. The LEM, by focusing on
processual issues, will fill a lacunae in understanding the
Oh No! Syndrome.

Further, the LEM focuses on the undergraduates’

cognitive, affective, and behavioral processes only during

26
their first interaction with their foreign TA. There is
substantial evidence that we form impressions of others very
early in our initial interaction with others. For example,
Schneider, Hastorf, and Ellsworth (1979) suggest there are
six steps in interpersonal perception -- attention, snap
judgment, attribution, trait implications, impression
formation, and prediction of human behavior. They note that
one could form an impression of another within the first few
minutes of an interaction. In addition, Gudykunst (1991)
argues that in most situations, and especially in initial
intercultural encounters, people tend to interpret and
evaluate incoming messages pefope describing them. There
are times, he explains, where people don’t describe at all.
Gudykunst (1991) offers an example where a girl refuses to
maintain eye contact during a conversation. A U.S. person,
he argues, is most likely to evaluate, "She’s lying," before
trying to describe (she did not look me in the eyes when we
talked) and interpret (any number of possible
interpretations from "she is shy" to "she is lying") the
event.

In the specific case of the Oh No! Syndrome, studying
the initial interaction is particularly important as almost
40% of the undergraduates drop from their classes when they
encounter a foreign TA on the first day of class (Rubin &
Smith, 1990). The LEM focuses, therefore, only on the first

day of class when the undergraduates and the foreign TA meet

27

for the first time.

Language Expectation Model

The cognitive, affective, and behavioral processes of
the undergraduates during their interaction with their
foreign TA is mapped out in a flow chart or a decision tree
form. Each of the links in the flow chart is discussed to
its logical conclusion before going to the next possible

route.

Identified as Foreign TA
When the teaching assistant walks into class first day

of the term (or semester), s/he has to be recognized as a
"foreign TA" (See Figure 1). This happens when the TA looks
"foreign," that is, has physical features of a person from a
country outside the U.S. The confirmation of the TA as
foreign happens only when s/he starts speaking with an
accent dissimilar to what the students expect in the U.S.
One of two things happens here.

[4If the students have had experiences with foreign TAs
before in a class room, or heard about them from peers,
specific expectations are evoked. These expectations could

be positive (foreign TAs are good teachers) or negative 2
(they have terrible accents, and I will not learn anything
from this class). As most of the foreign TAs from Korea,

China, and Taiwan do have a thick accent, the expectations

28

Identify as'Foreign TA

 

Language E ectations

 

Strong Weak

gigpge : Language Expectation Model -- Meeting the foreign
TA on the first day of class

29
generally are negative. The LEM focuses only on these
negative expectations.

If the students: (1) have had no prior experience with
a foreign TA as an instructor, (2) the university or their
peers do not inform them that they will encounter a foreign
TA, (3) have limited or no exposure to the media (especially
television, and/or (4) they have had no experience with
people from other countries, undergraduates will have no
language expectations of their foreign TA. They do have
other expectations like they would from any native TA - good
knowledge of the subject, effective and entertaining
communicator, and good interpersonal skills (Shepherd &
Trank, 1986). With increasing number of foreign TAs
teaching classes in the science departments (Rittenberg,
1992), publicity for this issue in most universities, and
undergraduates telling each other about the Oh No! Syndrome,
it is highly probable that there are very few undergraduates
who have no language expectations of their foreign TA.
However, it is still important to map out the processes for
those undergraduates with no expectations.

The two possibilities, where students either have or
don’t have any expectations of their foreign TA are now
explored. The processes for these two groups are very
similar, barring a few exceptions. Again, each route is
explained as the strength of the expectation varies across

the two groups.

30

Ne gpio; Lenguage Expectations for Foreign TAs

If undergraduates do not expect their foreign TA to
have a thick and different accent compared to their own,
current information will play a significant role in their
evaluations of the TA. iAlloy and Tabachnik (1984), in their
Expectation by Situational Framework, argue that social
judgments of others is mediated by the existing
knowledge/expectation of the situation and the incoming
information. So, if there is low expectation in a
situation, current information plays a significant role in
evaluating the situation and the people involved.\\

When undergraduates have no language expectations of
their foreign TA (see Figure 2), the foreign TA’s evaluation
is influenced by undergraduates’ immediate perception of
their TA’s language skills. If the accent is perceived as
understandable, then the undergraduates’ expectations of the
foreign TA is fairly positive (this foreign TA is o.k.). If
undergraduates perceive the accent as severe, the Oh No!
reaction is seen. The Oh No! or o.k. reaction is moderated
by at least three factors (personality traits, course
content; a d peers’ reactions) resulting in one of these
behaviors. Almost 40% of these students will drop from this
class (Rubin & Smith, 1990). Of others who stay, some
complain constantly, and there are a few others who follow

the TA’s accent without trouble. The role of some of the

undergraduates’ personality traits, the course content, and

31

No Prior Expectations for Fore gn TAs! Language Skills

Current Information Pla s a Significant Role

 

Perceive evere

 

 

Perceive
AccTnt Understandable
Acpent
Oh No! o.k.

 

Moderated by:
Personality Traits
Course Content
Peers! Reaction

zigpge : Language Expectation Model -- No prior language
expectations for foreign TAs

32
peers’ reaction are now discussed.

Bresnahan (1990) explains that personality traits like
dogmatism, authoritarianism, and communal orientation
influence undergraduates’ reaction to foreign TAs. With the
Oh No! reaction, if the undergraduates are open-minded, not
authoritarian, and are group-oriented, they are mepe
receptive to the foreign TAs’ severe accent than those who
are close-minded, more authoritarian, and not group-
oriented.

Course content appears to moderate undergraduates’
reactions to their foreign TA’s severe accent (Rubin &
Smith, 1990). If the subject taught by the foreign TA is a
major for the undergraduate, s/he is more likely to pay
attention, generally have a better grasp of the material,

and complain less about the accent. There is, in addition,

 

a motivation to learn as this is the undergraduate’s major.
Non-majors, however, are more likely to be frustrated, and
maybe even drop the class.

The reactions of the undergraduate peers in the class
room is a third and final moderating factor in the students’
reaction to their foreign TA. There is substantial evidence
that our social identity (e.g., Tedeschi & Reiss, 1981) and
group dynamics (Sherif & Sherif, 1966) influences our
decision making. Tedeschi and Ross (1981) explain that we
engage in certain types of self-presentations to gain social

awards, and avoid punishments. From their famous boys’ camp

33
study, Sherif and Sherif (1966) showed that most of us want
to belong to certain group(s) and be accepted by its
members. This may, in addition, make us behave in ways we
would consider wrong in other situations. In a series of
experiments, Asch (1956) asked subjects to judge which of a
set of three lines matched a fourth. The answer was pretty
clear. This experiment was conducted in groups, and in
certain conditions, every member (all confederates) apart
from the subject answered incorrectly. Almost 75% of the
subjects in this condition also answered incorrectly. Asch
(1956) argues that it is extremely difficult to maintain a
deviant view in the face of what appears to be an universal
belief.

On the first day of the class, if there is a general
murmur of disapproval by the undergraduates, it is very
likely to influence other undergraduates who have no
language expectations of their foreign TA. Such
undergraduates (without expectations) may be inclined to
believe that the Oh No! reaction is the appropriate behavior
and may even decide to drop the class. It is also possible,
however, that undergraduates in the class show no
perceptible reaction, allowing the undergraduates without
language expectations of their foreign TA make their own

judgments.

34

Strong versus Weak Prior Expectations for Foreign TAs

Undergraduates may expect their foreign TA to have poor
language skills (in other words a strong accent) because
they have prior personal experience with a foreign TA who
had a strong accent. This personal experience could be
direct, that is, they have interacted with foreigners who
have strong accents, or the experience could be indirect,
that is, they heard about it from their peers. The source
of their expectation, whether it is direct or indirect, is
vital for attitude change. Fazio (1990) argues that
although attitudes based on direct and indirect experience
may be expressed equally strongly, attitudes created by
indirect experience are easier to change. Similarly, the
undergraduates may have a strong or weak language
expectation of their foreign TA. Undergraduates who have
only heard about the foreign TA’s strong accent will have a
weak expectation as it based on indirect experience.
However, undergraduates who have taken classes from foreign
TAs with a strong accent are likely to have strong
expectations. Although the processes for undergraduates
with strong and weak expectations are the same, the
evaluation of the foreign TA varies across these two
conditions. The impact on evaluation is discussed in the
flow-charts.

The expectations of the foreign TAs language skills

(strong or weak) are either confirmed or violated. They are

35
confirmed when the undergraduates perceive the foreign TA’s
accent to be severe. The expectations are violated when the
perceived accent is not as severe as expected. The
undergraduates’ cognitive, affective, and behavioral

processes are different for these two decision routes.

Expectations confirmed. When undergraduates’
expectations of foreign TAs’ language skills are confirmed,
there is an immediate negative affective response of anger,
anxiety, and frustration (see Figure 3). The students are
angry that they will not be able to understand their TA; not
learn anything; and they don’t deserve this as they have
paid to be taught well. Bodenhausen (1993) offers an
excellent analysis of how emotions and arousal interface
with cognitions to influence stereotypic judgments of
outgroup members. lIn an intergroup situation with high
anxiety, like the interaction between undergraduates and
foreign TAs, Bodenhausen (1993) offers substantial evidence
that people are more likely to use heuristic strategies to
evaluate their communicator.: Heuristic strategies are
mental short-cuts people take in making decisions. The
heuristic model of persuasion suggests that, " ... people
exert little cognitive effort in judging the validity of a
persuasive message, and, instead, base their agreement with
a message on a rather superficial (italics included)

assessment of a variety of extrinsic persuasion cues ..."

36

Expectations of Foreign TAs Language
Skills (Strong vs Weak)

 

 

 

Experimental—7’
manipulation
Confi ed vi lated
Negat ve Affective Positive A fective
Retponse ResponTe
Negative Attitudinal Positive Attitudinal
Reaction Reaction
Reinforce Behavioral Sub-t. ing Behavioral
Existing Intention Intention
Expectations (drop class) (stay in class)

 

Moderated by:
Personality Traits

Course Content
Peers' Reactions

zigpge : Language Expectation Model - Expectations of
foreign TAs! Language Skills

37
(Chaiken, 1987, p. 3). Further, in their Elaboration
Likelihood Model, Petty and Cacioppo (1986) argue that a
person’s motivation and ability to process messages
influences whether people systematically evaluate messages
(central route), or use short-cuts (peripheral route).

It is likely that anxious and angry undergraduates may
have the ability to evaluate their foreign TA more
systematically. There is generally little motivation to do
so, however. Undergraduates often assume that it is the
foreign TAs’ responsibility to change their accent to make
themselves understandable (Bresnahan, 1990). That they use
heuristic cues (e.g., severe accent) to evaluate their
foreign TA, therefore, is not very surprising.

Bodenhausen (1993) notes that a key component of the
stereotyping process is the final judgment made by the
member of one group about the member of another group. With
the Oh No! Syndrome, the probable response would be a1
creation or reinforcement of a dislike towards all foreign
TAs, and dropping the class if there is an option to take it
with an American instructor laterf‘ This reaction would be
consistent with other research on ingroup-outgroup
interactions. For example, Wilder (1990) found, not
surprisingly, that ingroup members exerted more influence on
other members within the group than outgroup members. With
the existing ethos that it is alright to drop a class taught

by a foreign TA (Rubin & Smith, 1990), the undergraduates’

38
decision to drop a class is reinforced by other
undergraduates who do the same.

The relationship between the emotional response and the
cognitive evaluation/behavior, however, is not that simple.
Bodenhausen (1993) explains that mediating processes,
individual differences, and situational moderators influence
the extent to which anger and anxiety results in negative
evaluations of the communicator. With the Oh No! Syndrome,
the strength of these negative evaluations differ for
undergraduates with strong or weak expectations.

Alloy and Tabachnik (1984) argue that prior
expectations interact with current information to influence
social judgments of others. Smith (1991) offers an adapted
model of the Expectation by Situational Information
Interaction Framework which is shown in Table 1.
Smith(1991) explains that "the effects of current
information is strong when the data are nonambiguous,
salient, or vivid, and weak when the data are ambiguous or
inconsistent .... the effects of prior expectation will be
strong when a schema is activated at the time of perceptual
processing ..., and weak when no particular schema is
available" (p. 7).

Alloy and Tabachnik’s (1984) model can be adapted to
explain the undergraduates’ processes during the Oh No!
Syndrome. When undergraduates have strong prior

expectations of their foreign TA’s poor language skills, and

39

Table 1
Ap expectapiop by situationel informational intexeepionai
o o soc' 'ud ent
Current Situational Information
WEAK STRONG
Cell 1: Cell 3:
Refrain from social Social judgment
judgment or social highly influenced
LOW judgment made with by current
low confidence situational
information
221—0:
Expectation
Cell 2: Cell 4:
Social judgment Case 1: Both prior
highly influenced by expectation and
prior expectation current situational
HIGH information imply

similar social
judgment -- social
judgment made with
high confidence

Case 2: Prior
expectation and
current situational
information imply
different judgments
-- cognitive
dilemma

40
it is confirmed, they make a judgment with the highest
confidence that their foreign TA has poor language skills
and is an ineffective teacher (similar to Cell 4, Case 1 in
Alloy & Tabachnik’s framework). If undergraduates have weak
prior expectations of their foreign TA’s language skills,
and it is confirmed, their negative judgment of the foreign
TA is strongly influenced by the immediate foreign TA’s
language skills (similar to Cell 3 in Alloy & Tabachnik’s
framework). In both cases, the negative expectation of
foreign TAs’ language skills is reinforced, with
undergraduates in Cell 4 with a stronger negative attitude
towards foreign TAs than undergraduates in Cell 3.

Across the strong and weak conditions, the
undergraduates’ evaluation of their foreign TA is further
moderated by: 1) undergraduates’ traits like dogmatism and
authoritarianism (Bresnahan, 1990); 2) course content (Rubin
& Smith, 1990), and 3) peers’ reactions (Asch, 1956; Sherif
& Sherif, 1966; see discussion from an earlier section).

The impact of course content across strong and weak

expectations is explained in the hypotheses section.

Expectations violated. When a foreign TA violates the
undergraduates language expectations by speaking in an
understandable accent, there is generally a sigh of relief
and happiness. Bodenhausen (1993) explains that like anger

and anxiety, this positive arousal should also result in the

41
use of heuristic strategies to judge foreign TAs. He also
offers evidence to conclude, " ... happiness is likely to
reduce the capacity for systematic, elaborative thought and
to reduce the motivation for such mental activity" (p. 19).
Undergraduates’ reliance on heuristic cues (e.g.,
understandable accent) may result in generally positive
evaluations of the foreign TA. Bodenhausen (1993) offers
some evidence that happiness may result in negative
evaluations of outgroup members. These studies, however,
looked only at the incidental effect of emotion (unrelated
to the context) on evaluations. It is likely that if the
affect is integral to the context (like the undergraduate-
foreign TA interaction), happiness may result in positive
judgments of the foreign TA. The strength of happiness will
depend on whether the undergraduate’s prior expectations are
strong or weak.

Using Alloy and Tabachnik’s (1984) framework, when
undergraduates’ strong expectations of their foreign TA’s
poor language skills are violated, prior expectations should
strongly influence evaluations to sub-type the foreign TA as
an exception. Weber and Crocker (1983) define sub-typing as
a process where "initial knowledge about the group is
represented by superordinate stereotypes in which uniform
trait attributions are made to the entire group. As
discrepant information is acquired, discriminations within

the group are made, leading to the development of subtypes"

42
(p. 962). In other words, all foreign TAs are seen as
having poor language skills. When a foreign TA violates
this stereotype, s/he is seen as an exception and subtyped.
This suggests that the negative expectation of foreign TAs
in general does not change. Undergraduates with weak
expectations, however, are more likely to generalize their
positive affective response to other foreign TAs. In other
words, they are less likely to subtype their foreign TA as
Van exception.

1 Like before, the relationship between undergraduates
happiness and the evaluation of the foreign TA is further
moderated by: 1) undergraduates’ traits like dogmatism,
authoritarianism (Bresnahan, 1990): 2) course content (Rubin
& Smith, 1990), and 3) peers’ reactions (Asch, 1956; Sherif
& Sherif, 1966; see discussion from an earlier section).

The impact of course content across strong and weak
expectations is explained in the hypotheses section.

The arguments offered so far can be concisely
summarized. First, I argued that when undergraduates have
no language expectations of foreign TAs, the language skills
of the immediate foreign TA will play a significant role in
how s/he is evaluated. If the foreign TA’s accent is
perceived as understandable, her/his communication skills is
judged favorably. If it is perceived as unintelligible,
negative evaluations are made. These positive and negative

evaluations of the foreign TA are moderated by the

43
undergraduates’ personality traits (like dogmatism and
authoritarianism) and the course content.

Next, I pointed out how most undergraduates have
expectations that their foreign TA will have inadequate
language skills. These expectations are weak when they are
based on hearsay from others, and strong when they are based
on direct experience with foreign TAs as instructors and/or
based on hearsay from others. When this expectation is
confirmed, undergraduates become angry and anxious. This
arousal influences the undergraduates to use stereotypical
heuristic strategies to negatively evaluate their foreign
TA. When this expectation is violated, the undergraduates
are pleasantly surprised and happy. Again, this results in
the undergraduates using heuristic cues to favorably
evaluate their foreign TAs’ intellectual and communication
skills. The strength of these evaluations are influenced by
the undergraduates’ prior expectations (strong or weak) of
their foreign TAs, and further moderated by situational and

personality factors.

EXPQLngégé
The Language Expectation Model (LEM) offers several

testable hypotheses. In an effort to be parsimonious and
thereby get a better understanding of the processes
involved, only a part of the model is tested in this thesis.

The hypotheses here focus only on undergraduates with

44
language expectations (strong or weak) as it appears that
most undergraduates expect their foreign TAs to have poor
language skills. It is my wish to empirically test other
parts of the model in the future, making modifications based
on feedback given by foreign TAs, undergraduates,
colleagues, and research relevant to this model.

The primary dependent variable in the LEM is the
foreign TA’s evaluation by the undergraduates. Shepherd and
Trank (1986) framework, which explains how teachers are
evaluated within three domains, is modified to measure the
dependent variable. In the first domain, students rate
their teacher’s effectiveness on their fulfillment of the
peex goals. Shepherd and Trank (1986) explain that
"teachers are expected to ‘teach’ a certain body of material
that the students are expected to ‘learn’" (p. 7). In other
words, teachers are expected to be experts in the course
content, and that content should be important intrinsically
or pragmatically. The task goals are often clearly outlined
in the class syllabus.

Secondly, teachers are judged on their fulfillment of
reietional goals. Shepherd and Trank (1986) explain:

Teachers are expected to evidence caring for their

students. Students expect teachers to evidence this

caring in various ways: teachers are to be interested
in what they do -- they are to give the sense that they

want to be there and like their jobs; teachers are to

45

make themselves available to the student outside of the

classroom for individual help; ...: teachers should be

relatively "easy to talk to" and so on: ... (p. 7)
Undergraduates evaluate the extent to which teachers fulfill
these relational goals.

Finally, undergraduates judge their teachers on their
commupicative goals, that is, how well teachers communicate
their expectations of the students in the class; explain
concepts and offer interesting and relevant examples; answer
questions; offer comments to assist students’ learning; and
do it a level undergraduates understand and enjoy (Shepherd
& Trank, 1986).

In their study Shepherd and Trank (1986), using a
constructivist approach to communication (see Delia,

O’Keefe & O’Keefe, 1982 for explanation), argue that
undergraduates who are cognitively less complex tend to
evaluate their teachers similarly on task, relational, and
communicative skills. In this case, cognitive complexity
refers to the number of different constructs the students
have for their teachers. Conversely, undergraduates who are
cognitively more complex evaluate their teachers differently
on three skills. Based on complexity-extremity theory
(Linville, 1982; Linville & Jones, 1980) which explains how
stereotypes influence target evaluations, it is very likely
that the same argument applies to undergraduates evaluating

their foreign TA.

46

Complexity-extremity theory (Linville, 1982; Linville &
Jones, 1980) argues that people have more complex
representations of ingroup members than outgroup members as
we have more experience with ingroups than outgroups.
Further, there is a lesser chance to negatively evaluate an
ingroup member as one has more dimensions for her/him, and
negative dimensions are balanced by positive ones. For
outgroup members, there are fewer dimensions to define them.
When these dimensions are predominantly negative, there is a
greater likelihood for extreme negative evaluations of
outgroup members. With the Oh No! Syndrome, foreign TAs
belong to an "outgroup" as undergraduates have limited
interaction with them. In addition, as undergraduates
appear to have negative stereotype-based expectancies of
their foreign TAs language skills (Bailey, 1982: Bresnahan,
1990), it is likely that the foreign TAs’ evaluations on
task, relational, and communicative skills will be highly
correlated. For example, the undergraduates’ reasoning
could be as follows: "I cannot understand my foreign TA.
S/he has such a strong accent. She knows nothing about the
subject, and does not know how to teach this class." When
students’ expectations are violated, evaluations of a
foreign TA along task, relational, and communicative
competence should not be highly correlated.

It is true, however, that undergraduates’s prior

expectations of their foreign TA’s language skills (strong

47
or weak), the affective response of undergraduates, and
other personality and situational variables influence
undergraduates’ evaluations of their foreign TAs. In Figure
3, a flow chart maps out when the negative expectations of
foreign TAs’ language skills are confirmed or violated.
Several testable hypotheses are now offered, starting from
the top of the chart going down. Further, within each set,
hypotheses move from the simplest to more complex

predictions.

Hypotheses about AffethEmotion

When a teaching assistant walks into class on the first
day of the term, and is labelled a foreign TA because of the
way s/he looks and/or speaks, there is an immediate
affective response. The valence of the affective response,
positive or negative, will depend on whether the foreign TA
subsequently confirms or violates the student’s language
expectation. Students should have a negative affective
reaction (anger and anxiety) when the foreign TA confirms
their expectations, and a positive affective response
(relief and happiness) when the foreign TA violates the
students’ expectations. The affective response, therefore,
is driven by the valence of the expectation.

Hi: Students will have a more negative affective reaction
towards a foreign TA when s/he confirms rather than

violates their language expectations. Specifically:

48

a. students will experience higher levels of
anger and anxiety when the foreign TA
confirms rather than violates their language
expectations.

b. students will experience higher levels of
happiness when the foreign TA violates rather
than confirms their language expectations.

Further, the intensity of students’ affective reaction
should depend upon the strength of their expectation (strong
or weak prior expectations about the foreign TA’s language
skills). In general, students with strong expectations
should have more extreme affective responses than students
with weak expectations. In a related field, Sherif, Sherif,
and Nebergall (1965) argue that people with high levels of
ego-involvement react differently than people with low ego-
involvement. Specifically, people with high ego-involvement
tend to see viewpoints expressed close to their own as
closer than they actually are (assimilation effect), and
viewpoints expressed opposed to the receiver as further away
than it actually is (contrast effect). When students’
expectations are confirmed, they should therefore have a
strong negative affective response and a weak positive
affective response. Again, when students’ expectations are
violated, they should have a strong positive affective
response and a weak negative affective response as it is a

contrast effect in the opposite direction.

49

Confirmation versus violation of expectations will

exert stronger effects on the emotions of students who

possess stronger prior expectations than on those with

weaker prior expectations. Specifically:

a.

when a foreign TA confirms prior
expectations, students with strong prior
expectations will experience higher levels of
anger and anxiety than will students with
weak expectations.

when a foreign TA violates prior
expectations, students with strong prior
expectations will experience lower levels of
anger and anxiety than will students with
weak expectations.

when a foreign TA violates prior ‘
expectations, students with strong prior
expectations will experience higher levels of
happiness and relief than will students with
weak expectations.

when a foreign TA confirms prior
expectations, students with strong prior
expectations will experience lower levels of
happiness and relief than will students with

weak expectations.

50
es out alu t'o f h o

The simplest prediction is that evaluations of the
foreign TA should the consistent with the students’ prior
expectations. Specifically:

3;: Students will have a less favorable evaluation of a
foreign TA when s/he confirms rather than violates
their language expectations.

However, three factors -- strength of prior expectations,

confirmation versus violation of expectation, and course

content -- interact to influence how a foreign TA is
evaluated.

When a foreign TA confirms prior expectations, for
nonmajors, the evaluation of their foreign TA should be
negative. There should be no significant difference,
however, in evaluation between students with strong and weak
expectations. For example, if students who are
communication majors take a class in chemistry, both
students with strong and weak expectations should evaluate
the foreign TA negatively at the same level as their
expertise in the area is limited, and it is likely that
having this knowledge is not crucial to their careers.
However, for chemistry majors, students with strong
expectations should evaluate the foreign TA more negatively
than students with weak expectations. Students with strong
prior expectations perceive/feel that their inability to

comprehend the foreign TA influences their careers more

51

adversely than students with weak prior expectations.

He: Strength of prior expectations, confirmation versus
violation of expectations, and course content (major
versus nonmajor) will interact in their effects on
students’ evaluation of a foreign TA. Specifically,

a. when a foreign TA confirms prior
expectations, students with strong prior
expectations will make more unfavorable
evaluations of the foreign TA than will
students with weak prior expectations,
especially when the TA lectures on material
that is within rather than outside of the
students’ major.

When a foreign TA violates students prior expectations,
majors with strong or weak expectations should evaluate the
foreign TA positively, and at similar levels. For example,
chemistry majors with strong or weak prior expectations
should be equally happy/relieved that they can understand
their foreign TA fairly easily. However, nonmajors with
strong prior expectations should be more happy/relieved than
those with weak expectations. For example, communication
majors taking a class in chemistry with strong expectations
should be more relieved than those with weak expectations.

b. when a foreign TA violates prior expectations,

students with strong prior expectations will make

more favorable evaluations of the foreign TA than

52
will students with weak prior expectations,
especially when the TA lectures on material that
is outside of rather than within the students’

major.

fiypophesis about Subtyping the Foreign TA
These evaluations of a foreign TA are generalized to

other foreign TAs when a foreign TA confirms rather than

violates their language expectations. When the expectation
is violated, the positive evaluation of a foreign TA is
generalized to other foreign TAs only when students have
weak prior expectations. When students have strong prior
expectations, they are more likely to sub-type the specific
foreign TA by arguing that this foreign TA is an exception
by assuming that s/he has language skills atypical of the
whole group (Weber & Crocker, 1983). Consistency theories

(like cognitive dissonance, balance theory) would concur

with this view that people who are highly involved are more

likely to find ways to retain their original views.

3;: Confirmation versus violation of expectations and
strength of expectations (weak versus strong) will
interact in their effects on the extent to which
students subtype their foreign TA, such that:

a. When a foreign TA confirms prior language
expectations, students are likely not to

subtype their foreign TA whether they have

53
strong or weak prior expectations.

b. When a foreign TA violates prior language
expectations, students with strong prior
expectations will be more likely to engage in
sub-typing than will students with weak

expectations.

Hypothesis about the Dimensions of Evaluation
The beginning of the hypothesis section explained how

Shepherd and Trank (1986) differentiated the dimensions of

evaluation. Confirmation or violation of prior expectation

should influence the evaluation of a foreign TA along the
three dimensions of task, communicative, and relational
competence. In general, when a foreign TA confirms prior
expectations, most students may agree that the fereign TA is

competent but lacks communication skills. Thus, when a

foreign TA confirms students prior expectations, students’

evaluations of their foreign TA should be more favorable
along the task dimension than for the relational and
communicative dimensions.

HQ: Confirmation versus violation of prior expectations
should influence the effects on evaluations of a
foreign TA along three dimensions (task, communicative,
and relational competence). Specifically,

a. when a foreign TA confirms prior

expectations, judgments of task competence

54
should be significantly more favorable than
judgments of communicative and relational
competence.
b. when a foreign TA violates prior
expectations, judgments about the foreign TA
should not differ across the three

dimensions.

Hypeppeses about Behavioral Intentions

In general, it is more likely that students will drop a
course if the foreign TA confirms rather than violates their
expectations.

H1: Students will be more likely to intend to drop a course
from a foreign TA when s/he confirms rather than
violates their language expectations.

However, confirmation versus violation should interact with

strength of expectations (weak versus strong) to influence

behavioral intentions. When expectations are confirmed,
students with strong expectations are more likely to drop
the class taught by a foreign TA than students with weak
expectations. When expectations are violated, students with
strong expectations are less likely to drop the course than
students with weak expectations.

He: Confirmation versus violation of expectations should
interact with strength of expectations to influence

students’ intentions to drop the course. Specifically:

55

a. When expectations are confirmed,
students with strong prior expectations
are more likely to drop the course than
students with weak expectations.

b. When expectations are violated, students
with strong prior expectations are less
likely to drop the course than students
with weak expectations.

Confirmation versus violation of expectations also
should interact with strength of expectations to influence
whether students will take a class with a foreign TA in the
future. When a foreign TA confirms prior expectations,
students with strong or weak expectations are likely not to
take a class with a foreign TA in the future. When
expectations are violated, however, students with strong
expectations are likely not to take a class with a foreign
TA than students with weak expectations because the former
students are more likely to subtype their current foreign TA
as unusual and unlike other foreign TAs.

H2: Confirmation or violation of expectations and strength
of expectations (weak or strong) will interact in their
effects on whether students will take a class with a
foreign TA in the future. Specifically:

a. When a foreign TA confirms prior language
expectations, students are likely not to take

a class with a foreign TA in the future,

56
whether they have strong or weak
expectations.
b. When a foreign TA violates prior language
expectations, students with strong prior
expectations are more likely not to take a
class with a foreign TA than students with
weak expectations.
The nine hypotheses reflect the LEM as we move from the
top to the bottom. They are not, however, exhaustive.
Through empirical research, and alternate perspectives,

other factors and issues need to be addressed.

CHAPTER FOUR

METHODS

W

This study was conducted in two phases. In Phase 1,
the degree of accentedness and undergraduates’ expectations
of foreign TA’s language skills were pretested. One hundred
and twenty four pretest participants watched one of two
videotapes to determine two types of accent -- one that
participants perceive as difficult to understand, and the
other as easy to follow. In the first run (N=60), all the
participants perceived the accent of both foreign TAs as
difficult to follow; they saw no differences between the two
foreign TAs’ accents. In the second run (N=64), students
did perceive one of the foreign TAs as easy to follow, and
the other difficult as to understand. In Phase 2, 330
experimental participants watched a three to four minute
videotape of a foreign TA on the first day of class, and
evaluated him on task, relational, and communicative goals.
Relevant statistical analyses were used to explain the
relationship between undergraduates’ expectations of their

foreign TAs and the evaluations of foreign TAs.

57

58

Eheee i; gxetest for Stimulps Materials

HesearcH participants. A total of 124 (60 in the first
run and 64 in the second run) students from a large
southwestern university participated in this phase. They
were told that this was an exercise designed to learn
students’ perceptions of instructors on campus. The
participants were offered extra credit for volunteering

their time and responses.

Eyoceduxes. In Round 1, two classes were chosen for
the pretest. The procedure was the same in both classrooms.
A research assistant (White Caucasian Male) walked into the
classroom at the beginning of class, and thanked the
participants for participating in the research. A White
male was chosen to collect data as the author is a foreign
TA himself, and was anxious not to bias students’ responses.
After signing the consent forms, the participants were told
that they would answer a questionnaire, watch a videotape of
an instructor for a minute, and answer a questionnaire again
to complete the experiment. In classroom 1 (N=28),
partidipants watched a videotape of a foreign TA with a
severe accent which is difficult to understand. In
classroom 2 (N=32), participants watched a videotape of a
foreign TA with an accent easy to follow.

The whole process (questionnaire - videotape -

questionnaire) took about 15 minutes. After collecting both

59
questionnaires, the participants were debriefed, thanked
again for their participation, and allowed to leave. During
debriefing, it was explained that this was a pretest to see
whether participants perceived the foreign TA’s accent to be
difficult or easy to follow. Participants were also
requested not to share their experiences with other students
as the researcher was planning to conduct similar research

in the near future.

Stimulus materials. As most of the foreign TAs are
from Asian countries, the stimulus for the pretest
(videotape) was created by two Chinese male TAs, one whose
accent was easy to follow and the other with a thick accent
difficult to understand. The names of these TAs were
suggested by staff from the International Teaching
Assistants Training Center on Campus. After conversing with
each TA for a few minutes, they appeared appropriate for the
task. A script was created to simulate the first day of
class in an interpersonal communication class. It included
some personal information about the TA, and a brief
introduction to the class and the syllabus (see Appendix B).
Both TAs delivered their script like an instructor would on
the first day of class (looking at their notes
occasionally), and took about a minute to complete the
narration. To maintain consistency, these two samples were

videotaped in the same classroom, and around the same time

vi

th

th

th

fr

an

pa

st

CO.

So

111'

1‘81

Pa]

30(

SEE

COr

C0!-

the

be,

60
in the evening (6 P.M.). There were no students in the

class, and the TAs were videotaped from waist up only.

Heesurement instruments. Before watching the
videotape, participants answered a questionnaire about: (a)
their demographic information, (b) the number of classes
they have had from a foreign TA, (c) questions on whether
they had heard anything about foreign TAs from roommates or
friends, and if so, what the participants learned from them,
and (d) four questions to measure how the participants saw
themselves as students. After watching the videotape,
participants responded to: (a) four questions to see if
students’ expectations of foreign TAs language skills were
confirmed or violated, and (b) Troidahl’s (1972) Dogmatism
Scale. All questions were on a 5-point Likert-type scale,
with 1 being the weakest reaction and 5 being the strongest

reaction.

Hesults. Prior to viewing the videotapes, the 60
participants expected their foreign TA to speak with an
accent fairly difficult to follow (H=3.8: §Q=.88). After
seeing the videotapes, a comparison of the cell means for
confirmation of expectations (with 5 being expectations
confirmed and 1 being expectations violated) suggested that
there was no significant difference, p(58)=.88; p>.05,

between the confirmed (Classroom 1; H=3.4) and violated

61
(Classroom 2; H=3.6) conditions. In other words, data from
the two classes indicated that participants did not perceive
the two foreign TAs as differing in terms of confirmation of
expectations.

In Round 2, one of the two foreign TAs was replaced
with another Chinese TA who has lived in the U.S. for the
past 20 years. His accent was more comparable to the
"American" accent. This TA was filmed with the same script
in the same classroom, and around the same time in the
evening (6 P.M.). Using the same procedures and
measurements listed above, the pretest was conducted again
in two new classes.

Prior to viewing the videotapes, the 64 participants in
Round 2 also expected their foreign TA to speak with an
accent fairly difficult to understand (H=3.7). As
anticipated, after viewing the videotapes, a comparison of
the cell means for confirmation of expectations (with 5
being expectations confirmed and 1 being expectations
violated) suggested that there was a significant difference,
p(62)=4.8; p<.05, between the confirmed (Classroom 1; H=3.4)
and the violated (Classroom 2: H=1.1) conditions. In other
words, in Classroom 1, as expected, the foreign TA was
perceived as difficult to follow, while in Classroom 2, the

foreign TA was perceived as easy to follow.

62

Egese z; Expeximentel Heasures and Procedures
R§§§8IQD pegticipants. A total of 330 students (from

15 different classes) in a large southwestern university
participated in this phase. They were told that this was an
exercise to learn students’ perceptions of instructors on
campus. Only 2 of the 15 instructors offered extra credit
for participating. Preliminary analyses revealed no
systematic differences between responses from students who
received extra credit and those who did not on any of the
dependent variables; hence responses have been combined in

all analyses.

Exoceduxe. The procedures, which were similar to those
used in the pretest, were the same in all 15 classrooms.
Three research assistants (one White male and two White
females) collected the data. All three research assistants
went through a one hour training session where the data
collection process was simulated. A step-by-step
instruction sheet provided to all three assistants discussed
ways to handle possible difficulties.

In each class, the research assistant began by thanking
the students for participating in the research. After
signing the consent forms, the participants were told that
they would answer a questionnaire, watch a videotape of an
instructor for three to five minutes, and answer a

questionnaire again to complete the experiment. If students

63
had participated in the same experiment in another class,
they were requested not to participate again. The
participants watched one of four videotapes: (a) the foreign
TA with an accent difficult to follow teaching a chemistry
lab, (b) the foreign TA with an accent difficult to follow
teaching an interpersonal communication class, (c) the
foreign TA with an accent easy to follow teaching a
chemistry lab, and (d) the foreign TA with an accent easy to
follow teaching an interpersonal communication class.
Procedures for creating these four tapes are described in a
later section.

The whole process (questionnaire - videotape -
questionnaire) took about 25 minutes. After collecting both
questionnaires, the participants were debriefed, thanked
again for their participation, and allowed to leave. The
debriefing process explained that this was an experiment to
see how participants’ language expectations of foreign TAs
influenced their judgments of foreign TAs when these
expectations were confirmed or violated. The research
assistants answered questions that they were trained for,
and for questions they could not answer, they provided the
author’s phone number. Participants also were requested not
to share their experiences with other students as the

researcher was collecting data in other classes.

64

Independent variables. The independent variables in
this study are: (a) students’ prior language expectations
(strong or weak): (b) expectation confirmed or violated
(severe or weak accent); (c) Lecture topic --chemistry or
communication (major versus non major).

Students’ prior expectations were measured by a set of
six questions, with two questions each on expectations of
foreign TA’s communication competence, relational
competence, and task competence. Further, students’ strong
versus weak expectations of foreign TA’s language skills
were created by using a median split (strong expectations =
3.5 or greater on the 5-point scale, and weak expectations =
less than 3.5 on the 5-point scale) on the items measuring
expectations.

Confirmation versus violation of expectations was
operationalized by videotaping one foreign TA whose accent
is difficult to follow and another foreign TA whose accent
is easy to comprehend. The same two foreign TAs from the
pretest were used for this part of the study. The scene in
the tape was a simulation of the first day of class where
the TA introduces himself, discusses the syllabus, and
offers a brief introduction to the subject matter. Like the
pretest, the videotaping was done in the same classroom, and
at the same time in the evening (6 P.M.).

The lecture topic was operationalized by having each

foreign TA teach both the chemistry lab and the

65
interpersonal communication class. The script for the two
classes had a similar structure (personal introduction,
discussion of the syllabus, and a brief introduction to the
topic). The script for the Chemistry Lab (3 minutes; see
Appendix C) was slightly shorter than the script for the
interpersonal communication class (4 minutes; see
Appendix D). Both TAs delivered their script like an
instructor would on the first day of class, by looking at

their notes and at the class intermittently.

erendent variables. After watching the video tape
(about three to four minutes), students answered: (a) a
modified version of Watson, Clark, and Tellegen’s (1988)
positive and negative affect PANAS scales [a composite score
was created for positive (6 items) and negative (6 items)
emotions]; (b) a modified version of Shepherd and Trank’s
(1986) scales to evaluate instructors on task, relational,
and communicative goals: and (c) a set of questions to
assess whether students engaged in sub-typing (Weber &
Crocker, 1983), whether they would drop a class taught by
the foreign TA on videotape, and whether they would be
likely to take classes with foreign TAs in the future.
Students also answered: (a) four questions as a manipulation
check to measure if the participants’ expectations were
confirmed or violated, and (b) ten comprehension questions

to determine if they recalled what the foreign TA talked

66
about. All the questions were in a 5-point Likert-type
format (with 1 as low and 5 as high), except the
comprehension questions where the students respond true or
false to a set of statements (see Appendix E for pre and

post video questionnaires).

CHAPTER FIVE

RESULTS

Ovexyiew

The results section begins with a general description
of the profile of the participants, and descriptive data on
the experimental variables. This is followed by explaining
the results for each of the nine hypotheses. All the
hypotheses were analyzed using factorial analysis of
variance (ANOVA), except hypothesis 6 which was analyzed
using repeated measures analysis of variance.
Enegile of the participants

Table 2 summarizes important characteristics of the
participants. The total number of participants was 330, of
which 192 were female and 137 male. The average age (in
years) was 25. The over representation of females and the
higher than normal average age is comparable to the
Department’s male/female ratio (40/60) and the average age
of the students (27). There is a larger representation of
juniors and seniors (95 and 146) compared to the freshmen
and sophomores (21 and 67).

Based on an earlier survey, it appeared that it would

be possible to have an equal number of communication and

67

Table 2

Prefiie of the participants

 

 

 

 

 

 

Variable Number
Topic watched on videotape Communication 173
Chemistry 157
Sex Female 192
Male 137
No Response 1
Age (average, in years) 25
Level Freshmen 21
Sophomore 67
Junior 95
Senior 146
No Response 1
Ethnicity Hispanic 78
Caucasian 197
African American 8
Asian 10
Native American 15
Other 20
No Response 2
Participants’ Major Communication 107
Natural Sciences 42
Other 176
No Response 8

 

Note: N=330

69
natural science majors. This, however, was not reflected
inthe final sample. Of the 330 participants, 107 were
communication majors and only 42 were natural science
majors. As this is an important factor for a few
hypotheses, only data from these 149 students will be
analyzed for hypotheses involving type of major.

Table 3 summarizes descriptive data (mean, standard
deviation, and reliability) of all the experimental
variables. All items are 5-point Likert-type questions with
5 indicating stronger responses to the question. The
reliability of the items were fairly high, ranging from .74
to .89.

Henipnietion Check

A manipulation check was performed to determine if
students’ expectations of foreign TAs’ language skills were
negative, and that these expectations were either confirmed
or violated by the two foreign TAs in the study. Consistent
with the predictions, students had a fairly strong negative
expectation of their foreign TA’s language skills before
viewing the videotape (H=3.5). After viewing the videotape,
the cell means for the confirmation of expectations
(5=confirmed; 1=violated) indicated that there was a
significant difference between the confirmed (H=4.01) and
the violated (H=1.80) conditions, F(1,327)=525.46; p<.001.
Confirmation of expectations explained a significant portion

of the variance (eta-squared=.62).

70

Table 3

Qeeezippiye data on experimental variables

 

Variable Mean SD Reliability

Language Expectation

of foreign TAs 3.31 1.05 .74
(2)

Negative Emotions 2.25 1.03 .88
(6)

Positive Emotions 2.09 .85 .86
(6)

Confirmed Expectations 2.84 1.54 .93
(2)

Violated Expectations 3.01 1.35 .90

(2)

Foreign TA’s Task
Competence 3.18 .78 .77

(5)

Foreign TA’s
Relational Competence 2.98 .82 .75

(5)

Foreign TA’s
Communication Competence 3.19 .99 .84

(5)

Foreign TA’s Competence

on All Three Dimensions 3.12 .73 .89
(16)

Drop this Class 2.67 1.33 --
(1)

Subtyping 3.17 .98 --

(1)

Classes with Foreign TAs
in the Future 2.95 1.25 --

(1)

 

Note: N=330; SD=Standard Deviation: Reliability=Alpha;
Numbers within parentheses indicate the number of
questions measuring the variable.

71
Hypotheses about AffethEmotion
o sis

The first hypothesis predicted that students will have
a more negative affective response when the foreign TA
confirms rather than violates their language expectations.
There were two parts to Hypothesis 1. Hypothesis ia
predicted that students will experience higher levels of
anger when expectations are confirmed rather than violated.
To test this hypothesis, a oneway analysis of variance,
analyzing the effects of confirmed versus violated
expectations on negative emotions was performed. Results
suggested that the data were consistent with the hypothesis.
Students’ level of anger was higher when the foreign TA
confirmed (H=2.62) rather than violated (H=1.88) prior
expectations, F(1,312)=45.24; p<.001. Confirmation of
expectations, however, explained only a moderate portion of

the variance on the levels of anger (eta-squared=.13).

Hypothesis 1b predicted that students will experience
higher levels of happiness when the foreign TA violates
rather than confirms students’ language expectations. To
test this hypothesis, a oneway analysis of variance which
analyzed the main effects of confirmed versus violated
expectations on positive emotions was performed. Results

suggest that the data are not consistent with the

72
hypothesis. Levels of happiness did not differ when the
foreign TA confirmed (H=2.12) or violated (H=2.06) prior
expectations, E(1,313)=.30; p>.05.
Overall, the results indicated that while confirmation
versus violation of expectations exerted main effects on
negative emotions, there were no main effects for positive

emotions.

EMMA;

The second hypothesis predicted that expectations
(confirmed versus violated) should interact with strength of
expectations (strong versus weak) to influence emotions
(positive and negative). The strength of expectations
measure was created by a median-split on an aggregate score
for prior expectations. Based on the distribution of scores
for prior expectations (mean=3.31; median=3.5), students
whose expectations were 3.5 or higher were grouped under
"strong expectations," and students whose expectations were
less than 3.5 were grouped under "weak expectations." Based
on this grouping, there were 174 students with strong
expectations, and 150 with weak expectations.

There are four parts to the second hypothesis.
Hypopnesis 2e and 2b, dealing with negative emotions,
predicted that when a foreign TA confirms prior
expectations, students with strong expectations will feel

higher levels of anger and anxiety than students with weak

73
expectations. Further, when a foreign TA violates prior
expectations, students with strong expectations should
experience lower levels of anger and anxiety than students
with weak expectations. To test this hypothesis, an
analysis of variance which analyzed the effects of
expectations and strength of expectations on negative
emotions was performed.

The data were consistent with hypothesis 2a and 2b.

The analysis of variance analysis indicated that both
confirmation of expectations, H(1,304)=47.18; p<.001, and
strength of expectations, H(1,304)=5.38; p<.03, exerted
significant main effects on negative emotions. Further,
there is a significant interaction between confirmation of
expectations and strength of expectations, H(1,304)=7.49;
p<.01. As interaction effects can override main effects,
the interaction effect is interpreted first.

The cell means for negative emotions indicate that when
expectations are confirmed, students with stronger
expectations have stronger negative emotions than those with
weak expectations (see Table 4). Further, when expectations
are violated, students with stronger expectations have lower
levels of anger and anxiety than those with weak
expectations.

To reconfirm these conclusions, contrasts were fitted
specifically to test hypothesis 2a and 2b (see Table 4).

Results indicated that the data were consistent with the

74
predicted model, H(1,304)=51.67; p<.05, and the residual
explained variance was not statistically significant,

2(2,304)=2.39; p>.05.

Table 4

Qell neens for negative emotions py confinnetion pf
expecpations and strength of expectations

 

 

Strength of expectations

 

 

 

Strong Weak
Expectations
Violated 1.86 1.91
(91) (64)
—2 -1
Confirmed 2.87 2.32
(79) (74)
2 1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test the hypothesis specifically.

Hypothesis 2c and 2d, dealing with positive emotions,
predicted that when a foreign TA confirms prior
expectations, students with strong expectations will feel
lower levels of happiness and relief than those students
with weak expectations. Further, when a foreign TA violates
prior expectations, students with strong expectations should
experience higher levels of happiness and relief than those
students with weak expectations. To test this hypothesis, a

two-way analysis of variance exploring the effects of

75
expectations and strength of expectations on positive
emotions was performed.

Results indicated that the data were not consistent
with the hypothesis. While there was a main effect for
strength of confirmation of expectations, H(1,305)=13.76:
p<.001, there was no main effect for expectations,
£(1,305)=.16; p>.05. There was a significant interaction
between confirmation of expectations and strength of
expectations for positive emotions, H(1,305)=10.63; p<.01.
Although there is a significant interaction, the cell means
(Table 5) clearly indicate that the data are not consistent
with the predicted interaction for students with weak
expectations. For students with strong expectations, the
confirmation versus violation manipulation did significantly
affect levels of positive emotions, H(1,168)=4.46; p<.04.
For students with weak expectations, those who heard the
foreign TA confirm expectations unexpectedly reported nexe
positive responses than those who heard the foreign TA
violate expectations, H(1,137)=5.80; p<.02.

Overall, like Hypothesis 1, the data are consistent
with the hypothesis for negative emotions and not for
positive emotions. Indded, the most positive response comes
from students with weak expectations when a foreign TA

confirms their expectations.

76

Table 5

Qeii neans fog positive emotions by confirmation of
expectations and strength of expectations

 

Strength of expectations

 

 

Strong Weak
Expectations
Violated 2.05 2.09
(91) (65)
Confirmed 1.81 2.47
(79) (74)

 

Note: Numbers in parentheses reflect number of students.

Hypotheses about Evaluation of the Foreign TA

Hypothesis 3

Hypothesis 3 predicted that the foreign TA is likely to
be evaluated more negatively if s/he confirms rather than
violates students expectations. To test this hypothesis a
oneway analysis of variance, which analyzed the main effects
of confirmed versus violated expectations on evaluations of
the foreign TA was performed. A composite measure of
evaluation, including all three dimensions of task,
relational, and communication competence, was used.
The results were consistent with the hypothesis. The
analysis of variance indicated that there was a significant
main effect for confirmation of expectations,
£(1,313)=87.15: p<.001, such that students whose

expectations were confirmed evaluated the foreign TA less

77
favorably (H=2.78) than students whose expectations were
violated (H=3.45). Confirmation of expectation explains a
moderate portion of the variance (eta-squared=.21).

A secondary analysis was performed to analyze the main
effects of confirmed versus violated expectations on
evaluations of the foreign TA on the three dimensions of
competence (task, relational, and communication) separately
(A factor analysis indicated that the three dimensions of
competence were not unidimensional). The analysis of
variance indicated that there was a significant main effect
for confirmation of expectations such that students whose
expectations were confirmed evaluated the foreign TA less
favorably on task competence, H(1,318)=10.73; p<.002; eta-
squared=.03, and communication competence, H(1,318)=357.78,
p<.001; eta-squared=.53, than students whose expectations
were violated. There were no significant effects, however,
for confirmation of expectations on relational competence,
H(1,318)=1.49; p>.05: eta-squared=.01.

Overall, students evaluate the foreign TA less
favorably when their expectations are confirmed rather than
when violated. They evaluate the foreign TA less favorably
on task and communication competence when their expectations
are confirmed rather than when violated. Specifically, they
evaluate the foreign TA significantly more positively when
their expectations are violated than confirmed. Students,

however, evaluate the foreign TA similarly on relational

78
competence whether the foreign TA confirms or violates their

expectations.

Megs—4.

Hypothesis 4 predicted that expectations (confirmed
versus violated), strength of expectations (strong versus
weak), and course content (major versus nonmajor) should
interact in their effects on students’ evaluations of their
foreign TA. There were two parts to the fourth hypothesis.

Hypothesis 4a suggested that when a foreign TA confirms
prior expectations, students with strong expectations will
make less favorable evaluations of their foreign TA than
students with weak expectations, especially if the foreign
TA lectures on material that is within rather than outside
the students’ major. To test this hypothesis, a.three-way
analysis of variance crossing the factors of strength of
expectations, confirmation of expectations, and course
content was performed. The dependent measure was a
composite measure of the three dimensions of foreign TA
competence -- task, relational, and communicative
competence. Course content was operationalized by matching
the student’s major with the topic they listened to on the
video tape. Natural science majors who watched the
chemistry lecture, and communication majors who watched the
communication lecture were grouped under "Match."

Conversely, natural science majors who watched the

79
communication lecture, and communication majors who watched
the chemistry lecture were grouped under "Mismatch." Hence,
only a subset of the entire sample was included in this
analysis (N=149).

Results from the analysis of variance suggested that
the data were not consistent with the hypothesis. There was
a significant main effect for confirmation of expectations,
[(1,131)=49.56; p<.001, and strength of expectations,
£(1,131)=7.93; p<.01. The main effect for course content was
not significant, H(1,131)=.17; p>.05, nor were the
confirmation of expectation by course content, E(1,131)=.l9;
p>.05, or strength of expectation by course content
interactions, H(1,131)=2.59; p>.05. The 3—way interaction
for strength of expectations, confirmation of expectations,
and course content was also not significant, H(l,131)=.91;
p>.05. The confirmation of expectation by strength of
expectation interaction, however, was significant,
3(1,131)=5.68; p<.02.

Since neither the main effect nor the interactions
involving course content were significant, this hypothesis
was reanalyzed using the whole sample of 330. It was
predicted that when a foreign TA confirms prior
expectations, students with strong expectations should
evaluate the foreign TA less favorably than students with
strong expectations. When a foreign TA violates prior

expectations, students with strong expectations should make

80
more favorable evaluations of the foreign TA than students
with weak expectations.

The results for the entire sample were consistent with
this part of the hypothesis. There were significant main
effects for strength of expectations, H(1,306)=12.36:
p<.002, and confirmation of expectations, E(1,306)=90.16;
p<.001. The 2-way interaction of strength of expectation by
expectation on evaluation was statistically significant,
H(l,306)=4.32; p<.04. Cell means are shown in Table 6. A
contrast fitted to test the revised hypothesis indicated
that the data are consistent with the hypothesis,
£(1,306)=91.25: p<.05, but the residual explained variance

was also significant, H(2,306)=6.21: p<.05.

Table 6

Qell neens for foreign TA evaluation by confirmation gf
expectations and strength of expectations

 

Strength of expectations

 

 

 

Strong Weak
Expectations
Violated 3.41 3.51
(90) (68)
2 1
Confirmed 2.59 3.00
(77) (75)
-2 -1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test hypothesis specifically.

81

A secondary analysis was performed to analyze the
effects of confirmation of expectation and strength of
expectations on the three dimensions of competence (task,
relational, and communication) separately. The 2-way
interaction of strength of expectation by confirmation of
expectation on evaluation was statistically significant for
communication competence, H(1,311)=7.51: p<.05, and not
significant for task competence, H(1,311)=.85; p>.05, and
relational competence, H(1,311)=3.03; p>.05. When contrasts
were fitted to test the hypothesis directly for task and
relational competence (similar to the ones used in Table 6),
the data were consistent with the hypothesis for task
competence, e(311)=3.46; p<.002, and not for relational
competence, e(310)=1.68; p>.05.

Although the contrasts indicated that the data were
consistent with Hypothesis 4, a study of the cell means for
the effects of confirmation of expectations and strength of
expectations on task and communication competence suggested
that the mean differences between strong and weak
expectations is about the same for the violated condition.
Contrasts were fitted to test this alternate model for task
(Table 7) and communication competence (Table 8). Results
indicated that the alternate model explained more variance
(sum of squares for task competence=9.36; sum of squares for
communication competence=164.43) than the original model

(sum of squares for task competence=6.60; sum of squares for

82

Table 7

Hell means for foxeign TA evaluation on tesx competence by
eonfinnation of expectations and strength of expectations

 

Strength of expectations

 

 

 

Strong Weak
Expectations
Violated 3.25 3.44
(92) (68)
2 2
Confirmed 2.89 3.24
(77) (78)
-3 -1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test hypothesis specifically.

Table 8

Hell means for foreign TA evaluation on communication
competence bv confirmation of expectations and strengtn of
W

 

 

 

Strength of expectations

 

 

 

Strong Weak
Expectations
Violated 3.89 3.91
(93) (69)
2 2
Confirmed 2.25 2.69
(77) (76)
-3 -1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test hypothesis specifically.

83
communication competence=157.95).

Overall, course content does not appear to have any
significant impact on the evaluation of the foreign TA. As
this decision is based on fairly small sample sizes, it
should be accepted cautiously. Results initially indicated
that when expectations were confirmed, students with strong
expectations did evaluate the foreign TA less favorably than
students with weak expectations. Similarly, when
expectations were violated, students with strong
expectations evaluated the foreign TA more favorably than
students with weak expectations. A secondary analysis
suggested that there was no significant interaction between
confirmation of expectation and strength of expectation on
relational competence. For task and communication
competence, however, when expectations were confirmed,
students with strong expectations evaluated the foreign TA
less favorably than students with weak expectations. When
expectations were violated, there were no significant
differences between students with strong and weak
expectations in their evaluation of the foreign TA on task
and relational competence. Evaluation of foreign TA on
communication competence differed significantly when
expectations were violated. In other words, Hypothesis 4
was supported fully when the foreign TA confirmed
expectations, and partially supported when the foreign TA

violated prior expectations.

84

Hypothesis about Subtyping the Foreign TA
We;

Hypothesis 5 predicted that strength of expectations
(strong versus weak) and expectations (confirmed versus
violated) should interact to affect the extent to which
students subtype their foreign TA. Subtyping occurs when
students see a foreign TA as an exception: "This foreign TA
is not like other foreign TAs." In this study, it is
defined specifically that for situations when a foreign TA
violates exceptions, s/he is seen as exception. Hypothesis
5 has two parts that were tested with an analysis of
variance for investigating the effects of strength of
expectations and confirmation of expectations on subtyping.

Hypotnesis 5a predicted that when a foreign TA confirms
expectations, students with weak and strong expectations are
equally likely nee to subtype their foreign TA. Hypotnesis
en predicted that when a foreign TA violates expectations,
students with strong expectations are more likely to subtype
than students with weak expectations. Results indicated
that the data were not consistent with the hypothesis. Both
the main effects for confirmation of expectations,
F(1,308)=1.46: p>.05, and for strength of expectations,
H(1,308)=.48; p>.05, were not statistically significant.

The predicted 2-way interaction between expectations and
strength of expectations also was not statistically

significant, H(1,308)=2.88: p>.05. As ANOVA picks up only

85

cross-over interactions, specific contrasts were fitted to
test the hypothesis. Cell means are shown in Table 9, and
results indicated that the data still were not consistent
with the hypothesis, p(308)=1.34; p>.05. 1

Overall, the results suggested that students’
likelihood of subtyping their foreign TA will not be
affected by either whether the foreign TA confirmed or
violated their expectations, or whether students had strong

or weak expectations.

Table 9

Cell neans for subtyping by confirmation of expectations and
stpength of expectations

 

Strength of expectations

 

 

 

Strong 1 Weak
Expectations
Violated 2.95 2.84
(92) (67)
2 0
Confirmed 2.64 2.91
(78) (75)
-1 -1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test hypothesis specifically.

86
Hypepneeis epent tne Dimensions of Hvalnepion
MM

Hypothesis 6 predicted that expectations (confirmation
versus violation) should exert different effects on
evaluations of a foreign TA along the dimensions of task,
relational, and communicative competence. Specifically,
H222£h§§i§_§§ predicted that when expectations were
confirmed, judgments of task competence should be
significantly more favorable than judgments of communicative
or relational competence. Further, Hypothesis 6b suggested
that when expectations were violated, judgments about the
foreign TA will not differ across the three dimensions.

To evaluate this hypothesis, a repeated measures
analysis of variance was performed crossing the between-
subjects factor of expectations (confirmed versus violated)
and within-subjects factor of type of evaluation (task,
relational, and communication). Results indicated
significant main effects for the confirmed and violated
conditions, H(1,313)=72.9; p<.001, and for the
three types of competence, £(2,626)=150.37; p<.001. Both
findings were qualified by a significant interaction between
confirmation of expectations and the three types of
competence, H(2,626)=175.18; p<.001.

Contrasts were fitted to interpret the interaction and
test the hypothesis more directly. When the expectations

were confirmed, results indicated that both the linear and

87
quadratic contrasts were statistically significant.
However, consistent with the prediction (Hypothesis 6a), the
linear contrast, e(316)=10.39: p<.001, explained more
variance than the quadratic contrast, p(316)=3.46; p<.002.
The linear contrast indicates that when expectations were
confirmed, judgments of task competence (H=3.05) were higher
than judgments of both relational (H=2.90) and communication

competence (H=2.45: see Table 10).

Table 10

Qell means fox task, relationall and communication
competence by confirmation of expectations

 

 

 

Competence
Task Relational Communication
Expectations
Violated 3.33 3.04 3.90
(159) (159) (159)
Confirmed 3.05 2.90 2.45
(156) (156) (156)
Note: Numbers in parentheses reflect number of students.

When expectations were violated, the data were not
consistent with the hypothesis. Contrary to the prediction
(Hypothesis 6b), there were significant differences between
the task, relational, and communication competence. Both

the linear, e(316)=10.14; p<.001, and quadratic contrasts,

88
e(316)=12.91; p<.001, were again statistically significant.
These results indicate, therefore, that when expectations
are violated, students’ evaluations of their foreign TA
differ significantly across the three dimensions of task,
relational, and communication competence. In particular,
students evaluate the foreign TA more favorably on
communication competence when he violates rather than
confirms prior expectations.

Overall, for Hypothesis 6, there was a main effect for
confirmation of expectations, and an interaction between
expectations and the three dimensions of competence.
However, the data were consistent with the hypotheses only
when expectations were confirmed. Students evaluated their
foreign TA more favorably on task competence than on
relational and communication competence under this
condition. When expectations were violated, contrary to the
prediction, students evaluated the foreign TA on task,

relational, and communication competence differently.

Hypotheses about Behavioral Intentions
We;

Hypothesis 7 predicted that students are more likely to
drop the class taught by the foreign TA on video tape when
s/he confirms rather than violates expectations. To test
this hypothesis, a oneway ANOVA testing the effects of

confirmation of expectations on dropping the class was

89
performed. Results indicated that the data are consistent
with the hypothesis. A significant main effect for
confirmation of expectations, H(1,315)=35.23: p<.001,
indicated that students are more likely to drop the class
when their expectations are confirmed (H=3.76) than when
expectations are violated (H=2.92). Intentions to drop the
class explained a moderate portion of the variance (eta-

squared=.10).

Hypgtnesis 8

Hypothesis 8 predicted that confirmation of
expectations and strength of expectations should interact to
influence students’ intentions to drop the course.
Specifically, Hypothesis 8a predicted that when expectations
were confirmed, students with strong expectations are more
likely to drop the class taught by the foreign TA than
students with weak expectations. Hypothesis 8b predicted
that when expectations were violated, students with strong
expectations are less likely to drop the course than
students with weak expectations. To test this hypothesis, a
two-way analysis of variance examining the effects of
confirmation of expectations and strength of expectations on
dropping the class taught by a foreign TA was performed.

Results indicate that while there is a main effect for
confirmation of expectations, H(1,310)=34.67; p<.001, there

is no main effect for strength of expectation, H(1,310=1.42;

90
p>.05. However, there is a significant interaction between
confirmation of expectations and strength of expectations
for dropping the class, H(1,310)=4.07; p<.05. Contrasts
were fitted to test the hypothesis directly (see Table 11).
Results indicated that the data were consistent with the
hypothesis, H(1,310)=36.29; p<.05, and the residual
explained variance was not statistically significant,
F(2,310)=1.51: p>.05. Specifically, when expectations were
confirmed, students with strong expectations were more
likely to drop the class than students with weak
expectations. Similarly, when expectations were violated,
students with strong expectations were less likely to drop

the class than students with weak expectations.

Table 11

Qell means for dropping the claes by confirmation of
expectations and strength of expectations

 

Strength of expectations

 

 

 

Strong Weak
Expectations
Violated 2.87 2.99
(92) (68)
-2 -1
Confirmed 3.97 3.51
(78) (76)
2 1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test hypothesis specifically.

91
Metal

Hypothesis 9 predicted that confirmation of
expectations and strength of expectations should interact to
influence whether students intend to take classes with a
foreign TA in the future. There were two parts to the ninth
hypothesis. Hypothesis 9a predicted that when a foreign TA
confirms expectations, students are equally likely not to
take a class with a foreign TA in the future whether they
have strong or weak expectations. Hypothesis 9b predicted
that when a foreign TA violates expectations, students with
strong expectations are more likely not to take a class with
a foreign TA in the future than students with weak
expectations. To test this hypothesis, a two-way analysis
of variance investigating the effects of expectations and
strength of expectations on avoiding taking classes with
foreign TAs in the future was performed.

Results from the analysis of variance initially
indicated that the data were not consistent with the
hypothesis. While there were main effects for confirmation
of expectations, §(1,309)=13.15; p<.001, and strength of
expectations, §(1,309)=25.13; p<.001, the interaction effect
was not significant, F(1,309)=.06: p>.05. As ANOVA picks up
only cross-over interactions, however, contrasts were fitted
to test the hypothesis (see Table 12).

Results from this analysis indicated that the data were

consistent with the hypothesis, p(309)=4.67; p<.001.

92
Specifically, when expectations are confirmed, students with
weak and strong expectations are equally likely not to take
classes with a foreign TA in the future. However, when
expectations are violated, students with strong expectations
are more likely to avoid taking classes with a foreign TA in

the future than students with weak expectations.

Table 12

Cell means for taking a class with a foreign TA in tne
fntnne by congixmation of expectations and strength of
expectations

 

Strength of expectations

 

 

 

Strong Weak
Expectations
Violated 3.10 2.46
(91) . (68)
0 -2
Confirmed 3.62 2.91
(78) (76)
1 1
Note: Numbers in parentheses reflect number of students.

Numbers below parentheses indicate contrasts used
to test hypothesis specifically.

Although the contrasts indicated that the data were
consistent with Hypothesis 9, the mean differences between
strong versus weak expectations looks about the same for
both confirmed and violated conditions. As a subsidiary

analysis, a oneway ANOVA exploring the effects of strength

93
of expectations on avoiding taking classes with foreign TAs
in the future was performed for confirmed and violated
conditions separately. Results indicated that there was a
main effect for strength of expectations under both
violated, H(1,157)=16.09; p<.002, and confirmed
expectations, H(1,152)=12.16; p<.002. Violated
expectations, however, explained a little more variance
(eta-squared=.08) than confirmed expectations (eta-

squared=.06), indicating a weak support for Hypothesis 9.

CHAPTER SIX

DISCUSSION

Ove ew

The discussion of the results is divided into several
sections. It begins with a description of general patterns
of findings across confirmed and disconfirmed hypotheses.
This is followed by discussing the implications of these
findings for the Language Expectation Model (LEM) and toward
designing programs for students to communicate more
effectively with their foreign TAs. Suggestions for future
research are offered in the last section. Limitations of

current research are noted within several sections.

Genemel Hatterns of Findings

Qonfirmed hypotheses. Most of the predictions in the
LEM relate to expectations (confirmed versus violated) and
the strength of expectations (strong versus weak).
Students, in general, had fairly strong negative
expectations that their foreign TA will have an accent that
is difficult to follow. The LEM (Figure 3) predicts that
(all other factors remaining constant) students will have a
more negative affective, cognitive, and behavioral response

when a foreign TA confirms rather than violates students’

94

95
language expectations.

Across most of the hypotheses, students express
stronger negative reactions when a foreign TA confirms
rather than violates students’ prior language expectations.
Specifically, students: (a) exhibit higher levels of anger
and anxiety, (b) evaluate the foreign TA less favorably on
task and communication competence, (c) are more likely to
drop a class taught by a foreign TA, and (d) are more likely
not to take a class with a foreign TA in the future, when a
foreign TA confirms rather than violates students’ language
expectations of foreign TAs.

The LEM also predicts that strength of expectations
(strong versus weak) will interact with expectations
(violated versus confirmed). When a foreign TA confirms
prior expectations, students with strong rather than weak
expectations: (a) exhibit higher levels of anger and
anxiety, (b) evaluate the foreign TA less favorably on task
and communication competence, (c) are more likely to drop a
class taught by a foreign TA, and (d) are more likely not to
take a class with a foreign TA in the future.

However, when a foreign TA violates prior
expectations, consistent with the predictions, students with
strong expectations vary in their responses across different
variables. Specifically, when a foreign TA violates prior
expectations, students with strong rather than weak

expectations are less likely to take a class with a foreign

96

TA in the future. In other cases, however, students with
strong rather than weak expectations: (a) feel less anger
and anxiety, and (b) are less likely to drop the course
taught by a foreign TA than students with weak expectations.

Overall, the LEM’s predictions for expectations
(confirmed versus violated) and strength of expectations
(strong versus weak) were confirmed. There were a few

predictions, however, that were not confirmed.

Disconfirmed hypotheses. Data are not consistent with
the hypotheses for positive affective responses. The LEM
predicts that students would be more happy and relieved when
a foreign TA violated rather than confirmed their
expectations. Students, however, were equally happy and
relieved whether the foreign TA confirmed or violated their
expectations. It is understandable that the students were
not happy or relieved when a foreign TA confirmed their
expectations (mean=2.12). The positive affective response
was, in fact, marginally lower when the foreign TA violated
the expectations. A plausible explanation is that students
felt that the foreign TA had done nothing extraordinary by
violating their expectations. The foreign TA had just
fulfilled the students’ right to have a teacher whose accent
is easy to follow. With strong negative expectations,
students’ reaction may have been, "About time we had more

foreign TAs who can speak English in a way that is easy to

97

follow. We do pay lots of money to get this education,"
rather than, "Wow! we finally have a foreign TA who can
speak fluently." In other words, the students’ reaction may
have been one of rightful indignation rather than happiness.

The LEM also predicts that expectations (confirmed
versus violated) should interact with strength of
expectations (strong versus weak) such that when a foreign
TA confirms prior expectations, students with strong
expectations will be less happy than students with weak
expectations. Also, when a foreign TA violates prior
expectations, students with strong expectations should be
more happy than a student with weak expectations. While
there was a significant interaction between expectations and
strength of expectations, the data were not consistent with
the model. When expectations were confirmed, students with
strong expectations had the least positive affective
response, as predicted. When expectations were confirmed,
however, students with weak expectations had the most
positive affective response. It is likely that the
students’ happiness had more to do with the confirmation of
expectation and little to do with the foreign TA’s severe
accent. In other words, students with weak expectations
were happy that what other students had mentioned was, in
fact, true.

The LEM predicts that the students’ area of expertise

(operationalized as major versus nonmajor; chemistry versus

98
communication) should interact with expectations and
strength of expectations to influence evaluation of their
foreign TA. There are no significant main or interaction
effects for major versus nonmajor. However, the
researcher’s expectation of obtaining a large sample for the
two groups (chemistry and communication) was violated. The
total sample is 330, of which 107 are communication majors,
and only 42 are chemistry majors. This small sample’s
further division into one of six conditions (expectations x
strength of expectations x confirmation of expectations)
made the cell sizes very small, making interpretation of the
results less useful.

Expectations (confirmed versus violated) did not
interact significantly with strength of expectations (strong
versus weak) to influence subtyping. In other words,
students think that the foreign TA they watched on video is
like other foreign TAs on campus. This was true whether the
foreign TA confirmed or violated their expectations, and
whether students had weak or strong expectations. As this
study only simulated a classroom encounter, and only with a
foreign TA on video tape, students’ impressions of this
foreign TA may not have generalized to other foreign TAs on
campus. Also, subtyping was measured by only one question,
and that item may not have captured the essence of the
concept.

As noted before, when expectations were confirmed,

99
students with strong expectations evaluated the foreign TA
less favorably on task and communication competence than
students with weak expectations. However, when expectations
were violated, contrary to the prediction, there were no
significant differences between students with strong and
weak expectations evaluating their foreign TA on task and
communication competence. Consistent with the earlier
findings on positive affective response, it is likely that
students’ rightful indignation (about time we had a foreign
TA that could speak English fluently!) rather than prior
expectations influences their judgment of foreign TAs when
expectations are violated.

While there were a few significant effects for task and
communication competence, there were no effects for
relational competence. Students evaluated their foreign
TA’s on relational competence equally across confirmed and
violated conditions, and across strong versus weak
conditions. This is not surprising as students did not get
a chance to evaluate the foreign TA after interacting with
him. They only watched him on videotape. All the questions
on relational competence were focused on interacting with
students while task and communication competence items
focused on the foreign TA’s ability in his subject matter
and his ability to communicate with the students in the

classroom.

100
lmplieetions for the Language Expectation Hodel

Since data were consistent with several of the
hypotheses, there are several significant implications for
the Language Expectation Model (LEM). ’

First,;once the teaching assistant is identified as a
"foreign TA:W expectations regarding his/her accent are
evoked (Figure 1). These expectations are clearly negative.
That is, students expect the foreign TA’s accent to be
difficult to followi_Furthermore, these negative
expectations range from moderately weak to very strong, with
an average "fairly strong" response. From this study, it is
clear that there are very few students who have no language
expectations of their foreign TA.

Second, students’ prior language expectations of their
foreign TA are either confirmed or violated (see Figure 3).
Data were altogether consistent with the model when
students’ expectations are confirmed rather than being
violated. Students are more angry and anxious, evaluate the
foreign TA less favorably, are more likely to drop the class
taught by this foreign TA, and feel more strongly that they
do not want to take classes with a foreign TA in the future.

However, when expectations are violated, students do
not have a significantly stronger positive affective
response than when their expectations are confirmed. At
best, they are less angry and anxious when their

expectations were violated. The model, therefore, has been

101

revised to reflect the students’ sentiment when expectations
are violated (see Figure 4). Even though students did not
feel very happy and relieved, they evaluated the foreign TA
more favorably, and were more likely to stay in class when
expectations were violated rather than being confirmed.
Although the students did not see their foreign TA as an
exception (that is, no subtyping), it is still retained in
the model as there was only a weak test of this variable
(one-item measure) in this study (see Figure 4).

Of the three moderating variables, course content
(major versus nonmajor) was the only variable tested in the
study. As there were no significant effects for this

variable, it has been removed from the model.

Snggestions for Training Students
The crux of the LEM is that students have strong

negative expectations of foreign TA’s language skills. The
training of students, therefore, should focus on addressing
students expectations of foreign TAs. Students can be
trained (and are, in some universities) in at least three
stages: (a) before they arrive at the university, (b) at
freshmen orientation, and (c) through the four years in a
university.

Before the students arrive at the university, it is
important to send a letter explaining that they will have

instructors from other countries. The number and profile of

102

Expectations of Foreign TAs Language

Skills (Strc

 

Experimentalc

 

manipulation

Confirmed

More Negative
Affective Response

ng vs Weak)

/\

violated

Le 3 Negative
Affective Response

 

Negative Attitudinal

Reaction
Reinforce Behavioral
Existing Intention
Expectations (drop class)

 

Positive‘Attitudinal
Reaction

Beha ioral
Intention
(stay in class)

Subtyping

 

Moderated by:
Personality Traits

Peers' Reactions

:ignxe : Revised Language Expectation Model -- Expectations
of Foreign TAs' Language Skills

103
foreign TAs should be provided. It should also be
emphasized that foreign TAs’ accent is gifferent, and
therefore mey be difficule to follow. Strategies students
can use when a foreign TA is not easy to follow need to be
defined. For example, meeting a foreign TA during office
hours helps.)Foreign TAs are generally more relaxed and
easier to folIow outside of the classroom. It is also
important to point out the benefits of having a class taught
by a foreign TA (for example, learning about a new culture).

u—WLL.‘

:It helps having a testimonial from the Provost’s office that

//

some of the best TAs on campus are foreign. )If a foreign TAU:
has won an excellence-in-teaching award, his/her photograph
along with the letter adds a positive note to the issue.
This letter should provide a more realistic expectation to
students entering the university, and offer them ways to
cemmnnicate more effectively with their foreign TAs.

At freshmen orientation, the information discussed
above should be reinforced. In some universities, students
meet a couple of foreign TAs in an informal setting, and
learn more about the foreign TAs’ background and culture.

At this orientation, students are also told the classes they
are most likely to have taught by a foreign TA. Again,

where students can find help is explained (for example, some
departments have "help hours," where a student can get help

from any of the TAs available at the time). It helps if

freshmen meet senior students who talk about both their

104
positive and negative experiences with foreign TAs, and more
importantly wnat steps phey took to communicate meme
w't he' 0 .

During the four years at the university, these
realistic expectations should be reinforced. Training
programs already are available in a few universities to
address this issue. The University of New Mexico has a
"buddy" program, where a foreign TA is paired with a
undergraduate student for 10 weeks. Each team has a number
of goals to reach during these 10 weeks. They learn about
each others’ culture. The undergraduate student teaches the
foreign TA about classroom etiquette, and what U.S.
undergraduates expect in a classroom. The student also
helps the foreign TA with his/her language skills. The
undergraduate student gets 1 or 2 credits for participating
in this program.

A program in Michigan State University meets groups of
students in residence halls who have problems with their
foreign TA. A foreign TA and an undergraduate student
facilitate this session. The program is about an hour long,
and a significant portion of the time is spent acknowledging
the problems students have with foreign TAs (generally
language and classroom teaching style problems). In a non-
threatening atmosphere, students are asked if they know the
problems foreign TAs face teaching in a different culture.

Few students have thought about the Oh No! Syndrome from the

105
foreign TA’s perspective. Students also are asked what
foreign TAs and undergraduates can do to communicate more
effectively with their foreign TAs. Again, while students
have a number of suggestions for foreign TAs, they have not
thought about what students themselves can do §e_eemmnnieepe
mere effeetively with thei; fereign T . This program’s
goals is to raise awareness about the issue, let the
students do some perspective-taking, and offer potential
solutions. An important assumption of this program is that
the Oh No! Syndrome ethos was created by students, and it is

best changed by them.

implications for Future Research

This study tested a model (LEM) to explain students’
affective, cognitive, and behavioral processes interacting
with their foreign TA on the first day of class. Apart from
the interesting findings, there were many lessons learned
about how to this study could have been conducted
differently to make it a better study. Suggestions are also

made to extend the scope of the LEM.

How woulg I conduct this study differently now? First,

the simulation would be made more realistic for the
students. In an actual classroom, students’ "Oh No!"
reaction is partly due to the difficulty in understanding

the accent, and partly because their grades in the class

106
depend on the foreign TA. In other words, the foreign TA
can reward or punish the student with grades. To simulate
this in the future, data could be collected only in classes
where the instructor agrees to give extra credit for
participation. Students will be told that they will watch
an instructor on the first day of class. Some of the
questions will require students to recall what they watched
on tape. If they got all the answers right, students will
get the whole 5 points. For getting most of it right,
students will get 3 points. If they have only a few right,
students will get only one point.

Apart from simulating the reward/punishment concept in
the classroom, offering extra credit contingent on recall
performance also may reflect on the students’ learning
styles (more motivated students eager to earn more points,
and less motivated ones not so eager to get points). After
the students are debriefed, they will be told that this was
just an experiment and that they will get all the 5 points.

Second, measurement problems should be addressed. More
items would be added to measure subtyping, intention to drop
the class, and to take classes with a foreign TA in the
future. Further, Watson, Clark and Tellegen’s (1988) PANAS
scale could be revised to make the items more appropriate
for the interaction between students and foreign TAs. To
make this revision, several focus groups could be conducted

with students to learn what they feel when a foreign TA

107
confirms or violates their expectations.

Finally, as much as access to students is a difficult
issue, data should be collected in the chemistry (and other
natural science departments) to see if there is an
interaction between expectations, strength of expectations,

the subject of the class, and the students’ major.

Hnet should we do next? A fruitful next step is to
conduct a similar study in a different university for a
comparative analysis. For example, the results from the
University of New Mexico (UNM) could be compared to results
from Michigan State University (MSU). There are several
reasons for comparing results from different student bodies.
First, students from UNM are from one of three cultures
(Native American, Hispanic, and Anglo), and these three
groups interact a lot. Students at MSU (Peltier, 1994) are
predominantly White (73%), along with a moderately small
representation of African American (7.4%), International
Students (6.0%), Asian/Pacific Islanders (3.3%), and
Hispanics (2.1%). Second, the average age of students at
UNM is 28, while at MSU it is approximately 22 (Peltier,
1994). Third, UNM has a total enrollment of around 27,000
students while the enrollment at MSU is around 40,000. This
influences class size, which is around 30 in the Department
of Communication and Journalism at UNM, and it is around

100-150 students in freshmen and sophomore level classes,

108
and 50-75 students in the junior and senior level classes in
the Department of Communication at MSU (Barkman, 1994).
Finally, there are 700 foreign TAs at MSU (Rittenberg,
1992), while there are only 150 foreign TAs at UNM
(International Student Office, UNM).

Given these student profiles at UNM and MSU, it is
likely that students at MSU have stronger negative
expectations of foreign TAs than students at UNM. A
comparative analysis is also useful to show how different
cultures (UNM versus MSU) view the world, and to design
training programs to meet the needs of specific cultures.

It would help also to study the effects of peer
responses on students’ evaluations of foreign TAs. For
example, a study might tape the accent of two foreign TAs
(same gender, and from the same country) outside of class,
and who are evaluated equally by different students. These
two foreign TAs would lecture on the same topic to two
different classes of about 40 students each. Each class has
six students who are actually the researcher’s confederates.
In one class, the confederates provide only negative
feedback verbally and nonverbally (making inappropriate
comments, interrupting, etc.). In the other class, the
confederates provide only positive feedback (supportive,
friendly, etc.). Students (who are not confederates)
evaluate the foreign TA in both classes. If the peer

response heuristic is true, the foreign TA’s accent in the

109
first class should be evaluated less favorably than his/her
evaluation originally (more negatively by students with
strong rather than weak expectations). Also, the accent
should be more difficult to follow than from the taping
outside of class (that is, without any peer response). The
foreign TA’s accent in the second class should be evaluated
equally or more favorably than his/her original evaluation,
and should be as easy or easier to follow than the taping
done outside of class (and more favorably by students with
strong rather than weak expectations).

1 These predictions are similar to Axsom, Yates, and
Chaiken’sr(1987) predictions on the role of audience
response as a heuristic cue in persuasion. They found that
when a audience member’s involvement in the message is low,
other audience members’ reaction would exert a greater
persuasive impact than argument quality. Conversely, when a
audience member’s involvement in the topic is high, argument
quality rather than other audience members’ reaction would
be more persuasive. The LEM would predict that whether
expectations are confirmed or violated, students with weak
expectations are more likely to watch other students’
reactions to make up their mind about the foreign TA in
question. Students with strong expectations are likely to
show their feelings (anger or relief) immediately, and not
depend on other students’ feelings to express their own

feelings.

110

Hey ean we extend the scope of the LEH? The LEM’s
scope can be extended in several ways. First, LEM’s
predictions are restricted to students’ affective,
cognitive, and behavioral responses to a foreign TA on the
first day of class. The model could be revised to study
students’ responses over time. For example, students’
affective, cognitive, and behavioral responses could be
measured three times in a semester (at the beginning,
middle, and end) for two foreign TAs -- one who confirms
students’ expectations, and another who violates students’
expectations. This will be very useful to track changes in
student responses (if any) over time.

Second, the usefulness of LEM in explaining how
students react to other types of teaching assistants could
be studied. For example, do students react differently to
female foreign TAs compared to male foreign TAs? How do
students react to foreign TAs from countries in Europe that
were under Soviet rule till recently (for example, former
East Germany, Poland, etc.)? These foreign TAs do not look
"foreign," but have accents that may not be easy to
understand. Also, how do students react to teaching
assistants who are Native American, African American, or
Hispanic?

On the other hand, does the ethnicity of the student
make a difference in their reactions to a foreign TA? Do

Hispanics, Native Americans, Asian Americans, and

111
international students react differently to foreign TAs
compared to Anglo students? Is there a difference between
male and female students among these different ethnic groups
to a foreign TA?

Third, the LEM could be applied to other contexts. For
example, many of the doctors today in general practice in
the U.S.A. are from countries other than the U.S.A. There
is anecdotal evidence that patients refuse to be attended by
doctors whose names look "foreign," or after the doctor
walks into the waiting room. This doctor-patient
interaction is similar to the foreign TA-student interaction
in many ways. The doctor and the foreign TA are: (a)
providing a service, health care and education, (b) from
outside the U.S., and likely to have accents different from
the accents common in the U.S., and (c) generally highly
competent in what they do. The student and the patient, on
the other hand: (a) have to pay a significant amount of
money to receive the service, (b) are not familiar with new
and different accents, and (c) may not be aware that the

service provider is very competent.

Going beyond the LEM. A next step would be to see how
students’ language expectation influences their interaction
with a foreign TA in the classroom. For example, the first
day of class with a foreign TA could be taped (both video

and audio) to see if the self-fulfilling prophecy prediction

112
(mentioned in an earlier chapter as an alternate explanation
for the Oh No! Syndrome) is true. Students expect their
foreign TA to speak with an accent difficult to follow.
Consistent with this expectation, students may exhibit their
anger and anxiety verbally (asking questions that may be
personally offensive to the foreign TA) and nonverbally
(interrupting the foreign TA, asking questions in an
aggressive tone, etc.), and make the foreign TA apprehensive
and anxious. The foreign TA’s accent may not be normally
difficult to follow. However, the anxiety and apprehension
created by the students may make the foreign TA’s accent
more difficult to understand. In other words, the foreign
TA’s difficulty in communicating clearly may have to do more
with students’ expectations of the foreign TA’s accent than
with the foreign TA’s accent itself.

The LEM, however, may provide more precise predictions
about foreign TA - U.S. student interaction than the general
self-fulfilling prophecy hypothesis. The latter hypotheses
would seem to suggest that students with strong
expectations, in general, should be more likely than
students with weak expectations to engage in behaviors
promoting self-fulfilling prophecies. In contrast, the LEM
predicts that students with strong expectations should be
nee; likely to create self-fulfilling prophecies only when a
foreign TA confirms prior expectations. This would be

consistent with the self-fulfilling prophecy hypothesis.

113
When expectations are violated, however, students with
strong expectations should be leeee likely to create self-
fulfilling prophecies. The LEM predicts that students with
strong expectations will evaluate the foreign TA more
positively than students with weak expectations, but are
less likely to show it openly. This would reduce the
possibility of creating a self-fulfilling prophecy.

It would also be useful to study the Oh No! Syndrome
using Uncertainty Reduction Theory (Berger & Calabrese,
1975: Gudykunst, 1983) which suggests that we feel uncertain
in initial encounters with other individuals. Also, we
attempt to reduce uncertainty during initial encounters
either proactively (before we meet the individual for the
first time) or retroactively (after meeting the individual
for the first time). Undergraduates and foreign TAs, having
been socialized in different cultures, are sure to feel
uncertain in initial encounters, and probably use a variety
of ways to cope with this uncertainty. Nelson (1989) found
that undergraduates’ felt less uncertain when foreign TAs
used more personal examples. This does not suggest, however,
what undergraduates need to do to reduce foreign TAs’
uncertainty. It would be useful, for both and for training
purposes, to understand the nature of uncertainty
undergraduates and foreign TA feel interacting on the first
day of class, and to analyze how they negotiate this

uncertainty in the classroom.

114
ma 5

This study began with a personal interest to see why
students often evaluate their foreign TAs poorly. There was
enough research predicting that if we only took care of the
foreign TAs language skills, the "Oh No! Syndrome" would go
away. After more than ten years of research, it is clear
that training only the foreign TAs will not suffice.

As an initial step to support the argument that the Oh
No! Syndrome is at least a two-way process (including both
foreign TAs and students), this study offered the Language
Expectation Model to explain the students’ affective,
cognitive, and behavioral processes when they encounter a
foreign TA on the first day of class. As this study
indicates, there is a lot of work yet to be done.

With more research and training of both the students
and the foreign TAs, this researcher hopes to see the day
when Sugmin (our friend from beginning of the study) walks
into a class, and the students unanimously say, "Wow! a TA

from Taiwan. We are so lucky!"

APPENDICES

 

APPENDIX A

APPENDIX A

Michigan Senate Bill No. 518

A bill to provide that certain instructors be orally
proficient in the English language; and to prescribe the

powers and duties of certain public officials.

THE PEOPLE OF THE STATE OF MICHIGAN ENACT:

(a) "Governing board" means a board of regents, board of
trustees, board of governors, board of control, or
other governing body of an institution of higher

education.

(b) "Institution of higher education" or "Institution"
means a college or university listed in section 4 of
article VIII of the state constitution of 1963 or a
community college or junior college established
pursuant to section 7 article VIII of the state

constitution of 1963.

(c) "Instructor" means a teaching assistant except those
teaching a foreign language who provides classroom

instruction to students enrolled full-time or part-time

115

 

KIA. \. ‘ U -

116

in an institution of higher education. Instructor does

not include a visiting scholar to the institution.

Section 2.

The governing board of an institution of
higher education shall ensure that, not later
than the commencement of the 1988-89 academic
year, each instructor who is not orally
proficient in the English language attains
such proficiency before providing classroom

instruction to students.

 

APPENDIX B

 

APPENDIX B

Pxetest Interpersonal Communication Lecture Script
This is C&J 321, Interpersonal Communication, and if
you don’t belong to this class, please feel free to leave

any time.

 

 

My name is and I am from .
I came to the U.S. to do my (M.S., Ph.D.) in
. Back home in , I studied

 

 

. I have (mention family members here).

 

First, we will take a look at the syllabus. And then I
will explain some basic issues about interpersonal
communication.

Let us look at the syllabus. Like I said before, my

name is , and please call me .

 

This class meets Monday and Wednesday, 11 A.M. to 12:15 P.M.
My offices are in 126 Marron Hall, and my phone number is
277-7571. My office hours are from 2:30 to 3:30 P.M.
Tuesday, Wednesday, and Thursday, and also by appointment.
If you want to leave a message with the department
secretary, the number is 277-5305.

The objectives of this class are to introduce you to
various theories in interpersonal communication, and see how

they apply to real-life situations. The required text book

117

is 1
wil

Now

118
is Trenholm and Jensen’s Interpersonal Communication. I
will provide additional readings as the class progresses.

Now, let us turn to Page 2 of the syllabus.

APPENDIX C

V1

APPENDIX C

lntexpersonal Communication Lecture Script for the Study
This is C&J 321, Interpersonal Communication, and if
you don’t belong to this class, please feel free to leave

any time.

 

 

My name is and I am from .
I came to the U.S. to do my (M.S., Ph.D.) in
Back home in , I studied

 

 

. I have (mention family members here).

 

First, we will take a look at the syllabus. And then I
will explain some basic issues about interpersonal
communication.

Let us look at the syllabus. Like I said before, my

name is , and please call me .

 

This class meets Monday and Wednesday, 11 A.M. to 12:15 P.M.
My offices are in 126 Marron Hall, and my phone number is
277-7571. My office hours are from 2:30 to 3:30 P.M.
Tuesday, Wednesday, and Thursday, and also by appointment.
If you want to leave a message with the department
secretary, the number is 277-5305.

The objectives of this class are to introduce you to
various theories in interpersonal communication, and see how

they apply to real-life situations. The required text book

119

120
is Trenholm and Jensen’s Interpersonal Communication. I
will provide additional readings as the class progresses.

There will be three exams and five quizzes. Each quiz
has ten questions, and is worth 10 points. Each exam will
have 50 questions -- multiple choice and short essay -- and
worth 50 points. The grading scale is explained clearly in
the syllabus, and is based on a straight scale.

I don’t take attendance, but from my experience, I know
there is a direct relationship between attendance and your
final grade. The more you attend, the better your grade.
The tentative schedule for the class is listed finally.

Any questions? Wait for a couple of seconds and then
proceed.

This class is interpersonal communication. Let us take
a couple of moments to understand what we mean by the two
key words "communication" and "interpersonal."

Communication is the symbolic transactional process to
create, modify, or reinforce the attitudes and behaviors of
others. Some key terms in this definition. Sympolic stands
for the fact that we use symbols, or a language to
communicate. When we use language, we are sending verbal
and nonverbal messages. Transactional stresses that there
are at least two parties communicating with one another. It
is a pxeeeee because there are set of activities occurring
over time. We communicate primarily to persuade others,

that is, create, reinforce, or modify the attitudes and

121
behaviors of others. For example, right now, I am trying to
create a positive attitude towards interpersonal
communication.

The other half of the picture deals with
"interpersonal." This looks at the communication between
two people. This communication could be face—to-face or
mediated. When it is mediated, it could be over the
telephone, fax, e-mail, and so on.

I have given you a brief introduction to interpersonal
communication. For next class, please read chapters one and
two from the Trenholm and Jensen book. If you don’t have any

questions, we will meet on Wednesday.

APPENDIX D

 

APPENDIX D

enemisnny Leb Scxipt fo; the Stndy
This is Analytical Chemistry 254, Section 3. If you

don’t belong to this class, please feel free to leave any
time.

My name is and I am from .
I came to the U.S. to do my (M.S., Ph.D.) in

. Back home in , I studied

 

 

. I have (mention family members here).

 

First, we will take a look at the syllabus. And then I
will explain some basic issues about analytical chemistry.
Let us look at the syllabus. Like I said before, my

name is , and please call me _ .

 

My office hours are from 2:00 to 3:00 P.M. every
Wednesday. If you have any questions, please don’t hesitate
to come over and ask. Please feel free to ask me any
question about the lecture also.

Because this is a chemistry lab, you know we will be
dealing with a lot of chemicals and most of them are
dangerous and toxic. So you have to be careful and your
safety is our primary concern. Because of that you have to
wear goggles all the time. You also have to wear lab coats

all the time in the lab. Otherwise you could be hurt.

122

 

123

Let us talk about the grading system a little bit.
There will be nine labs. Each lab is worth ten points,
totalling up to 90 points. Labs 1 and 2 are compulsory, and
after doing these labs, you are allowed to do the remaining
7 labs in any order you please. 10 more points are given
for maintaining your record in a neat and orderly manner.

Any questions? Wait for a couple of seconds and then
proceed.

Today, I will introduce you to how to do the first lab.
The first lab is about titration. And as you know for
titration we have to use a burette. A burette has four
significance levels. We also have to weigh the primary
chemical Calcium Carbonate. And you can get Calcium
Carbonate from Gary in the stock room. You have to weigh
the calcium carbonate using the balance. You need four
significance figures because the burette has four
significance levels. If you get only three, it is not
perfect, and you need to try it again.

That is all I have to say today. So, good luck and let

us get started.

 

APPENDIX E

 

APPENDIX E

Pre-Video Questionnaire
Number:

Please understand that the responses to these questions are
voluntary. You are free to stop answering questions at any
point in time. For those questions you answer, however,
please react as you would normally and complete all the
answers. Please do not write your name on any part of this
questionnaire to insure anonymity.

Please circle or put a check mark for your responses

 

1. Sex: Male Female 2. Age
3. Level: Freshmen
Sophomore
Junior
Senior
4. Ethnicity: Hispanic
Caucasian/Anglo

African American
Asian/Pacific Islander
Native American

Other

 

Please specify

5. What is your major?

 

The next set of questions assess your expectations of
instructors here at the University of New Mexico. While
answering the questions, we would like you to predict the
instructors! behaviors. That is, we want your answers to
reflect what you expect your instructors gill do rather than
what you think they should do.

In general, I think the instructors will:

 

6. Have a good grasp of the subject they teach.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

124

 

10.

11.

125
Make the class fun and interactive.
Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree
Allow and answer questions enthusiastically.
Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree

disagree

Speak with an accent that is easy to follow.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

Have poor communication skills.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

Offer relevant examples to explain each important
issue.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

Now, let us take the specific case of foreign teaching

nsgistants.

In general, you expect your foreign teaching assistants

will:

12.

Have a good grasp of the subject they teach.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

 

Again, while answering the questions, we would
like you to predict instructors! behaviors, that is, answer
what they will do rather than what you think they should do.

13.

14.

15.

16.

17.

18.

126

Have poor communication skills: s/he will be difficult

to understand.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

Allow and answer questions enthusiastically.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

Speak with an accent that is easy to follow.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

Make the class fun and interactive.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree' disagree
disagree

Offer relevant examples to explain each important
issue.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

How many of your classes so far (including the one you

are taking now) have been taught by foreign teaching
assistants?

0
1
2
3
4

or more classes

 

127

19. Has anybody (friends, classmates, room mates, etc,.)
talked to you about what it is like to take a class
taught by a foreign teaching assistant?

No Yes. If yes, what did they tell you
about these classes?

Einelly. e eenple e; gnestions abont how you see yenneelfi es
Lattices:

20. You always come prepared for class by reading the
required chapters in the text book.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

21. You believe that learning is a two-way process; both
the instructor and you as student expend equal amounts
of energy to make it an enriched learning experience.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

22. You rarely participate in class discussions because you
feel that it is the instructor’s responsibility to
impart information.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

23. You would label yourself as an active learner, one who
puts a lot of your own effort in to the class.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

 

128

Number:
Postvideo Questionnaine
Yo s wa bed a rei n t ach n as 'stant on e t

day of class. Please answer the following gnestions
essuming thet this was your instructor {or ehe eemestex, ane
tnis in pne fins; gay o; class.

Please circle or put a check mark for your responses.

ow' scale consists a umb o wor s
gescribe different feelinqe and emotions. Read each item
and then mark the appropriate answer in the space next to
enat more. Indicate to what extent you feel this way ebout
yo u; instructor in the videotape right now.

 

 

 

1 = extremely

2 = quite a bit

3 = moderately

4 = a little

5 = very slightly

or not at all

interested irritable
distressed alert
excited ashamed
upset inspired
strong nervous
guilty determined
scared attentive
hostile frustrated
enthusiastic relieved
proud afraid

21. This foreign TA was easier to understand than I
expected.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

 

22.

23.

24.

25.

26.

27.

129

I expected this foreign TA to speak with a thick
accent, and just as I expected, he was very difficult
to follow.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

My expectations were violated; this foreign TA spoke
with an accent much easier to follow than I expected.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

This foreign TA confirmed my expectations; he spoke
with an accent that was not at all easy to follow.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

This foreign TA does not appear to know the subject
matter at all.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

It is easy to understand this foreign TA’s accent.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

The foreign TA did not seem very interested in teaching
the course.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

 

28.

29.

30.

31.

32.

33.

34.

130

The foreign TA made the course goals clear to you

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

This foreign TA’s accent is difficult to understand.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

This foreign TA appeared relaxed, open, and
comfortable.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

You think that your knowledge and skills will increase

significantly after this class.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

This foreign TA communicated at a level you could
easily follow rather than talk over your head.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

This foreign TA’s accent is not very severe.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

You would feel very comfortable asking questions in
this class.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

35.

36.

37.

38.

39.

40.

131

This foreign TA was not clear about what he considered
to be important.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

This foreign TA would encourage you to see him if you
were having difficulty in class.

Strongly Moderately Neither Moderately Strongly

agree agree agree nor disagree disagree
disagree

This foreign TA’s presentation was interesting and
challenging.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

This foreign TA has a totally unintelligible accent.

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

This foreign TA is an expert in what he is teaching

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

This foreign TA created a friendly and relaxed class
room atmosphere

Strongly Moderately Neither Moderately Strongly
agree agree agree nor disagree disagree
disagree

The following set of questions ask you to recall specific
issues about the foreign TA or what he talked about.

True or False?

41.

This foreign TA is from China.

 

132
His name is Wu Ching.

His office hours are from 2:30 to 3:30,
Tuesday, Wednesday, and Thursday.

He came to the U.S. to do his Master’s in
Communication. '

Interpersonal communication could be face-to-
face or mediated.

The objective of this class is to only learn
how to apply interpersonal communication
theories to real-life situations.

There will be three exams and five quizzes in
this class.

Symbolic stands for the fact we use language
to communicate with each other.

The foreign TA said, "This is C&J 325,
Interpersonal Communication."

Communication is a process because we have at
least two people interacting with each other.

These last set of questions are some general questions about
the foreign TA and the class.

51. I would be very likely to drop this class.

Strongly Moderately Neither Moderately Strongly

agree

agree agree nor disagree disagree
disagree

52. This foreign TA is an exception. I am sure other
foreign TAs are not like him.

Strongly Moderately Neither Moderately Strongly

agree

agree agree nor disagree disagree
disagree

53. I will avoid taking classes with a foreign TA again.

Strongly Moderately Neither Moderately Strongly

agree

agree agree nor disagree disagree
disagree

LIST OF REFERENCES

LIST OF REFERENCES

Alloy, L. B., & Tabachnik, N. (1984). Assessment of
covariation by humans and animals: The joint
influence of prior expectations and current

situational information. Psyenolegicel Heyiew, 2;,
112-149.

Altmeyer, B. (1988). Enemies of freedom: Undenstanding

xignt-wing antnoritarianism. San Francisco: Jossey-
Bass Publishers.

Anderson-Hsieh, J. (in press). Teaching
suprasegmentals to international teaching

assistants. English fox Specific Purposes.

Arthur, B., Farrar, D., & Bradford, G. (1974).
Evaluation reactions of college students to dialect
differences in the English of Mexican-Americans.

Lengnage and Speech, l1(3), 255-270.

Asch, S. (1956). Studies of independence and
conformity: l. A minority of one against a

unanimous majority. gsyenelogieel Honogxapns, 19,
1-70 e

Axsom, D., Yates, S., & Chaiken, S. (1987). Audience
response as a heuristic cue in persuasion. Journal of

Bezeonelity end Social Psychology, £2. 30-40.

Bailey, K. M. (1982). Teachinqxin a second language:
The commnnicetive competence of non-native speaxing
neagning assistants. Unpublished dissertation,
University of California, Los Angeles.

Bailey, K. M. (1983). Foreign teaching assistants in
U.S. universities: Problems in interaction and

communication. TESOL Quarterly, 17(2), 308-310.

Bailey, K. M. (1984). The "foreign TA problem." In
K. M. Bailey, F. Pialorsi, & J. Faust. (Eds.)
Horeign teaching assistants in U.S. universities
(pp. 3-14). Washington, D.C.: National Association
for Foreign Student Affairs.

133

 

134

Barkman, M. (1994). Conversation with secretary, Department
of Communication, Michigan State University.

Berger, C., & Calabrese, R. (1975). Some explorations
in initial interaction and beyond. Human

Qemmnnieetion Heseaxch, l, 99-112.

Bodenhausen, G. V. (1993). Emotions, arousal, and
stereotypic judgments. In D. Mackie & D. Hamilton

(Eds.), Affect, cognition, and stereotyping (pp 13-
37). San Deigo, CA: Academic Press.

Boyd, F. A. (1989). A perspective derived from professional
education. English fox Specifie Puppesee, §(2), 195-203.

Bresnahan, M. (1990). Attitndes toward foreign
teaching essistents at a migyespenn nnivensipy.

Unpublished manuscript, Michigan State University.

Bresnahan, M., & Kim, M. (1991a, May). Tne effeep pf

enthoritanienism in bias toward foreign teacning
essistants. Paper presented at the 4lst Annual

Conference of the International Communication
Association.

Bresnahan, M., & Kim, M. (1991b, August). Amenicen
e rad a iv' o e' n each'

s' tants: n issue of En lis oficienc ? Paper
presented at the 4th International Conference on
Social Psychology and Language, Santa Barbara,
California.

Briggs, S., Hyon, S., Aldridge, P., & Swales, J.
(1990). The international teacning assistant: An

enneneted eritical bibliognaphy. The University of
Michigan: The English Language Institute.

Brown, K. (1990). College student attitudes towarg
nen-nepive instructors. Unpublished manuscript.

Burgoon, J. K. (1978). A communication model of
personal space violations: Explication and an

initial test. Human Communicetion Researen, e, 129-
142.

 

Burgoon, J. K. (1983). Nonverbal violations of
expectations. In J. M. Weimann & R. P. Harrison

(Eds.) Nonverbal interaction (pp 77—111). Beverly
Hills, CA: Sage.

135

Burgoon, J. K. (1985, May). Expecneneies, newande.
yielatiens and outcomes: Application to tne
insnructional environment. Paper presented at the

annual meeting of the International Communication
Association, Honolulu.

Burgoon, J. K., & Aho, L. (1982). Three field
experiments on the effects of conversational

distance. gommunication Honognaphs, 42, 71-88.

Burgoon, J. K. & Hale, J. L. (1988). Nonverbal
expectancy violations: Model elaboration and
application to immediacy behaviors. Communicapien

Monographs. 55. 58-79.

Burgoon, J. K., & Jones, S. B. (1976). Toward a theory
of personal space expectations and their violations.

Hnmen Qommnnication Reseazen, 2, 131-146.

Burgoon, J. K., & Walther, J. B. (1990). Nonverbal
expectancies and the evaluative consequences of

violations. Human Communication Researcn, 11(2),
232-265.

 

Byrd, P., & and Constantinides, J. C. (1988). PTA
training programs: Searching for appropriate
teaching styles. English for Specific Purposes,

2(2), 123-129.

Callan, V. J., Gallois, C., & Forbes, P. A. (1983).
Evaluative reactions to accented English. Journal

of Cross-Cultural Psychology, 14(4). 407-426.

Chaiken, S. (1987). The heuristic model of persuasion.
In M. P. Zanna, J. M. Olson, & C. P. Herman (Eds.),

The nenzistic model of persuasion (Vol 5. pp. 3-39).
Hillsdale, NJ: Erlbaum.

Constantinides, J. (1987). The "foreign TA" problem:
An update. NAFSA Newsletter, ;H(5), 3-6.

Delia, J. G., O’Keefe, B. J., O’Keefe, D. J. (1982).
The constructivist approach to communication. In F.

E. Dance (Ed.), Human Communication Theony:
Cempenative essays. New York: Harper & Row.

Edwards, J. R. (1982). Language attitudes and their
implications among English speakers. In E. B. Ryan &

H. Giles (Eds.), ntgitudes towargs language

veniepion;Social and applied contexts. London:
Edward Arnold.

136

Fazio, R. H. (1990). Multiple processes by which
attitude guides behavior: The MODE model as an
integrative framework. Adv es ' x e 'menta

Seeiel_2§22helegx. 2;. 75-109.

Giles, H., & Powesland, P. F. (1975). _§E§§Qh_§§¥1§
eng soeiel evaluapion. London: Academic Press.

Gill, M. M., & Badizinski, D. M. (1992). The impact of
accent and status on information recall and

perception information. Commnnicepion Heports, §(2),
99-106.

Gudykunst, W. (1983). Uncertainty reduction and
predictability of behavior in low and high context

cultures. Communication Querterly, el, 49-55.

Gudykunst, W. B. (1991). Bridging differences:
e iv i t on ommun' ion. Newbury Park,
CA: Sage Publications.

Hewstone, M., & Brown, R. (1986). Contact is not
enough. In M. Hewstone & R. Brown (Eds.), Contact

end eonflict in intergroup encounters (pp. 29). Oxford:
Blackwell.

Inglis, M. A. (1988). Variables that affect
nndergnaduetes’ evaluations of non- native speaking

peecning assistants’ instrucpion. Unpublished
doctoral dissertation, Memphis State University.

Jenkins, S. A. (1989, November). A comparison of the
eptitudes of American and international teaching

essistants to their undergraduates. Paper presented
at the Second National Conference on the Training

and Employment of Teaching Assistants, Seattle,
Washington.

Johncock, P. (1991). PTA tests and University FTA
testing policies. Working Papens of tne l987
Hniyexsity o; WyomingZNAFSA Institute on FTA
Training-

Lambert, W. E., Hodgson, R. E., Gardner, R. C., &
Fillenbaum, S. (1960). Evaluational reactions to
spoken languages. gonrnal of Abnormal and Sociel
Esychology, pg, 44-51.

Linville, P. W. (1982). The complexity-extremity effect

and age-based stereotyping. Journal of Personality
ang Soeial Psychology, 42, 193-211.

137

Linville, P. W., 8 Jones, E. E. (1980). Polarized

appraisal of out-group members. genxnel_eg_
Peregnelit y end Sociel Es yenelegy, 15. 689-703.

Lyczak, R., Fu, G. S., 8 Ho, A. (1976). Attitudes of
Hong Kong bilinguals toward English and Chinese

speakers. gennnal of Cross-Cultnral Esychology,
425-438.

Nelson, G. L. (1989). The nelapionship pepween the use

e so am fore n teac n s ' ’
tures nd uncertaint reduction student

eppitude. spudent recall, and ethnocentxism.

Unpublished dissertation, University of Minnesota.

Peltier, L. (1994). Conversation with Manager, Planning and
Budget, Michigan State University.

Petty, R. E., & Cacioppo, J. T. (1986). Communica 'on

e su s'on: Central a e ' ral outes to
epeipnge_enenge. New York: Springer-Verlag.
Rao, N. (1993, November). h No! 0. .: eac
e adua es to commun'cat e ect'vel w't

pneir foreign TAs. Paper presented at the Fourth
National Conference on the Training and Employment
of Graduate Teaching Assistants, Chicago, Illinois.

Rittenberg, B. (1992). Conversation with the person in
charge of the Foreign Teaching Orientation at
Michigan State University campus.

Rosenthal, R., & Jacobson, L. F. (1968). Teacher
expectations for the disadvantaged. Seienpifiie

Amalgam. 213(4). 18-23.

Rubin, D. L., & Smith, K. A. (1990). Effects of
accent, ethnicity, and lecture topic on
undergraduates’ perceptions of nonnative English-

speaking teaching assistants. International Hennnal
e: Innercultural Relations, l1, 337-353.

Ryan, E. B., & Giles, H. (1982). Attitudes towards
a e var'ation: Social and a lied cont ts.
London: Edward Arnold.

Schneider, D. J., Hastorf, A. H., & Ellsworth, P. C.

(1979). Eerson Perception. Reading, Mass.:
Addison-Wesley Publishing Co.

138

Sequiera, D., & Constantino, M. (1989). Issues in ITA
training programs. In J.D. Nyquist, R.D. Abbott, &

D.H. Wulff (Eds.) Teecning essistant treining in
the 19905 (pp. 79-86). San Francisco: Jossey-Bass.

Shepherd, G. J., & Trank, D. M. (1986, November).
Construce dififexentiation and multidimensional
jnegments of effectiyeness. Paper presented at the

Speech Communication Association, Chicago, Illinois.

Sherif, M. & Sherif, C. W. (1966). Qzenpe_in_nezmeny_
eng_peneien. New York: Octagon Books.

Sherif, C. W., Sherif, M., & Nebergall, R. E. (1965).

Ateitudes eng attitude change: The social-judgmene
involvement approach. Philadelphia: W. B. Saunders
Company.

Smith, S. W. (1991, May). The ef ects f s ron

nonvenbal relational data on perceptual processing

and resulting social judgments. Paper presented at
the annual meeting of the International

Communication Association, Chicago, Illinois.

Staff. (1989, October 27). Ombudsman: Foreign TAs
understood. The State Neye, Michigan State
University.

Stevens, S. G. (1989). A "dramatic" approach to
improving the intelligibility of ITAs. English
fer Specific Purposes, §(2), 181-194.

Tedeschi, J. T. & Reiss, M. (1981). Identities, the
phenomenal self, and laboratory research. In J.T.

Tedeschi (Ed.). Impression Management Tneoxy and
§eeiel Psychological Research (pp. 4-10). New York:

Academic Press.

Thomas, C. F., & Monoson, P. (1989, November). State
mandated oral English language proficiency
megnirements: Should we be concerned? Presented at
the Second National Conference on the Training and
Employment of Teaching Assistants. Seattle,
Washington.

Thomspson, L., & DeHarpport, T. (1990, June).
Negotiation in long-term relationships. Paper
presented at the Third annual meeting of the
International Association for Conflict Management,
University of British Columbia, Vancouver, Canada.

 

139

Triandis, H.C., Bontempo, R., Villareal, M., Assai, M.,
8 Lucca, N. (1988). Individualism-collectivism:
Cross-cultural perspectives on self, in-group

relationships. gournal o: Pensonaliny ang Sociel
Esycnelogy, 55, 323-338.

Troidahl, P. 8 Powell (1972). Short degmatism scale.
In Robinson 8 Shaw (Eds.) Measures of social

psyenelogical attitude. Ann Arbor: Institute for

Social Research.

Watson, D., Clark, L. A., 8 Tellegen, A. (1988).
Development and validation of brief measures of
positive and negative affect: The PANAS scales.

Heuxnal of Personality and Social Psychology, §4(6),
1063-1070.

Weber, R., 8 Crocker, J. (1983). Cognitive processes
in the revision of stereotypic beliefs. Journal

0: Personality and Social Psychology, 45. 319-322.

Wilder, D. A. (1990). Some determinants of the
persuasive of in-groups and out-groups: Organization
of information and attribution of independence.

Joumnal of Personality and Social Psychology, H2,
1202-1213.

Wol-Young, S. (1989). A profile of communicanion
sxills of foreign teeching assistants in a major
migwespenn nnivensity. Unpublished doctoral

dissertation, University of Cincinnati.

"171111111111111111111“