University of Southern California Dissertations and Theses

Acculturation styles, gender differences, and meanings of smoking in Asian-American college students

(USC Thesis Other)

Acculturation styles, gender differences, and meanings of smoking in Asian-American college students

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ACCULTURATION STYLES, GENDER DIFFERENCES, AND MEANINGS OF
SMOKING IN ASIAN-AMERICAN COLLEGE STUDENTS
by
Fan-Ni Hsia

A Dissertation Presented to the
FACULTY OF THE ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION

December 2006

Copyright 2006 Fan-Ni Hsia
ii
ACKNOWLEDGEMENTS
I would like to acknowledge the support and guidance of all the members of my
dissertation committee. Dr. Maurice Hitchcock, my dissertation chairperson, has been
particularly supportive and patient as I have negotiated this perplexing research process.
Dr. Donna Spruijt-Metz, my dissertation co-chair, who is most responsible for
helping me complete the dissertation. She has been a friend and mentor. She is always
there for me and gives me guidance in not only my academic life but also my personal
life. Without her encouragement and constant guidance, I could not have finished this
dissertation. Dr. Melora Sundt and Dr Linda Hagedorn, my dissertation committee, asked
me challenging questions and gave me insightful comments.
I thank my family, especially my parents, Lien-Sen Hsia and Yi-Ning Yu, for
educating me, and for giving me unconditional support to pursue my interest. I also thank
my friends, Qin Haung and Ashley Chen who accompanied me when I was feeling down
and stressed during the process of writing the dissertation.
A special thank to Tobacco-Related Disease Research Program (TRDRP). I was
funded by TRDRP, grant number 11DT-0161.
iii
TABLE OF CONTENTS

ACKNOWLEDGMENT ii

LIST OF TABLES v

ABSTRACT vii

CHAPTER 1: INTRODUCTION 1
Problems to be Investigated and the Significance of These Problems 1
Purpose of the Study 19
Organization of the Study 20
Overall Research Design 21
Definition 28

CHAPTER 2: ACCULTURATION AND THE MEANINGS OF SMOKING
AMONG ASIAN-AMERICAN COLLEGE STUDENTS 33
Introduction 33
Method 41
Results 54
Discussion 63

CHAPTER 3: GENDER DIFFERENCES IN MEANINGS OF SMOKING
AMONG ASIAN-AMERICAN COLLEGE STUDENTS 72
Introduction 72
Method 79
Results 88
Discussion 96

iv
CHAPTER 4: FACTOR STRUCTURE AND INV ARIANCE OF THE
MEANINGS OF SMOKING INDEX FOR ASIAN AMERICAN COLLEGE
STUDENTS (MSI-AACS) 104
Introduction 104
Figure 1: The Proposed Theoretical 3-factor Structure of Meanings
of Smoking 111
Method 112
Results 124
Discussion 128

CHAPTER 5: DISCUSSION 134
Discussion of the meanings of smoking 134
Discussion of the factorial structure of the meanings of smoking 138
Limitations and suggestions for future research 140
Unique contributions of the study 142
Conclusions 144

REFERENCES 146

APPENDICES 161
APPENDIX A. Demographic Items 161
APPENDIX B. Acculturation Scale for Asian-American College Students 164
APPENDIX C. Items for Smoking Behaviors 167
APPENDIX D. The Meanings of Smoking Index for Asian-American
College Students (MSI-AACS) 171
v
LIST OF TABLES

Table 1: Demographic Characteristics of Respondents and Descriptive
Statistics across Colleges 55

Table 2: Comparisons between Rankings of Meanings of Smoking
across Colleges (Spearman’s Correlation Coefficients) 56

Table 3: Demographic Characteristics of Respondents and Descriptive
Statistics across Groups (Spearman’s Correlation Coefficients) 57

Table 4: Meanings of Smoking by Subgroups According to Salience 58

Table 5: Association between Group Rank Orderings of Meanings of Smoking 59

Table 6: Meanings of Smoking by Subgroups According to Relevance
(Odds Ratios) 60

Table 7: Association Between Rank Orderings According to Salience and
Relevance Per Group (Spearman’s Correlation Coefficients) 61

Table 8: Demographic Characteristics of Respondents and Descriptive
Statistics across Colleges 89

Table 9: Comparisons Between Rankings of Meanings of Smoking
across Colleges (Spearman’s Correlation Coefficients) 90

Table 10: Demographic Characteristics of Respondents and Descriptive
Statistics across Genders 90

Table 11: Meanings of Smoking by Genders according to Salience 92

Table 12: Association Between Group Rank Orderings of Meanings
of Smoking (Spearman’s Correlation Coefficients) 92

vi

Table 13: Meanings of Smoking by Genders According to Relevance 93

Table 14: Association Between Rank Orderings according to Salience
and Relevance across Genders (Spearman’s Correlation Coefficients) 95

Table 15: Demographic Characteristics of Respondents and Descriptive
Statistics across Colleges 125

Table 16: Demographic Characteristics of Respondents and Descriptive
Statistics across Genders 126

Table 17: Factor Loadings of the Retained Items for Male and Female Students 127

Table 18: Results of Invariance Testing Between Genders 128
vii
ABSTRACT
This dissertation was grounded on The Theory of Meanings of Behavior to
investigate the meanings of smoking among the under-representated Asian American
college students. The meanings that were frequently attached to cigarette use by students
and the association between those meanings and smoking behaviors were examined. This
dissertation also proposed a theoretical three-factor model for the meanings of smoking
and evaluated whether and how the proposed model matched the data.
This study was a two-year design. In the first year, qualitative individual interviews
were undertaken to obtain an in-depth understanding of college smoking. Fifty Asian
American students (25 males and 25 females) participated in the interviews. Results of
the interviews served the subsequent development of a smoking survey. In the second
year, a survey was developed and administered to two post secondary institutions in
southern California. A total of 460 Asian students completed the survey.
To determine the salient meanings of smoking, and the associations between those
meanings and smoking behaviors, the percentage of the sample who endorsed each
meaning and the odds ratio of each meaning to students’ actual smoking behaviors were
calculated. Horizontal and vertical comparisons were made to compare similarities and
viii
differences across genders and acculturation patterns. Confirmative Factor Analyses
and invariance testing were conducted to evaluate the factor structure of the meanings of
smoking across genders.
Results of this study showed that personal and social meanings were mostly
endorsed by students and strongly associated with smoking behaviors. Important
differences of acculturation and gender on the meanings of smoking were revealed. These
results suggest that the meanings are powerful determinants of college smoking in Asian
populations, and a culturally appropriate smoking education curriculum tailored to
different genders are necessary.
1
CHAPTER 1
INTRODUCTION
Problems to be investigated and the significance of these problems
Negative consequences of cigarette smoking and the increases in college smoking
rates
Cigarette smoking plays a major role in the mortality of the U.S. population
(Centers for Disease Control and Prevention [CDC], 2004). It leads to serious health
problems, including an increase in the number and severity of respiratory illnesses,
decreased physical fitness, adverse changes in blood cholesterol levels, chronic heart
disease, stroke, and cancers of the lung, mouth, pharynx, esophagus, and bladder (Yang,
et al., 2004; Tonstad & Svendsen, 2005; Wen, et al., 2004; Robbins, et al., 2001). The
offspring of women who smoke during pregnancy may have increased risk for perinatal
mortality and sudden infant death syndrome. Postmenopausal women who smoke
regularly tend to have lower bone density and have an increased risk for fracture than do
women who do not smoke (Broussard & Magnus, 2004).
In order to prevent and control smoking, a number of smoking prevention programs
and control strategies have been developed and implemented over several decades. As a
2
result, the overall prevalence of smoking in the U.S. seems to be stabilized. During the
past 11 years, the overall smoking rates have declined from 26.0% in 1993 to 20.9% in
2004 (Johnston, et al., 2004; CDC, 2005). However, when examining population
subgroups, the smoking rates among college-age groups (aged 18-24 years) are on the
rise while other age groups continued to decline (CDC, 2004). Data from a large
nationally representative sample of U.S. college students (Johnston, et. al., 2005b)
indicated that 36.7 % of students had smoked in the past year, 24.3% had smoked in the
past 30 days, and 14% had smoked cigarettes every day in the past 30 days. Smoking
rates in this population are far above the 2010 national health objective- reducing
smoking rates to 12% or below (USDHHS, 2000).
The high prevalence in college smoking represents a significant “national concern”
(Wechsler, et al., 1998) and has received increased attention. College students are in the
transition from adolescence to early adulthood. They are in a decisive time in which their
behaviors can be changed, modified, and consolidated into lifetime patterns (Emmons et
al., 1998). Thus, it is important to investigate the determinants of college smoking and
develop control strategies for this group of people.

3
Need for research on determinants of smoking for Asian Americans at college level
Asian American is the fastest growing ethnic group in the U.S. Between 1990 to
2000, this population increased by 5.0 millions, or 72%, faster than the total U.S.
population (13%, Barnes & Bennett, 2002). Asian Americans represent 7.2% of the U.S.
college populations. Due to the fact that Asian Americans tend to be concentrated in
metropolitan areas (Barnes & Bennett, 2002), such as New York and Los Angeles, the
proportion of Asian college students to the population in these areas may be much higher.
Like the smoking trend for the overall U.S. college populations described above, the
smoking rates among Asian American students are on the rise although the prevalence of
smoking for the general Asian population has declined. According to Rigotti and his
colleagues (2000), smoking rates among Asian American college students rose from
16.7% in 1995 to 21.9% in 1999. Their smoking rates are expected to remain high due to
the following observations:
First, Asian Americans’ smoking rates rise markedly during adolescence. The
American Legacy Foundation (2004) reports that smoking rates for Asian Americans rose
steeply from 24% in middle school to 40% in high school. Among smokers who smoked
at least one day in the past 30 days, 26.9% of Asian Americans smoked more than 6
4
cigarettes per day (The American Legacy Foundation, 2001). This rate was higher than
rates for Hispanics (19.0%) and African Americans (22.6%). When these cohorts reach
college, the smoking rate on college campuses will probably increase.
Second, recognizing that Asian Americans represent millions of people in the U.S.
and Asian American community have increased consumer purchasing power compared to
other ethnic groups, tobacco companies and their marketing research firms have
developed specific promotional strategies to target Asian Americans (Muggli, et al.,
2002). A wide range of marketing strategies have been used over the years, including
conducting product promotions during community celebrations, using Asian American
models in advertisements, distributing tobacco products and samples at bars, nightclubs,
and college social events, and offering promotional materials to the retailers, along with
monetary incentives for cooperating with the promotional programs of the manufacturers
(Muggli, et al., 2002).
Studies comparing advertising in Asian, Hispanic, African American or White
neighborhoods reported that cigarette advertisements were three to five times more
prevalent in Asian American communities than in White communities (Luke, et al., 2000).
In addition, Asian stores were significantly more likely to have tobacco advertising
5
outside stores than other ethnic stores (U.S. Department of Health and Human Services,
1998). A similar study, using data from a large national sample of college students
reported that Asian American college students have the highest rates of exposure to
tobacco promotional activities at bars, nightclubs, and campus social events (Rigotti, et
al., 2005).
Students’ exposure to tobacco promotions has been shown to directly lead to the
initiation or progression of tobacco use (Katz & Lavack, 2002; Rigotti, et al., 2005).
Therefore, Asian American college students are at high risk of developing smoking habits.
In order to decrease and stabilize the prevalence of smoking and counter the effects of
tobacco marketing, it is necessary to develop a theory-based smoking prevention program
targeting Asian college populations.
Need for research to understand the role of acculturation in smoking
Acculturation has been defined as cultural changes which result when groups of
people having different cultures come into continuous, first-hand contact with one
another (Redfield, Linton, & Herskovits, 1936). The acculturation process was once
thought to be a uni-dimensional course of cultural change—a cultural group progresses
from a native or tradition-oriented state through a transitional stage to an elite
6
acculturated stage (Trimble, 2003). According to this view, group or individual changes
proceed away from one’s own life way in a linear manner and culminate in the full and
complete assimilation of another culture’s life way. However, more contemporary social
researchers claim that acculturation is not a uni-linear but a multi-linear process, because
group or individual changes occur in more than one direction, not only toward
assimilation (Phinney, 2003; Berry, 2003).
Berry’s multi-linear model of acculturation (2003) was widely used by most
contemporary social researchers. He suggested four directions or acculturation patterns
that immigrants might progress toward. When individuals do not wish to maintain their
cultural identity and seek daily interactions with the dominant culture, they are
proceeding towards assimilation. In contrast, when individuals place a value on holding
on to their original culture and, at the same time, wish to avoid interacting with the
dominant cultural group, they choose the separation alternative. When people have an
interest in both maintaining their original culture and interacting with the dominant
culture, they are proceeding towards integration. Finally, the direction of marginalization
is chosen when there is little possibility of, or interest in, cultural maintenance and little
interest in having relationship with the mainstream culture.
7
The impacts of acculturation on health-related behaviors are significant. Research
has demonstrated that acculturation to the American culture, or assimilation, was a risk
factor for unhealthy behaviors, including smoking, alcohol use, drug abuse, violence,
suicidality, eating disorders, and high-risk sexual behaviors (Jennings, et al., 2005;
Epstein, et al., 2001; Song, et al., 2004; Caetano et al., 2004; Kaplan, et al., 2002).
It is interesting that studies examining the association between acculturation and
cigarette smoking demonstrated opposite results among adolescents and adults (Ma, et al.,
2004; Hahm, Lahiff, & Guterman, 2003; Chen, et al., 1999a; 1999b). Among adolescents,
acculturation to the U.S. cultures seems to be a risk factor for smoking. For instance,
Unger, et al (2000) used English language usage as an indicator of acculturation and
examined the relationship between English language usage and smoking among Asian
American youth. They found that Asian adolescents who spoke only English at home had
approximately twice the risk of lifetime smoking as did the adolescents who spoke only
another language at home (OR = 1.94, p< .05).
However, when examining Asian American adults, a completely opposite pattern
emerges—acculturation to the US no longer appears to be a risk factor, but may be a
protective agent against smoking. Higher involvement in American cultures may lead to
8
lower likelihood of tobacco use. A logistic regression analysis (Ma, et al., 2004)
revealed that Asian adults who lived in the US fewer than 5 years, who were less
acculturated to the US, or who were foreign born, were more likely to 1) have family
members smoke in the home, and 2) allow visitors to smoke in the home.
Current studies on the association between acculturation and smoking in Asian
populations overwhelmingly focused on adolescents or adults aged over 25 years.
College populations who are in the transition from adolescence to adulthood are
understudied. It is important to investigate the association between acculturation and
smoking among Asian college students not only because their smoking rates are expected
to increase but also because different acculturation patterns may generate different
impacts on smoking behaviors. Understanding the ways that different acculturation
patterns influence cigarette smoking helps the development of culturally appropriate
smoking education programs for Asian American students and prevent negative
consequences brought by smoking.
Need for research to understand gender differences on smoking
Cigarette smoking was consistently higher among college males than females over
the past decade (Johnston, et al., 2005b). The gap of smoking prevalence between males
9
and females has narrowed since 1999. In 2003, the smoking prevalence between genders
exhibited very little difference, with 21.9% for males and 22.8% for females. However,
the gap enlarged again in 2004 because the use by male students increased to 29.8% and
female students declined to 21.1% (Johnston, et al., 2005b). When examining ethnic
groups on college campuses, gender differences were also found. For Hispanic, Asian,
and African Americans, male students smoked more than females. Asian groups
demonstrated the largest gap of smoking prevalence between genders (25.3% for men,
19.5% for women). In contrast to the minority groups, the prevalence among White
female students was slightly higher than male students (31.9% and 30.4% respectively,
Rigotti, et al., 2000).
Gender was found to play the role of moderator in the association between cigarette
smoking and certain factors such as self-esteem, stress, and acculturation levels (McKee,
et al., 2003; Whalen, et al., 2001; Koval, et al., 2000; Crump, et al., 1997). For instance,
research has shown that the associations between stressful life events (i.e., death of a
relative, divorce, personal financial problem, life-threatening illness) and smoking status
differed by genders (McKee, et al., 2003). When facing personal financial problems,
women were more likely than men to smoke cigarettes. Men were more likely than
10
women to quit smoking when experiencing life-threatening illness. Among
Korean-Americans, the effects of acculturation to the U.S. culture on smoking behaviors
were more significant in women than in men (Lee, Sobal, & Frongillo, 2000). For women,
those who were more acculturated to the U.S. culture smoked more. Among women,
higher familiarity with American culture and greater American social participation were
associated with more smoking. Conversely, higher familiarity with Korean culture and
greater Korean social participation were associated with more smoking in men (Lee,
Sobal, & Frongillo, 2000).
As to the influence of gender on smoking, it was suggested that Asian men and
women may have different cultural values or motivations, such as differences in sex-role
expectation or morality, to initiate and maintain their smoking habits (Lau, et al., 2003;
Lai, et al., 2004; Hsia & Spruijt-Metz, 2003). It probably was these differences that led to
the different patterns of smoking (Weiss, et al., 2006; Hsia & Spruijt-Metz, 2003;
Abernathy, et al., 1995). However, little is known about these cultural difference and the
associations between these cultural differences and smoking. It is critical to examine
gender differences on patterns of cigarette use and the underlying mechanisms that lead
to initiation or maintenance of smoking behaviors for Asian men and women. This
11
information is very important to the development of a smoking prevention program
which effectively reduces tobacco use in this population.
Limited effectiveness of currently used health behavioral models and a need
for a new perspective to augment existing smoking prevention approaches
The majority of school-based smoking prevention programs which have been
developed and implemented during the past three decades are based on the three
cognitive models of behaviors: the information deficit, affective education, and social
influence models. The “information deficit model” assumes that young people will
refrain from smoking if they are supplied with accurate and adequate information
demonstrating the negative consequences of cigarette smoking. Prevention programs
based on this model attempt to instill accurate knowledge of smoking in youth. The
“affective education model” believes that personal factors, such as reduced levels of
self-worth, and poor attitudes toward family and school, are the root causes of cigarette
smoking. Therefore, enhancing students’ self-esteem or teaching them stress management
skills are necessary to prevent smoking. The “social influence model” emphasizes the
strong effects of psychosocial factors (i.e., peer pressure, parent’s attitudes toward
smoking, perceived smoking prevalence) on cigarette use. Subsequent strategies
12
suggested by this model include equipping youth with specific skills (i.e., refusal and
communication skills) and deconstructing social myths of smoking, for instance, by
emphasizing the fact that a relatively small proportion of people are regular smokers
(USDHHS, 1994).
A wide range of evaluation results showed that many of these programs, whether
using one of the three models alone or combining the three models, had significant
short-term effects on reducing smoking prevalence and delaying smoking initiation, but
these effects tended to fade over time. Generally effects lasted for about 2 years (Institute
of Medicine, 1994). The high prevalence of lifetime smoking (53.7%) among 12
th
graders
(Johnston, et al., 2004) and the increase in smoking on college campuses reflect the
limitations of recent elementary and secondary school-based smoking prevention efforts.
Furthermore, even when smoking prevention programs based on the social influence
model are effective, there remains uncertainty as to why these programs work (Sussman,
2001).
One possible reason of the limited effectiveness of these cognitively oriented
programs is that they may fail to reflect the most critical determinants of smoking in
adolescents and young adults (Spruijt-Metz, 1999). Spruijt-Metz (1999) pointed out that
13
health-related knowledge, communication skills, attitudes toward smoking, and
self-efficacy do not necessarily determine smoking behaviors. Her suggestion was
supported by current evaluation studies, demonstrating that smoking rates among
students was not reduced with the improvement of student’s cigarette refusal self-efficacy,
health-related knowledge, and communication skills (Lantz, et al., 2000).
Another reason that may explain the lack of effectiveness of cognitively oriented
programs may be related to the ignorance of multicultural issues. Student populations in
schools are becoming increasingly ethically diverse in many areas of the U.S. today.
Research has shown that different cultures have different norms, values, and acceptability
of smoking, and these cultural factors play a strong role in the prevention or development
of smoking behaviors (Spruijt-Metz, et al., 2004, 2005; Johnson, et al., 2004; Hsia &
Spruijt-Metz, 2003). However, the majority of the large-scale smoking prevention
curriculum implemented in schools targets the mainstream White cultures and overlook
the multicultural issue (Johnson, et al., 2005).
In the light of the information that cognitive models may fail to reflect the most
critical determinants of smoking and current smoking prevention programs tend to ignore
the multicultural issues, subsequent modifications of currently used models and the
14
incorporation of multicultural factors are needed in order to develop a more effective
school-based smoking prevention curriculum for a population with diverse cultures
(Spruijt-Metz, et al., 2005; Johnson, et al., 2005).
The new perspective: the Theory of Meanings of Behavior
The Theory of Meanings of Behavior, developed by Spruijt-Metz (1999), attempted
to augment existing cognitively oriented behavioral models and provide new insights into
the field of determinants of behavioral research. This theory contends that adolescents
and young adults tend to imbue health-related behaviors with affective meanings and that
these meanings then become important determinants of those behaviors. The affective
meanings reflect one’s need of emotional balance or psychological comfort. In addition,
meanings attached to a behavior are often disconnected with any available knowledge
concerning that behavior. For instance, adolescents recognize that sweets or fried foods
are related to obesity and cardio-vascular diseases. However, they are likely to overlook
this knowledge and continue to eat a lot of sweets or fried foods because these foods may
represent a way of comforting themselves, rewarding themselves, or dealing with
frustration and stress (Weiss, et al., 2006).

15
The development of affective meanings is related to personal experiences in
families, communities, societies, and cultures. Thus, affective meanings are personal and
not fixed in time or space. They can be built, modified, and changed, and in turn, change
people’s behaviors. For example, a person who lives in a society where smoking is
perceived as a great way to establish good relationship with others may tend to smoke
because smoking means “survival and popularity” to him. However, when this person
moves to another society where smoking is not welcome and accepted, smoking does not
mean “survival and popularity” any more but switches to “stupid or selfish”. Under this
situation, he may quit smoking (Hsia and Spruijt-Metz, 2003).
Cognitively oriented models, as mentioned earlier, assume that cognitions
determine behaviors. People can behave rationally and make reasonable choices
according to objective, or scientific facts which they learn or have been told from outside
resources (Nader, et al., 1996; Spruijt-Metz, 1999). However, unlike cognitively oriented
models, the theory of Meanings of Behavior argues that for some people, particularly
adolescents and young adults, seeking immediate emotional balance or satisfying their
psychological needs (i.e., calming one’s anger, seeking the feeling of belonging) are
given the first priority and thus primarily motivate their behaviors (Spruijt-Metz, et al,
16
2004; 2005). In other words, when knowledge about a behavior contradicts the
meanings which they attach to that behavior, cognition becomes secondary to their
affective personal meanings. These meanings are the primary motivators for that
behavior.
Emphasizing affective meanings of behavior, psychological needs, and emotional
balance make this The Theory of Meanings of Behavior unique and distinguish it from
cognitive models. This theory provides a new perspectives on human behaviors and a
solution for the problems that recent school-based smoking prevention program are
facing.
Need for research on the associations between meanings of smoking and smoking
behaviors among Asian-American college students
Many interesting findings have been found when examining the associations
between meanings of smoking and smoking behaviors (Spruijt-Metz, et al., 2004; 2005).
Spruijt-Metz and her colleagues (2004) interviewed children in middle schools
concerning the reasons why children smoke. They counted the times each meaning was
mentioned by children and examined the associations between these meaning and
children’s actual smoking behavior across ethnic groups. Their results showed that
17
meanings which significantly related to smoking behaviors differed across ethnicities.
“Smoking helps me forget my problems” was most frequently mentioned by and strongly
related to lifetime smoking among Hispanic children. Among Asian students, “I don’t
want to make another person smoke alone” had high endorsement rates and significant
association with lifetime smoking. White students who endorsed “smoking helps me
study” or “Smoking gives me more energy” were more likely to try smoking than those
who did not.
Hsia and Spruijt-Metz (2003) examined meanings of smoking among newly
immigrated Asians and conceptually derived three types of meaning of smoking: personal,
social, and functional meanings. Personal meanings, originally conceptualized by Tappe
(1992) and re-conceptualized by Spruijt-Metz (1995), help the regulation of inner
affective states, such as reducing stress, forgetting annoying problems, or being
independent. Functional meanings, conceptualized by Jessor (1984) and Perry (1999),
were primarily associated with achieving physical or environmental goals, such as
keeping awake at night to prepare school exams. Social meanings (USDHHS, 1994) are
associated with achieving concordant rapport with others or the society, such as pursuing
the feelings of belonging and popularity. Researchers further developed the Meanings of
18
Smoking Index for middle school students and examined the factorial structure of
meanings of smoking in this population (Spruijt-Metz, et al., 2004; 2005). In addition to
personal, social, and functional meanings, they introduced an additional factor-weight
concern-into the structure of meanings of smoking. Of the four types of meanings,
personal and functional meanings were highly predictive of lifetime smoking among
middle school students.
Current research on meanings of behavior or smoking began to influence the
direction that a health promotion program takes. Programs incorporating meanings of
health-related behaviors have received preliminary success. Adolescents’ nutritional
practices were successfully changed by helping them develop positive meanings for
dietary behaviors (Spruijt-Metz, 1999). The amount and degree of physical activities
were also enhanced when adolescents imbued physical activities with positive meanings
(Spruijt-Metz, 1999; Tappe, 1992; Duda, et. al., 1989).
However, research on meanings of behavior or smoking overwhelmingly focuses on
adolescents. For college students, however, the reasons why they think that they smoke,
the relationships between these reasons and actual smoking behaviors, and the factorial
structure of meanings of smoking are not clear. This is worthy to be explored because this
19
information could not only advance our knowledge on college smoking but could also
guide the intervention efforts to more effective directions.
Purpose of the study
In the light of the need for research on determinants of smoking among Asian
Americans at college level, the present study attempted to investigate the meanings of
smoking among the underserved Asian American college students. Specific aims of the
present study included:
1. Meanings of smoking frequently endorsed by students.
2. The associations between the meanings endorsed by students and their actual
smoking behaviors.
3. Acculturation and gender differences, if any, in meanings of smoking.
4. Whether and how the factor structure of meanings of smoking corresponded
with the proposed theoretical three-factor structure in Asian American college
populations.
5. If so, whether the proposed three-factor structure was stable across genders.

20
Organization of the study
Due to the wealth of information from our data, the present study was divided into
three papers which are presented in chapter 2 through chapter 4.
Chapter 2 presents the first paper, entitled “Acculturation and meanings of smoking
among Asian American college students.” The purposes of this paper were to examine 1)
meanings of smoking frequently endorsed by Asian American college students, 2) the
associations between student’s actual smoking behaviors and meanings endorsed by them,
and 3) differences in meanings of smoking according to acculturation patterns, if any,
among Asian American college students.
Chapter 3 presents the second paper, entitled “Gender differences in smoking and
meanings of smoking in Asian-American college students.” This paper aimed to explore
gender differences, if any, in smoking rates, meanings of smoking, and association
between smoking and specific meanings in Asian college populations.
Chapter 4 presents the third paper, entitled “Factor structure and invariance of the
Meanings of Smoking Index for Asian American college students.” Based on the
suggestion (Spruijt-Metz, et al., 2004, 2005; Hsia & Spruijt-Metz, 2003) that the
meanings of smoking behaviors can be combined to yield three broad categories-personal,
21
social, and functional meanings-this paper evaluated the proposed theoretical
three-factor structure of meanings of smoking using The Meanings of Smoking Index for
Asian-American College Students (Hsia & Spruijt-Metz, unpublished). The specific aims
of this paper were to examine1) whether and how the factorial structure of meanings of
smoking corresponded with the proposed theoretical three-factor structure in Asian
American college populations, and 2) if so, whether the proposed three-factor structure
was stable across genders.
An overview of the overall research design was described below.
Overall research design
This was a 2-phase (2-year) study utilizing both qualitative and quantitative
methodologies to investigate the determinants of college smoking. In Phase 1, qualitative
individual interviews were undertaken in order to obtain in-depth understanding of Asian
college smoking. This methodology seems particularly well suited to the nature of the
questions in this study, which are discovery-oriented and deal with personal value or
meanings of smoking from the perspective of young adults. Results from the interviews
served to the subsequent development of a smoking survey for college students. In Phase
2, a survey was developed and administered to two postsecondary institutions in Southern
22
California in order to examine the relationships between meanings of smoking and
actual smoking behaviors. The two postsecondary institutions in Southern California
were selected for participation because they were culturally diverse and contained large
proportions of Asian American students.
Phase 1: Individual interviews
Fifty Asian students were recruited from the two participating sites, with 25 males
and 25 females. The interview protocol included discussions of 1) smoking experiences
and cessation efforts, 2) reasons to smoke or to refrain from smoking, and 3) salient
experiences of cultural differences on smoking between American and subject’s home
country.
A funnel-based strategy was chosen for interview discussion. In a funnel-based
interview, each session begins with a less structured approach that emphasizes free
discussion and then moves toward a more structured discussion of specific questions
(Morgan, 1997). This approach was chosen for the following reasons: 1) the relationship
between Asian American cultures and smoking behaviors is poorly understood. A less
structured part of discussion would be appropriate for the exploration of this relatively new
terrain with the participants because it helps interviewers learn what actually matters to the
23
participants themselves and guards against losing serendipitous findings (Morgan,
1997). 2) We attempted to investigate differences and similarities between cultures and
genders. A more structured discussion would be appropriate for making consistent
comparisons across groups because it encourages participants to respond to the
researcher's specific interests and makes the data easier to analyze across groups (Morgan,
1997).
The recruitment strategies included 1) posting invitations for participation on the
websites of student organizations in which a large proportion of their members were
Asian Americans, such as East Asian Languages and Cultures Undergraduate Association,
Asian-Pacific Students Outreach Program, Chinese, and Korean Student Associations,
and etc. and 2) posting flyers on the bulletin boards on campuses or in libraries. In this
way, students who were not members of any of the student associations and who may
have difficulty accessing the internet were recruited. The selection criteria for
participation included 1) subjects were 18 years old or above, 2) subjects were U.S.
citizens or residents of Asian origins, and 3) subjects were enrolled as undergraduate
students at one or both of two participating institutions. Students who were interested in
participation were asked to contact the researcher by e-mail or telephone. A screening
24
interview was undertaken to ensure if the selection criteria were met and the language
they preferred to use during interviews.
Interviews were conducted at school in conference or study rooms. The interviews
began with a general introduction followed by a review of the topic. The session ended by
answering any questions that the participants might have, and by thanking the participants.
Each participant received 25 dollar cash for reimbursement. Each session lasted for
approximately one hour and was audio taped with participants’ consents. One third of the
interviews were conducted in Mandarin and the rest were conducted in English.
The translation of the interviews, data coding, and data analysis were completed by
the bilingual moderator with assistance from a native English speaker, both trained in
qualitative data analysis. All interviews were first translated or transcribed into English by
the bilingual moderator. The English transcripts were then examined by both investigators.
Statements that seemed unclear were checked with the original tapes. In accordance with
our consent procedures, participants were contacted by e-mail or telephone to clarify tape
segments that could not be deciphered during data transcription and analysis.
Content analysis was conducted through the process of qualitative induction (Goetz
& LeCompte, 1984). Final themes and codes were derived from iterative review of the
25
transcripts and discussion with the members of research team. MAXQDA software
(VERBI Software, 2001) was used for storing, retrieving, displaying, and analyzing data.
Phase 2: The survey study
Taken together with the data from current individual interviews and the groups
interviews conducted by Hsia and Spruijt-Metz’s (2003), and existing literature related to
the meanings of smoking, the Southern California College Student Survey (consisting of
the measures in Appendices B-E) was developed. Measures on this survey included 1)
Demographics. Students were asked about their age, gender, ethnicities, marital status,
and etc. (See Appendix A). 2) Acculturation. A modified version of the Acculturation,
Habits, and Interests Multicultural Scale for Adolescent (AHIMSA) acculturation scale
(Unger, et al., 2002) was used to assess students’ acculturation patterns. Students were
asked about their preferences for friendship, food, and music. The familiarity with Asian
cultural values was also assessed (See Appendix B). 3) Smoking behaviors. This measure
was derived from a large school-based survey (USC Student Survey, Institute for
Prevention Research, 2002). lifetime use of cigarettes, smoking initiation, past-month
frequency, daily cigarette consumption, quitting behaviors were assessed using self-report
questions (See Appendix C). 4) Meanings of smoking. A 24-item Meanings of Smoking
26
Index for Asian American College Students (MSI-AACS) was used to assess meanings
that Asian American college students attached to cigarette smoking. Meanings, such as
“relaxation” and “refreshment” were included in this index (See Appendix D).
Data collection and recruitment strategies
There were two strategies to collect the survey data. The first strategy was to create
a web-based survey and distribute it to students through internet. This strategy was
selected because using the web-based survey to collect data was speedy and
cost-effective. Compared to students who participated in paper-and-pencil surveys,
students who answered surveys through the internet were more likely to self-disclose and
less likely to respond in a socially desirable way due to the feeling of safety and the sense
of distance associated with responding through the internet (Daley, et al., 2003; Erdman,
Klein, & Greist, 1995). In addition, evidence shows that web-based surveys produced
higher response rates than mailed hardcopy surveys among undergraduate students with
respect to cigarettes smoking and illicit drug use (McCabe, 2004).
When we experienced difficulty recruiting students using the internet due to lack of
infrastructure in one of the participating institutions, our secondary strategy-- going to the
classrooms and administrating paper-and-pencil surveys-- was used. This strategy was
27
time and labor consuming, but high response rates were assured (Herrero & Meneses,
2006).
The main approach to recruit students was to use electronic invitations. The
researcher posted invitations for participation on the websites of student organizations in
which a large proportion of their members were Asian Americans, such as East Asian
Languages and Cultures Undergraduate Association, Asian-Pacific Students Outreach
Program, Chinese American Student Association, Accounting Society, Taiwanese,
Japanese, and Korean Student Associations, and etc. The researcher also distributed flyers
to students walking on campus, and posted flyers on the bulletin boards in libraries and in
schools. Students who were interested in participation could either complete the survey
online or request a hard-copy survey. In addition to the above strategy, individual
professors were approached for the permission of survey distribution and data collection
during classes when the web-based recruitment was difficult.
The selection criteria of participants in Phase 1 were identical to those in Phase 2,
including 1) 18 years or older, 2) U.S. citizens or residents of Asian origins, and 3)
enrolled as undergraduate students at one or both of the two participating institutions
during the period of data collection.
28
Data Analysis
To achieve the first three aims of the study (examining salient meanings of smoking,
the associations between meanings and smoking, and differences and similarities on
meanings of smoking across subgroups), the percentage of the sample who endorsed each
meaning and the odds ratio of each meaning to students’ actual smoking behaviors were
calculated. Horizontal and vertical comparisons were made to compare similarities and
differences across genders and acculturation patterns. To achieve the forth and fifth aims
of the study (evaluating the factorial structure of meanings of smoking across genders),
Confirmative Factor Analyses and invariance testing were undertaken. SPSS for windows
version 12.0 (SPSS Inc., 2003) and Mplus software version 3.13 (Muthen & Muthen,
2004) were used to analyze data. Details of the methodologies were described in chapter
2 through chapter 4.
Definitions
Terms used in this dissertation were stated in the following:
Asian Americans
People who have origins in any of the original peoples of Far East Asia, Southeast
Asia, or the Indian subcontinent. Regarding people with mixed ethnic origin (i.e., one of
29
the parents is a Chinese and the other one is a White), they are identified as Asian
Americans if they identify themselves as Chinese or Asian Americans. If subjects identify
as multi-ethnic, they therefore consider the Chinese or Taiwanese background as part of
their identities, and thus qualify for the study.
Newly immigrant Asian Americans
Asian Americans who are U.S. citizens or permanent residents and have been living
in the U.S. for 2 or less than 2 years.
First generation Asian Americans
When Asian Americans and both of their parents were born outside the United
States, they are defined as 1st generation Asian Americans.
Second generation Asian Americans
When Asian Americans were born in the United States but at least one of their
parents was born outside the United States, they were defined as 2
nd
generation Asian
Americans.
Third generation Asian Americans
When Asian Americans and both of their parents were born in the United States,
they are defined as 3
rd
generation Asian Americans.
30
Meanings of smoking
Meanings of smoking are defined as those that smoking symbolizes for an
individual. For instance, smoking cigarettes may represent “looking cool” The meanings
adhering to a behavior are personal and affective in nature. These meanings are primarily
affective driven, and possess affect, emotion, and value elements (Spruijt-Metz, 1999).
Meanings of smoking are suggested to be divided into three broad categories (Hsia &
Spruijt-Metz, 2003):
Personal meanings. Meanings of smoking which have to do with intra-personal
relations, such as using smoking as a way to regulate negative emotions or alleviate
unpleasant affects (Tappe, 1992; Spruijt-Metz, 1995)
Socially relevant or social meanings. Meanings of smoking which have to do with
inter-personal relations, such as using smoking as a way to building good connections
with others (Perry, 1999).
Functional meanings. Meanings of smoking which are primarily associated with
achieving physical or environmental goals, such as refreshing minds and enhancing
concentration (USDHHS, 1994).

31
Lifetime smoking rates
The percentage of students who have ever tried cigarette smoking even one or two
puffs in their lifetime.
100-cigarette smoking rates
The percentage of students who have smoked at least 100 cigarettes in their
lifetime.
Past-30-day smoking rates
The percentage of students who smoked cigarette(s) on one or more days in the past
30 days preceding the survey.
Current smoking rates
The percentage of students who have smoked at least 100 cigarettes in their lifetime
and currently smoke daily or occasionally in the past 30 days. The rates of ever smoking
and daily smoking cannot both be the same. This needs to be clarified throughout.
Acculturation
Acculturation is defined as cultural changes which occur through continuous
interactions between people with different cultural backgrounds (Redfield, Linton, &
Herskovits, 1936). According to Berry (2003), acculturation process is multidimensional.
32
Not only one direction that immigrants can progress toward but four:
Assimilation. When people attempt to replace the home cultural orientation with
the host cultural orientation.
Integration. When people attempt to retain the home cultural values and identity,
and refuse to interact with the host culture.
Separation. When people try to maintain the home culture and at the same time be
open to the host culture.
Marginalization. When people make no attempt at culture maintenance and host
culture interaction.
33
CHAPTER 2
ACCULTURATION AND THE MEANINGS OF SMOKING AMONG
ASIAN-AMERICAN COLLEGE STUDENTS

Introduction
Current smoking prevalence
Although national adult smoking rates continue to decline, the smoking rates for
college students have remained high. According to a recent report issued by the Centers
for Disease Control and Prevention (CDC, 2005), the college smoking rate was 24.3% in
2004 which was 3.5% higher than the overall adult smoking rates in the US. A
longitudinal study tracking smoking behaviors among college students revealed that a
sizable proportion of college students who have never smoked before experiment with
cigarette smoking when they enter college and students who were occasional smokers are
at risk of becoming daily smokers in college (Wetter et. al., 2004). As indicated by Wetter
et al. (2004), over the course of 4 years in college, 11.5% of nonsmokers progressed to
occasional smokers; 14.4% of occasional smokers progressed to daily smokers.

34
The surprisingly high smoking rates and rapid development of smoking
behaviors on college campuses represent a significant public health issue (Patterson, et al.,
2004; Krohn, et al, 2005; Wechsler, et al., 1998). A growing body of research has been
conducted to investigate college smoking. However, investigations have overwhelmingly
focused on White or Hispanic populations. Asian Americans, an underserved ethnic group
at high risk for smoking, remain understudied.
Asian populations need to be taken into account in college smoking research
because Asian Americans have the highest rate of college attendance in the United States
(55.9%, Day & Jamieson, 2003). In addition, their rates of cigarette use are on the rise,
from 16.7% in 1995 to 21.7% in 1999 (Rigotti, et al., 2000). The rates in Asian college
populations were about 6.6% higher than the overall Asian population (15.1%, CDC,
2002). Given the fact that smoking contributes to 7 of 10 leading causes of death in this
population (U.S. Department of Health and Human Services [USDHHS], 1998),
examination of college smoking as well as prevention of cigarette use among Asian
Americans are urgently needed.

35
Acculturation and cigarette smoking
Acculturation is a process of cultural changes which occurs through continuous
interactions between people with different cultural backgrounds. (Redfield, et al., 1936).
The acculturation process is multidimensional. The direction in which people move
depends on the extent to which people attempt to maintain their home cultural orientation
and the extent to which people attempt to adopt aspects of the host culture. According to
Berry (2003), immigrants might progress toward one of four directions or acculturation
patterns: 1) Assimilation—attempt to replace the home cultural orientation with the host
cultural orientation, 2) Separation—attempt to retain the home cultural values and
identity, and avoid to contact with the host culture, 3) Integration—try to maintain the
home culture and at the same time remain open to the host culture, or 4)
Marginalization—little interest in both home cultural maintenance and having
relationship with the mainstream culture.
During the process of acculturation, immigrants experience changes in many areas,
such as social behaviors, style of speech, and health beliefs and behaviors (Hubert, et al.,
2005; Hahm, et al., 2003; Satia, et al., 2001; Lynn, et al., 1999). These changes on the one
hand help immigrants to develop new coping skills and help them function well in
36
diverse settings (Marin & Gamba, 2003; Unger, et al., 2000, 2004). On the other hand,
these changes can negatively impact self-esteem, identity, and interpersonal relations
(Kim & Omizo, 2005; Phinney, 2003; Bhui, et al., 2005; Silverstein & Chen, 1999;
Ouarasse, et al., 2005). Research has shown that acculturation to the American culture
tends to have a negative impact on immigrants’ healthy-related behaviors, including
alcohol use, drug abuse, violence, suicidality, eating disorders, and high-risk sexual
behaviors (Jennings, et al., 2005; Epstein, et al., 2001; Song, et al., 2004; Caetano et al.,
2004; Kaplan, et al., 2002).
As to cigarette smoking, the impact of acculturation to the U.S. culture on Asian
immigrants’ smoking appears to be bifurcated. When focusing on adolescents,
acculturation to the U.S. culture seems to be a risk factor for smoking. The more the
immigrants interact with the American cultures, the more likely they are to smoke
cigarettes (Ma, et al., 2004; Hahm, et, al., 2003; Chen, et al., 1999). For instance, Price, et
al (2002) indicated that for foreign-born Asian American adolescents, the risk of getting
drunk and smoking cigarettes was about 20% lower than for adolescents born in the U.S.
If their mothers were also foreign-born, the risk was 40% lower.

37
In contrast to the findings among adolescents, acculturation to the US culture
tends to be a protective agent against smoking among Asian adults. Fu et al., (2003) used
language usage as an indicator of acculturation and examined its relationship to cigarette
smoking in Asian adult populations. Their data showed that among Chinese American
men, higher English proficiency was associated with decreased current smoking rates.
Ma and her colleagues (2004) also reported that Asian adults who lived in the US fewer
than 5 years, who were less acculturated to the US, or who were foreign born, were more
likely to have family members smoke in the home and allow visitors to smoke in the
home.
Although some information is available about the effect of acculturation on
cigarette use for Asian adolescents and adults, little research has examined this
association among college students, those who are in the transition from adolescence to
early adulthood. It is important to understand smoking in college students not only
because their smoking rates are increasing but also because they are in a decisive time in
which their behaviors can be changed, modified, and consolidated into lifetime patterns
(Emmons et al., 1998). Understanding the role of acculturation in the development of
cigarette smoking among college students can help researchers or practitioners to design
38
culturally appropriate smoking education programs to prevent or stop smoking
behaviors on our culturally diverse college campuses.
The Theory of Meanings of Behavior
School-based smoking prevention programs implemented during the past three
decades were mostly based on cognitive models of behavior, including the information
deficit, affective education, and social influence models. These programs successfully
delayed smoking initiation and reduced smoking rates among children, but the impacts of
these programs tended to fade over time (Peterson, et al., 2000; Institute of Medicine,
1994). As to the lack of long-term effects, researchers pointed out those cognitively
oriented models assume that cognitions determine behaviors. People can behave
rationally and make reasonable choices according to objective or scientific facts which
they learn or have been told from outside resources (Nader, et al., 1996; Spruijt-Metz,
1999). However, this is not always the case (Spruijt-Metz, 1995). For adolescents and
young adults, immediately satisfying their psychological needs and adhering to their
personal meanings primarily motivate and determine their behaviors (Spruijt-Metz, et al,
2004; 2005). Thus, researchers indicated that the cognitively oriented models may fail to
reflect the most critical determinants of smoking (Spruijt-Metz, 1999).
39
Recognizing the problems that cognitively oriented models might involve,
Spruijt-Metz (1995) developed The Theory of Meanings of Behavior to augment the
existing cognitive models of behavior. This theory assumes that for some people,
particularly adolescents, meanings rather than cognitions determine their health-related
behaviors. Adolescents and young adults tend to imbue health related behaviors with
personal meanings. When these meanings are intrinsic, affective, emotional in nature,
they form a strong bond with those behaviors and become main determinant of those
behaviors. When knowledge that they receive contradicts their personal meanings and
psychological needs, cognition becomes secondary to their personal meanings and
psychological needs. These meanings bypass rationality and cognition to motivate
behaviors. Emphasizing affective determinants of behavior, psychological needs, and
personal meanings make this theory unique and distinguish it from cognitive models, and
may provide an alternative approach for school-based smoking preventions programs.
According to this theory, personal meanings are not fixed in time or space. They can
be built or modified through personal experiences, education, or under the influence of
the large environment, which in turn, change youth’s behaviors. Therefore, strategies
proposed by this theory to prevent risk behaviors is to help them build positive meanings
40
of a specific behavior (e.g., saying NO to smoke is cool and smart), and disengage
meanings from a specific risk behaviors by offering alternative healthy behaviors that can
assume this important meaning (e.g., doing exercise is a more effective way than
smoking to deal with stress). Results of several recent interventions based on the Theory
of Meanings of Behavior have shown that adolescents’ nutritional practices were
successfully changed by helping them develop positive meanings on dietary behaviors
(Spruijt-Metz, 1999). The amount and degree of physical activity was also enhanced
when adolescents imbued physical activities with positive meanings (Spruijt-Metz, 1995;
Tappe, 1992; Duda, et. al., 1989).
Studies on cigarette use also highlight a significant relationship between meanings
and smoking behaviors. Spruijt-Metz and her colleagues (2004) administered the
Meanings of Smoking Index in 24 middle schools in the greater Los Angeles area and
found a significant association between meanings on this Index and students’ actual
smoking behaviors. An overseas study on cigarette use among Chinese adolescents in
seven cities across China demonstrated the same pattern that positively valenced
meanings of smoking were highly predictive of the number of cigarettes smoked per day
in the past month (Weiss, et al., 2006). In addition, they found that meanings which were
41
related to smoking behaviors differed across ethnicities or cultures, suggesting that the
Theory of Meanings of Behavior may apply across cultures and meanings associated with
cigarette use may be culture-specific.
Study goals
The present study adopted the methodologies used by Spruijt-Metz and her
colleagues (2004) and extended their investigations on meanings of smoking to the
underserved Asian-American college student community. This study explored 1)
meanings of smoking frequently endorsed by Asian-American college students, 2) the
associations between these students’ actual smoking behaviors and the meanings that they
endorse, and 3) acculturation differences, if any, in meanings of smoking. Results of this
study may help in the development of appropriate and effective guidelines for smoking
prevention efforts for Asian-American college students.

Method
Sample
Participants in this study were 623 college students from two postsecondary
institutions in Southern California. The institutions were selected because they were
42
culturally diverse and contained large proportions of Asian American students. In
addition, research indicates that smoking patterns may differ according to college types
(Wechsler et al., 1998). In order to investigate the difference, two types of postsecondary
institutions were selected: a 4-year private university, and a 2-year community college.
The Format of the Survey
The survey distributed to students had two formats: one was web-based survey, and
the other one was conventional paper-and-pencil survey. The content of the surveys were
exactly the same. The web-based survey mode was used to collect data because it is
cost-effective and does not jeopardize a teacher’s instructional time and student’s learning
time. More importantly, the similarity of the results derived from the web-based and
paper-based survey modes on drug use among undergraduate students was affirmative
(McCabe, 2004).
However, considering 1) some students may have difficulty to access the internet, 2)
some student organizations that were key to recruitment of Asian American students for
this study did not have their own webpage which made the strategy of recruiting people
via internet difficult, our secondary strategy was going to the classrooms and
administrating paper-and-pencil surveys.
43
Student Recruitment
In the 4-year private college, the information technology services providing
networking, computing, and telecommunication resources are fully developed. The student
organizations are active. Most of them have their own web sites and send out newsletters
to their members frequently. In addition, most of the schools have student list serves and
strong computer networks connecting students and schools. All of these made our
recruitment easier and more efficient. The recruitment strategies in the private college
included 1) posting invitations for participation on the websites of student organizations of
which a large proportion members are Asian American, such as East Asian Languages and
Cultures Undergraduate Association, Asian-Pacific Students Outreach Program, Chinese
American Student Association, Accounting Society, Taiwanese, Japanese, and Korean
Student Associations, etc.; 2) distributing flyers to students walking on campus, and
posting flyers on bulletin boards in libraries and individual schools. In this way, students
who were not members of any of the student associations and who may have difficulty
accessing the internet could also be recruited. Students who were interested in participation
could either complete the survey online or request a hard copy of the survey. The targeted

44
number of subjects (300) was reached without going to the classrooms to administer
paper-and-pencil surveys.
Recruiting students through internet was difficult in the 2-year community college
because their student organizations frequently did not have their own on-line webpage.
Moreover, although accesses to students’ school email addresses was provided, these
school addresses were not their primary email addresses, making it inefficient to contact
students by using their school email addresses. Therefore, we approached individual
professors for the permission to distribute our survey and collect data during class in
addition to recruitment of students by sending electronic invitations to student
organizations and distributing flyers to students walking on campus.
Procedures
All of the participants in the 4-year private college completed the surveys through the
internet. Students who were interested in participation could click or type the link provided
on the electronic invitations or the flyers. After clicking the link, the Information Sheet
came up on the screen, which served as Consent Form. Students were asked to read the
Information Sheet thoroughly and they were encouraged to ask questions by emailing or
making phone calls to the researcher before they preceded with the survey. Another link
45
was provided at the bottom of the Information Sheet. Clicking the link led students to
the main page of the survey. Completion and return of the survey constituted consent to
participate in this research project. The information sheet clearly stated that participants
were required to be 18 or over, and no informed consent procedure was in place for parents
of minors. Therefore, data from respondents under the age of 18 were deleted. This study
was fully approved by the Institutional Review Board at both the private and the
community colleges.
In the 2-year community college, about half of the respondents completed the
surveys through the internet. Another half of them completed the paper-and-pencil survey
during classes. Students were instructed to read the Information Sheet and ask questions
before filling out the survey. Participants could withdraw at any time without
consequences of any kind.
Both the on-line and paper-and-pencil surveys took about 25 to 30 minutes. All
subjects were requested to provide email addresses. In order to reward the participants and
enhance the response rates, a lottery game was held. Cash prizes from $150 to $5 were
given to 86 winners who were chosen by computer random selection. Each winner was
informed individually via email. Winners had a 30-day period to claim their prizes. The
46
unclaimed funds were returned to the funding agency (Tobacco-Related Disease
Research Program, TRDRP).
Data were collected during a 45-day period from February to March 2005 in both
schools. In the private university, the electronic invitations and flyers were sent to 608
and 31 students respectively. A total of 322 students, excluding identified repeat
respondents, completed the web-based surveys and resulted in a 50.4% return rate. In the
community college, the response rates of the web-based and the paper-and-pencil survey
were 28.7% and 97.3% respectively. This resulted in a 49.0% response rate in the
community college and 49.7% for the entire sample. Compared with our previous work
using mailed surveys in college populations (Hsia & Spruijt-Metz, 2004), the
combination of web-based survey plus classroom administration produced a higher
response rate (49.7%) than a U.S. mail survey (15.3%).
Measures
Meanings of Smoking: A 24-item Meanings of Smoking Index for Asian American
College Students (MSI-AACS) was developed to assess meanings that were attached by
Asian American college students to cigarette smoking. In order to generate items
appropriate for this population, we first conducted individual interviews at the two
47
participating colleges, with 50 Asian students including newly immigrant, 1
st

generation, 2
nd
generation, and 3
rd
generation Asian Americans and over. Second, in order
to triangulate the findings from the current individual interviews, we also used data from
Hsia and Spruijt-Metz’s (2003) group interviews for Chinese and Taiwanese Americans,
and reviewed existing literature on meanings of physical activity, drug, alcohol, and
cigarette use (Spruijt-Metz et al., 2004). Combining results from the individual interviews,
the group interviews, and literature review, 24 meanings were generated. By adapting the
format of Meanings of Smoking Index (MSI) developed by Spruijt-Metz et al. (2004), the
24 meanings were written as 24 statements, each beginning with a stem "I smoke (or
might smoke) because: " followed by a meaning (e.g., "it helps me to fit in") Students
were asked to “check all that apply”.
Acculturation: Acculturation was assessed by a modified version of the
Acculturation, Habits, and Interests Multicultural Scale for Adolescent (AHIMSA, Unger,
et al., 2002). AHIMSA was developed for ethnically diverse adolescent populations.
which is comprised of 8 items, including 1) My favorite music is from…; 2) My favorite
TV shows are from. . . ; 3) The holidays I celebrate are from. . . ; 4) The food I eat at
home is from. . .; 5) The way I do things and the way I think about things are from. . . ; 6)
48
I am most comfortable being with people from. . . ; 7) My best friends are from. . . ; 8)
The people I fit in with best are from. . . ’’ Each of the 8 items includes four response
options: ‘‘The United States,’’ ‘‘The country my family is from,’’ ‘‘Both,’’ or ‘‘Neither.’’
In order to modify AHIMSA for college students, 50 individual interviews were
undertaken in both the private and community colleges. During the interviews, many
students, particularly 1
st
and 2
nd
generation Asian Americans, had difficulty responding to
the last three items on AHIMSA. For example, regarding item 7,“My best friend(s) are
from…,” when a respondent’s best friend was an Asian and was born and raised in the
United States, the respondent was hesitant about making a selection between “the United
States” and “ the country my family is from.” As one of the interviewees reported, “My
best friend, her family’s from Hong Kong, but um, like me, she was born and raised here
and we are very American. I’d like to say that she‘s from the United States but
considering the ethnicity, it seems I’d pick the second one (the country my family is
from). It’s really confusing…” Because the aim of these questions was to know the ethnic
background of the respondents’ best friends, the people whom they feel comfortable
being with and whom they fit in with, the response options for the three items were
modified as a list of ethnicities, including 6 Asian subgroups, 4 Hispanic subgroups,
49
Pacific Islander, Caucasian, African American, and other. “check all that apply”
strategy was used.
Participants’ responses of the three items were then recoded into “The United
States,” “The country my family is from,” “Both,” or “Neither,” by taking both the
respondents’ and their friends’/people’s ethnic backgrounds into account. For example,
when the respondents are Asian Americans, and the people whom they fit in with best are
Asian Americans only, then their answer to this item was recoded as “the country my
family is from”. When the people whom they fit in with best include both Asian and
White, then their answer to this item was recoded as “both”, and so forth.
In addition to the change of the response options for the above three items, one item
was added into the present scale—“Cultural values that I know pretty well are from...”
The response options are “The United States (e.g., freedom, independence, individualism,
etc.,)” “The country my family is from (e.g., respect parents and elders, collectivism,
etc.,)” “Both,” and “Neither.”
This modified 9-item scale was then pilot tested in a group of college students. All
items were judged to be comprehensible and distinct.

50
Smoking behaviors: Smoking behaviors were assessed using self-report questions
derived from a large survey (USC Student Survey, Institute for Prevention Research,
2002). Students were asked about their lifetime use of cigarettes, smoking initiation,
past-month frequency, daily cigarette consumption, quitting behaviors, and perceptions of
current college smoking rates. Examples of the questions are: “Have you ever tried
cigarette smoking, even one or two puffs? (Lifetime use of cigarette)” “How old were
you when you smoked your first whole cigarette? (Smoking initiation)” “Think about the
last 30 days. On how many of these days did you smoke cigarettes? (Past-month
frequency of cigarette use) Lifetime smokers were defined as students who had ever
smoked even one or two puffs, 100-cigarette smokers as those who had smoked at least
100 cigarettes in their lifetime, and past-30-day smokers as those who smoked on one or
more days during the 30 days preceding the survey. Biochemical research has established
the validity of self-reported smoking status among adult respondents (Caraballo, et al.,
2001). Therefore, we have confidence that this self-report approach accurately assessed
respondents’ smoking status.
Ethnicity: Ethnicity was assessed by a self-report question, “What is your
ethnicity?” using check all that apply strategy. There were 15 response options including
51
subgroups of Asian Americans, such as Chinese and Vietnamese Americans; subgroups
of Hispanic Americans, such as Mexican and South Americans; Pacific Islanders,
Caucasians/European Americans, and African Americans. Due to the aims of the overall
study and the corresponding recruitment strategies, samples of Hispanic Americans and
African Americans, Pacific Islanders were small, and were omitted from the present
analyses.
Marital status: Students’ marital status was measured with the question “Your
present marital status?” Response options includes 1) Single, 2) Married, 3) Living with a
partner, 4) Widowed, 5) Separated, and 6) Divorced.
Socioeconomic Status (SES): As indicated by Hagedorn and her colleagues (2002),
SES is very difficult to measure for college students because SES is compounded by a
jumble of factors such as whether students are still living with parents or other family
members, whether their parents or relatives fully or partially support student’s tuition and
living cost, whether a student receives financial assistance from schools (e.g., loan) or
other sources, or whether a student is the head of entire household. Researchers
(Hagedorn, Maxwell, Chen, Cypers, Moon, 2002) indicated that merely asking student’s
annual income may yield inaccurate information because students, particularly those who
52
are living with parents and receive financial support from parents, may not know their
parents’ income. This would tend to underestimate SES in those cases where students
earn a modest income but come from affluent family. In contrast, when students coming
from poor families are earning larger salaries, or when students are the head of household
for extended families, SES tends to be overestimated. Therefore, using income to assess
SES for college students is imperfect (Hagedorn, et al., 2002). Considering 1) Asian
cultures in which parents encourage their children to pay full attention to their studies
with no distraction from employment, and 2) research findings that of the indicators of
SES, parents’ educational levels had large and significant association to health outcomes
throughout the life span (Gellis, et al., 2005; Bradley & Corwyn, 2002), we used parents’
educational levels as the indicator of SES for college students.
Data Analysis
Descriptive statistics were run using SPSS for windows version 12.0 (SPSS Inc.,
2003). Chi-square Tests and Analysis of Variance (ANOV A) were used to evaluate
differences across colleges in age, gender, smoking status, marital status, and parents’
educational levels.

53
To understand the relationship between meanings of smoking and smoking
behaviors, we used the paradigm developed by Spruijt-Metz et al (2004) to understand
salient versus relevant meanings of smoking.
Salient meanings of smoking refer to meanings that are frequently attached by
students to smoking (Spruijt-Metz, et al., 2004). To determine the relative salience of the
meanings to the respondents, the rates of endorsement for each of the 24 meanings
(percent of the sample who endorsed each meaning) were counted first. The 24 meanings
were then rank-ordered from the highest to the lowest rates of endorsement. Each of the
participating colleges and genders were examined separately. Spearman’s correlation
coefficient (Spearman’s rho, or r
s
) was used to compare the rank orderings between
colleges and genders. Spearman’s rho is a nonparametric measure used to calculate the
correlation between two sets of rank orderings of the same variable and is often used to
compare items ranked by different judges or according to different heuristics.
Relevant meanings of smoking refer to meanings that have significant relationship
with actual smoking behaviors (Spruijt-Metz, et al., 2004). To determine the relative
relevance of the meanings to lifetime smoking, odds ratios were calculated for each of the
24 meaning first. Meanings with odds ratios significantly at the 0.05 level were then
54
ranked according to the magnitudes of the odds ratios. Each of the participating
colleges and genders were again examined separately. Spearman’s rho was computed to
make group comparisons in terms of the rank orderings by relevance.

Results
Table 1 shows the characteristics of study participants and descriptive statistics for
the 4-year private University and 2-year community college. Of the original sample of
623 participants, 163 were excluded from data analyses because they were under 18 years
old, they were not from the targeted institutions, they were identified as repeat users, or
they were not of Asian origin. The total sample in Table 1 was 460 Asian college students,
with 232 from the 4-year University and 228 from the 2-year community college. There
were no significant differences across colleges in the distribution of age, gender, and
marital status. Students from the private university tended to have parents with higher
educational levels compared to those from the community college.

55
Table 1. Demographic Characteristics of Respondents and Descriptive Statistics across Colleges
Total
(N= 460)
A 4-year
University
(N=232)
A 2-year
college
(N= 228)
ANOV A
test
χ2
test
Age [Mean (SD)] 20.9 (2.6) 21.0 (2.9) 20.7 (2.3) F= 2.1
P= 0.1
-
Gender (%)
Females/Males
55.5/44.5 51.1/48.9 59.9/41.1 - χ2= 3.6
P= 0.06
Marital status (%)
Single
Married/living with a partner
Separated/Divorced/Widowed

96.3
3.5
0.2

98.3
1.7
0

94.3
5.3
0.4

-

χ2= 5.3
P= 0.07
Parents’ educational levels
[Mean (SD)]

6.0 (2.9)

7.1 (2.8)

4.9 (2.7)

F= 70.5
P= 0.00

-
Lifetime smoking (%) 45.2 49.1 41.4 - χ2= 2.8
P= 0.10
100-cigarette smoking (%) 17.6 17.5 17.7 - χ2=0 .001
P= 0.97
30-day smoking (%) 18.7 16.8 20.6 - χ2= 1.1
P= 0.30
Although several studies indicate that smoking prevalence differs between college
types, with the public schools higher than the private schools (Wechsler, et al., 1998), our
data did not show this pattern. The lifetime smoking rates in the two participating
colleges were equivalent ( χ
2
= 2.8, P= 0.1). In contrast to previous studies, the 4-year
private college had higher rates of smoking than the 2-year community college (49.1%,
41.4%, respectively). In addition, Spearman’s correlation analyses demonstrated
comparable rank orderings of meanings of smoking between the two colleges (Table 2).
56
Due to lack of significant differences across colleges in smoking patterns and rank
orderings of meanings, we felt justified in running the subsequent analyses by combining
data from the two colleges without expecting to encounter cross-level interactions.
Table 2. Comparisons Between Rankings of Meanings of Smoking across Colleges (Spearman’s Correlation
Coefficients)
Salience Relevance
The 4-year private university vs. the 2-year
community college
.83** .52**
** r
s
is significant at the 0.01 level (two-tailed).
In the private university, 115 White students participated the on-line survey. We
included the 115 White students into our analysis in order to make comparison between
Asian and White students. The characteristics of the final sample and descriptive statistics
are presented in Table 3. Of the 460 Asian students, 56.9% belonged to Separation, 38.7%
belonged to Integration, and no one belonged to Marginalization. The assimilation group
was excluded from the analyses because the size of this group was too small (4.6%) to
justify its representativeness. Thus, the final sample were 440 Asian (262 belonged to
Separation, 178 belonged to Integration) and 115 White students. The Mean age for the
total sample was 20.9. ANOV A and Chi square tests demonstrated equivalent
distributions of age, gender, marital status, and parents’ educational levels across groups.

57
White students had the highest lifetime smoking rates (59.1%) and 30-day smoking
rates (20.0%).
Table 3. Demographic Characteristics of Respondents and Descriptive Statistics across Groups
Total
Sample
(N= 555)
Separation
(N= 262)
Integration
(N= 178)
White
(N= 115)
ANOVA
test
χ2
test
Age [Mean (SD)] 20.9 (2.7) 20.8 (2.3) 21.0 (2.8) 20.7(3.7) F= 0.4
P= 0.8
-
Gender (%)
Females/Males
57.5/42.5 60.0/40.0 57.3/42.7 46.1/53.9 - χ
2
= 8.0
P= 0.05
Marital status (%)
Single
Married/living with a partner
Separated/Divorced/Widowed

94.4
5.4
0.2

93.9
6.1
0

93.7
5.7
0.6

100
0
0

-

χ
2
= 4.3
P= 0.6
Parent educational level
[Mean (SD)]
6.5 (2.9) 6.6 (2.9) 6.2 (3.0) 6.5 (2.9) F= .81
P= .49
-
Lifetime smoking (%) 47.0 42.9 45.9 59.1 - χ
2
= 8.8
P= 0.01
100-cigarette smoking (%) 17.8 20.2 13.4 18.3 χ
2
= 3.6
P= 0.16
30-day smoking (%) 18.7 19.5 16.6 20.0 χ
2
= 2.6
P= 0.27
Salient meanings of smoking
Table 4 shows the 24 meanings of smoking and their rates of endorsement for the
three groups. For all three groups, the most salient meaning of smoking was “It helps me
to deal with stress,” Twenty eight percent of students endorsed this meaning of smoking.
In accordance with previous research (Hsia & Spruijt-Metz, 2003) suggesting that Asian
58
Americans tended to smoke for social reasons, many Asian students in our sample
perceived that they smoke or might smoke because smoking makes them feel comfortable
in some social situations. About 21% of Asian students endorsed this meaning of
smoking.
Table 4. Meanings of Smoking by Subgroups According to Salience (Rates of Endorsement %)
I smoke (or might smoke) because: Total Separation Integration White
1. It helps me deal with stress 28.1 30.1 23.4 27.0
2. It makes me feel more comfortable in social situations 21.4 24.8 15.8 19.1
3. It helps me to relax 19.6 20.2 15.8 20.9
4. It helps me calm down 16.5 17.7 13.9 15.7
5. It gives me something to do and keeps me from being
bored
16.0 12.8 13.9 23.6
6. It helps me deal with anger 13.7 14.5 12.0 12.2
7. It makes me feel more relaxed or high about things in
combination with drinking alcohol
12.8 10.3 7.0 23.5
8. It refreshes my mind 11.3 16.0 9.5 3.5
9. It puts me in better mood 11.1 10.0 10.8 12.2
10. It helps me to fit in 10.8 14.5 6.3 7.0
11. It helps me bond with others 10.4 6.4 10.1 20.0
12. It helps me to forget my problems 8.7 9.2 5.7 10.4
13. I don’t want to make another person smoke alone 8.2 10.0 5.0 9.7
14. It helps me keep my weight down 7.6 9.2 7.0 6.1
15. It makes me more cheerful, joyful, happier 6.4 10.3 1.9 2.6
16. It makes me feel I’m part of a group 5.9 6.0 4.4 7.0
17. I’m rebelling against authorities 5.6 7.8 3.2 4.3
18. It helps me concentrate 5.4 6.4 3.8 6.0
19. I don’t want to look stupid or weird 4.7 8.5 1.2 0.9
20. It makes me look good or cool 4.7 5.7 3.8 3.5
21. It makes me feel like I’m socially accepted 3.5 2.8 3.2 6.0
22. It makes me more popular 2.6 3.9 1.9 0.9
59
Table 4: Continued
23. It makes me feel closer to my boyfriend or girlfriend 1.4 1.4 1.9 0.9
24. It makes me feel like I’m making a good impression 1.2 1.4 1.9 0.0
The rank orderings of salient meanings were fairly stable across groups. Spearman’s
correlation coefficients between the three rankings were more than 0.63 and significantly
at p <0.01 between all three groups, indicating that the three rankings were significantly
correlated (Table 5).
Table 5. Association Between Group Rank Orderings of Meanings of Smoking (Spearman’s Correlation
Coefficients)
Salience Relevance
Separation vs. Integration .82** .06
Separation vs. White .63** .20
Integration vs. White .82** .29
** r
s
is significant at the .01 level (two-tailed)
Relevant meanings of smoking
Table 6 presents meanings of smoking which have significant associations (odds
ratios) to lifetime smoking. Of the 24 salient meanings, 17 meanings were significantly
related to lifetime smoking for the separation group, 11 for the Integration group, and
only 6 for White students.

60
Table 6. Meanings of Smoking by Subgroups According to Relevance (Odds Ratios)
** is significant at the .01 level (two-tailed); * is significant at the .05 level (two-tailed).
For all three groups, rankings ordered by relevance were different from rankings
ordered by salience. As shown in Table 7, all of the correlation coefficients were lower
than 0.64, with p values larger than 0.05, indicating the rankings by relevance and
I smoke (or might smoke) because: Total Separation Integration White
1. It gives me something to do and keeps me from
being bored
15.9** 19.8** 34.6** 5.5**
2. It makes me feel more relaxed or high about
things in combination with drinking alcohol
8.7** 6.1** 5.9** 28.5**
3. It puts me in better mood 5.6** 3.1** 10.9**
4. It makes me more popular 4.5* 6.4**
5. It helps me bond with others 4.5** 2.8* 6.0** 4.2*
6. It helps me calm down 4.5** 3.2** 6.8** 15.3**
7. It helps me deal with anger 4.1** 3.4** 7.8**
8. I don’t want to look stupid or weird 4.0** 5.8**
9. It helps me forget my problems 3.8** 10.5**
10. It makes me feel more comfortable in social
situations
3.7** 4.2** 3.0* 3.9*
11. It makes me more cheerful, happy, and joyful 3.6** 4.9**
12. I don’t want to make another person smoke
alone
3.5** 3.8**
13. It helps me deal with stress 3.3** 2.2** 6.5**
14. It helps me concentrate 3.3**
15. It helps me to relax 3.3** 2.8** 3.7** 6.6**
16. It makes me feel like I’m part of a group 3.2** 3.4**
17. It helps me to fit in 3.0** 3.0**
18. I’m rebelling against authorities 2.5* 3.1*
19. It helps me keep my weight down 2.0*
20. It makes me look good or cool 3.1*
21. It refreshes my mind 5.5*
61
salience were not significantly correlated. For instance, “dealing with stress” was the
top salient meaning for the separation group. However, when ranking meanings by
relevance, “dealing with stress” was no longer the top meaning of smoking. Instead,
“dealing with boredom” was the top ranking meaning. Students who attached this
meaning to smoking were about 19.8 times more likely to smoke than those who did
not. ”Dealing with stress” was ranked in the bottom of the ranking by relevance. On the
other hand, “It makes me popular” was ranked last when ranking meanings by salience
for the Separation group. However, this meaning had the fourth largest odds ratio in
relationship to smoking. Students who endorsed this meaning were 6.4 times more likely
to smoke than those who did not. Findings on the differences between rankings ordered
by salience and relevance were consistent with a prior work among adolescents
(Spruijt-Metz, et al., 2004).
Table 7. Association Between Rank Orderings According to Salience and Relevance Per Group (Spearman’s
Correlation Coefficients)
Salience vs. Relevance
(Meanings of Smoking)
P value
Separation -.27 .31
Integration -.21 .54
White .64 .17
Meanings which were related to smoking behaviors for one group may not have
significant associations to smoking for another group. For instance, “I don’t want to make
62
another person smoke alone” had a significant odds ratio to lifetime smoking for the
Separation group. Students in this group who attached this meaning to smoking were 3.8
times more likely to become a smoker than those who did not. However, this meaning
was not related to lifetime smoking for the Integration and White groups. “It helps me
deal with anger” was another example that was related to smoking for Asian group (both
the separation and integration groups) but not for White students.
Finally, rankings ordered by relevance were unstable between groups. Spearman’s
correlation coefficients between meanings rank-ordered by relevance were lower than
0.29 with p values large than 0.05 (Table 5). The top three meanings for each of the three
groups were different. For the separation group, the meaning that had the largest
magnitude of association to lifetime smoking was related to the feeling of boredom,
followed by the feeling of popular and alcohol drinking. Odds of being a smoker were at
least six-fold higher among students who attached any of the three meanings to smoking
compared with those who did not. For the Integration group, the top three meanings had
to do with dealing with boredom, dealing with bad mood, and forgetting problems.
Students who attached any of these meanings to smoking were at least 10 times more
likely to become a smoker than those who did not. For the White students, the meaning
63
with the largest odds ratio to lifetime smoking was related to alcohol drinking,
followed by “It helps me calm down,” and “It helps me to relax.” White students who
imbued smoking with any of these meanings were at least 6.6 times more likely to smoke
than those who did not.

Discussion
Intra-personal meanings, such as the feeling of boredom, alcohol use, and dealing
with a bad mood, stress, and anger, were endorsed by many students and strongly related
to smoking. Although the feeling of boredom has been reported as one of the most salient
meanings of smoking (Spruijt-Metz, et al., 2005; Freeman, et al., 2005; Hsia &
Spruijt-Metz, 2003; Lau, et al., 2003), serious attention to this meaning has not been paid
in smoking research or interventions. This might be because the association between the
feeling of boredom and actual smoking behaviors in previous studies were not explored
or the association was inconspicuous. The present study examined the relevance of each
meaning to smoking and revealed that the feeling of boredom has the strongest
association to lifetime smoking for both males and females for all three groups of
students. Our findings suggest that the feeling of boredom should be taken very seriously
64
as a risk factor, and strategies which help students to deal with boredom are needed for
smoking prevention among Asian Americans on college campuses.
The strong correlation between alcohol and tobacco use is well established (Field, et
al., 2005; McKee, et al., 2004; King & Epstein, 2005; Gulliver, et al., 1995). The present
study further demonstrated a strong relationship between these two risk behaviors across
acculturation patterns and ethnicities. For the entire sample, students who connected
“feeling enhanced high” to “smoking while drinking” were about 8.7 times more likely to
become a smoker than those who did not. The odds ratio for alcohol drinking to smoking
among White (OR= 28.5, p <0.01) was much larger than for the Separation (OR= 6.1, p
<0.01) or Integration students (OR= 5.9, p <0.01). According to our data, students
perceived that smoking while drinking enhanced the high that was not attained by using
tobacco or alcohol only. In order to enhance the high, students tried smoking together
with drinking alcohol. This concurrent use of alcohol and tobacco should be taken into
account in prevention efforts for Asian as well as White students.
Similar to prior research on meanings of smoking (Spruijt-Metz, et al., 2004; 2005;
Weiss, et al., 2006; Hsia & Spruijt-Metz, 2003; Whalen, et al., 2001), students in our
sample perceived that smoking provided a way to regulate negative emotions or alleviate
65
unpleasant affect. Personal meanings, such as dealing with a bad mood, anger, stress,
annoying problems were highly endorsed and strongly associated with smoking across
acculturation patterns and ethnicities. Students who attached any of these meanings to
smoking were at least 2.2 times more likely to become a lifetime smoker than those who
did not. For most college students, particularly new college students, school life is
challenging and stressful (Clark, 2005). These students often move away from their
parents’ home for the first time, and live independently for the first time, and some must
pay their tuition and living expenses by themselves. They also have to learn how to set
responsible personal goals, make educational choices, and make their way socially.
Emerging adulthood literature shows that a stressful transition to college can lead to
greater vulnerability (Alvan, et al., 1996) and therefore may increase the risk for illegal
drug use or smoking (Kadison & DiGeronimo, 2004; Lowery, et. al., 2005). Helping
students to adjust to college life may be necessary for an effective smoking prevention
program targeting Asian populations.
Socially relevant meanings, such as “smoking makes me more popular”, “smoking
helps me bond with others”, and “smoking makes me feel comfortable in social
situations”, also had high endorsement rates and strong associations to smoking.
66
Compared to Integration and White groups, students belonging to the Separation group
seemed to be more likely to smoke for social reasons. For instance, the association
between “Smoking makes me more popular” and actual smoking was significant for the
Separation students but not for the Integration and White students. There were also
significant associations between smoking and “I don’t want to look stupid in social
situations,” and “I don’t want to make another person smoke alone” for the Separation
students but not for the other two groups of students.
Taken together with Berry’s acculturation theory (2003) and previous observations
(Weiss, et al., 2006; Lai, et al., 2004; Hsia & Spruijt-Metz, 2003), Asian students
belonging to the Separation group tend to avoid contacting the host American cultures
and retain most of their home cultural values and customs which encourage social
smoking. Students belonging to the Integration group are open to the host American
culture where smoking is not as acceptable as it is in their home countries. This appears
to indicate that contact with the American culture may determine the importance of the
role that social meanings play in Asian smoking. It seems that the higher the degree to
which Asian students accept American cultural values concerning smoking, the lower the

67
probability that they smoke for social reasons. In other words, American culture may
play a protective agent against smoking for Asian immigrants.
The smoking rates among men in most Asian countries are higher than the rates in
America. For instance, in China, the smoking rates among men are 60.2% (Gu, et al.,
2004). It is about 36.8% higher than the smoking rates in the US (CDC, 2005). By
contrast, the smoking rates among women in Asia are lower than the rates in America. In
China, for instance, the smoking rates among women are 6.9% (Gu, et al., 2004). In
America, the smoking rates among women are about 18.5% (CDC, 2005). Many studies
reported that smoking among women is discouraged and unaccepted by the larger Asian
society, so female smokers are uncommon in most Asian countries. However, social
acceptability for smoking is roughly equal for men and women in the American culture.
It is possible that the American culture may be a risk factor for smoking for Asian
females. To understand the impact of acculturation to the American cultures on smoking
among Asian females, future research should examine the possible moderation effects of
gender on the association between acculturation and smoking in Asian populations.
Meanings with high endorsement rates may not have strong or significant
associations with cigarette use. This is consistent with previous findings (Spruijt-Metz, et
68
al., 2004) among adolescents. Taken together these findings suggest that for both
middle school and college students, the meanings of smoking which were perceived by
students as important may not be major determinants of smoking. The meanings of
smoking which were perceived by students as having less influence on their behavior
were nevertheless more strongly linked to smoking. Regarding the question of how to
best incorporate meanings in order to design an appropriate smoking prevention program,
focusing on major or actual determinants of smoking seems to be more effective.
However, as indicated by Spruijt-Metz and her colleagues (2004), ignoring what students
are perceived as important determinants of smoking “may jeopardize buy-in by the
targeted population” (Spruijt-Metz, et al., 2004, p. 204) and thus threaten the
effectiveness of smoking prevention programs. Therefore, addressing both meanings
endorsed by students and meanings significantly associated with actual smoking
behaviors might be most effective for smoking interventions.
It was also found that the associations of meanings of smoking to smoking
behaviors tended to vary across groups. Meanings which had strong influences on
smoking behaviors for one group may have limited influences on smoking for another
group, suggesting that the determinants of smoking may vary across acculturation
69
patterns and ethnicities. Conducting smoking studies under a broad “college” or a
single “Asian” category may mask differences on acculturation types or ethnicities. To
effectively prevent Asian college smoking, it is necessary to consider acculturation types
in what smoking means.
Limitations and suggestions for future research
The results of this study have to be considered within the context of its limitations.
First, meanings of smoking items were generated for Asian American college students.
The validity may be questionable when using these items in other ethnic populations or
age groups.
Second, samples were drawn from a 4-year private and a 2-year community
colleges in Southern California and therefore the generalizability of this study may be
limited by geographical correlates.
Third, in order to protect participants’ privacy, we did not use cookies (a small text
file written to users’ computers each time they visited the survey) to store information to
prevent internet users from taking a web-based survey multiple times. Therefore, it would
have been difficult to identify possible repeat participants if they provided different email
addresses. This may lead to an overestimation of the sample size.
70
Moreover, concerning the response rates, we calculated the ratio of the total
number of students whom we invited by email or flyers for participation to the total
number of the completed surveys. Although we could accurately calculate the number of
students to whom we invited, we were unable to accurately identify how many students
whom we invited by email did receive or have a chance to check our emails. This may
make the response rates skewed up or down.
Furthermore, in some instances, we had very large odds ratios exceeding 20. This
may be because we encountered small cell sizes in the 2 x 2 table while estimating odds
ratios. The small cell size may limit the precision of the estimates of odds ratios.
Finally, the present study is correlational in nature, so interpretations must be made
carefully. In order to further determine whether and how meanings predict smoking, more
advanced statistics, such as regression analyses or structural equation modeling, and
longitudinal experimental designs are needed.
Conclusions
In conclusion, the goal of the present study was to examine the meanings of
smoking in White college students and in Asian America college students with different
acculturation patterns. The results of this study showed that the personal meanings (i.e.,
71
dealing with boredom, stress or anger) and socially relevant meanings (i.e., want to be
more popular, or need to bond with friends) were highly endorsed by students and most
strongly related to cigarette smoking. It might be suggested that smoking interventions
should help students develop positive meanings that encourage health behaviors and
discourage cigarette smoking.
The results of this study also demonstrated the difference between the perceived and
actual determinants of smoking among students, and the differences on the determinants
of smoking between acculturation patterns and ethnicities. Our findings suggest that in
order to be effective, smoking prevention programs need to be culturally tailored and take
both the perceived and actual determinants of smoking into account.
A measurement tool-the MSI-AACS- for the under-representative Asian-America
college students was created by a carefully designed qualitative methodology. This tool
can contribute to future studies with respect to the interpretation and assessment of the
impact of the meanings of smoking on smoking behaviors in Asian populations. Future
research should involve longitudinal or experimental designs in order to examine causal
relationships between the meanings of smoking and smoking behaviors.
72
CHAPTER 3
GENDER DIFFERENCES IN SMOKING AND MEANINGS OF SMOKING IN
ASIAN AMERICAN COLLEGE STUDENTS

Introduction
The prevalence of smoking among college students has increased substantially
since the mid-1990s and remained high over the last few years (Centers for Disease
Control and Prevention [CDC], 2003; Rigotti, et al., 2000). Findings from the Monitoring
the Future study showed that 24.3% of college students were current smokers in 2004.
This prevalence was higher than the 22.5% in 2003. College smoking rates have also
been continually higher than the overall smoking rates in the U.S. For instance, in 2004,
approximately 20.9% of U.S. adults smoked currently, which was 3.4% lower than the
rates of smoking among college students (CDC, 2005).
The high prevalence in cigarette smoking on college campuses occurs broadly
across all ethnicities, including Asian Americans (Wechsler, et. al., 1998). However,
smoking research focusing on Asian-American college students is quite limited. Even in
studies involving large nationally representative samples of college students, Asian
73
populations are often overlooked. Although the trends and prevalence of college
smoking among Asian students are not clear, according to recent studies on adolescent
smoking and reviews of the tobacco industry’s marketing strategies, it is apparent that
Asian American students are at high risk of developing smoking habits.
Data from The National Youth Tobacco Survey indicated that the lifetime smoking
rates among Asian Americans adolescents rose steeply from 24% in middle school to
40% in high school (The American Legacy Foundation, 2004). The prevalence of
frequent smoking (smokers who reported smoking on 20 days or more in the past 30 days)
among Asian American high school students was second only to White high school
students. In addition, Asian Americans smoke more intensely than other ethnic groups. As
indicated by the American Legacy foundation (2004), the average number of cigarettes
smoked per day among Asian American students is over a pack (about 22 cigarettes).
Asian American smokers were also less likely to have tried to quit smoking (American
Lung Association, 2004). It is expected that when these cohorts move into college, they
will probably take their smoking behaviors along with them, and therefore the smoking
rates in Asian college populations may remain high in the near future.

74
A review of internal tobacco industry documents and archives revealed that
tobacco industries recently focused on Asian Americans for promoting tobacco products
(Muggli, et al., 2002). Recognizing that the Asian American community had increased
consumer purchasing power compared to other ethnic groups, tobacco companies and
their marketing research firms developed specific promotional strategies to target the
Asian-American market (Muggli, et al., 2002). A wide range of marketing strategies have
been used, including conducting product promotions during community celebrations,
using Asian American models in advertisements, distributing tobacco products and
samples at bars, nightclubs, and college social events, and offering promotional materials
to retailers in Asian American communities (Muggli, et al., 2002).
The impact of these tobacco promotions on the development of smoking habits
among college students is significant. Research has shown that Asian students who were
exposed to tobacco promotions were more likely to try smoking or continue smoking
than those who were never exposed to these promotions (Rigotti, et al., 2005). Data from
a large national sample of college students (Rigotti, et al., 2005) reported that Asian
American college students have the highest rates (10.7%) of exposure to tobacco
promotional activities at bars, nightclubs, and campus social events compared to African
75
Americans (4.6%), Whites (8.3%), and Hispanic (9.6%). Cigarette advertisements are
three to five times more prevalent in Asian American communities than in White
communities (Luke, et al., 2000). It is apparent that Asian Americans college students are
at high risk of developing smoking behaviors. A theory-based smoking prevention
curriculum is urgently needed in order to control the prevalence of smoking and counter
the impacts of tobacco promotions.
Gender differences in cigarette smoking
Adult smoking rates vary substantially across gender. In general, more men smoke
than women (CDC, 2005). In 2003, 24% of men and 19.2% of women reported current
smoking. The pattern of gender differences in smoking was also apparent across ethnic
groups (White, Black, Hispanic, American Indian/Alaska native, Asian), and Asian
American demonstrated the largest gap in smoking rates between genders, with 17.5% of
men and 6.5% of women reporting current smoking in 2005 (CDC, 2005). Although
some studies report equivalent smoking rates between genders when focusing on White
college students, we found gender differences in smoking rates among Asian Americans
on college campuses, with 14.9% female and 23.5% male students reporting current
smoking (Hsia & Spruijt-Metz, 2006, unpublished data).
76
Research has shown that gender may moderate the association between cigarette
smoking and several factors such as self-esteem, stress, and acculturation levels (McKee,
et al., 2003; Whalen, et al., 2001; Koval, et al., 2000; Crump, et al., 1997). For instance,
Lee, Sobal, and Frongillo (2000) conducted logistic models to assess whether the
relationship between acculturation and smoking was similar across Korean-American
men and women. Their results showed that the effects of acculturation on smoking
behaviors were more significant in women than in men. For women, those who were
more acculturated to the U.S. culture smoked more. For men, those who were more
oriented towards the Korean culture smoked more. In addition, they found that the
components of acculturation that were related to smoking behaviors differ by gender. In
women, higher familiarity with American culture and greater American social
participation were associated with more smoking. Conversely, higher familiarity with
Korean culture and greater Korean social participation were associated with more
smoking in men.
As to gender differences found in current smoking studies, researchers indicated
that Asian females may possess different motivations than males to initiate and maintain
their smoking habits, such as differences in sex-role expectation or morality (Weiss, et al.,
77
2006; Hsia & Spruijt-Metz, 2003; Abernathy & Bertrand, 1995). However, how
motivations differ and how they influence smoking differently in males and females
remains unclear. In order to effectively reduce cigarette use in Asian college populations,
it is necessary to obtain in-depth understanding of gender differences in the determinants
of smoking. (Madison-Colmore, 2003; Spruijt-Metz, et al., 2005; Unger, et al., 2004).
A recent developed theory- The Theory of Meanings of Behavior- highlights
differences in gender, ethnicity, and cultural values and has been suggested as an
appropriate basis for multicultural smoking interventions. The theory of Meanings of
Behavior was developed by Spruijt-Metz (1995), and is a synthesis of research on mood
(Slovic, 2001), affect (Ikard & Tomkins, 1973; Shor & et al., 1981; Perry, 1999; Perry &
Kelder, 1992), and affective neuroscience (Panksepp, 1998, 2003; Adolphs, et al., 2003).
According to this theory, the meanings of behavior are index what a specific behavior
symbolizes for an individual. For instance, smoking might be regarded as a symbol of
“popularity” or “independence” for young adults (Hsia & Spruijt-Metz, 2003). The
meanings attached to a behavior are personal and affective in nature. They often take
precedence over cognition or knowledge to determine that behavior. An example is that

78
many people continue smoking for stress relief even if they clearly know that smoking
may cause lung cancer and heart disease.
The emphasis on the personal, affective, and symbolic qualities of meanings of
behavior makes this theory unique and more able to capture differences in cultures,
ethnicities, and genders. Therefore, a growing body of research has investigated the
meanings attached to health-related behaviors such as physical activity, dietary choices,
and cigarette smoking across ethnicities, genders, and age groups (Spruijt-Metz & Metz,
unpublished manuscript; Spruijt-Metz, 1999; Spruijt-Metz, et al., 2004, 2005). Health
promotion programs also began to incorporate the Meanings of Behavior theory to
prevent risk behaviors. These programs preliminarily generated strategies specific to
different ethnicities, genders, and age groups to replace meanings which lead to risk
behaviors with those which help to avoid risk behaviors, and successfully helped people
developing healthy behaviors (Spruijt-Metz, 1995, 1999; Tappe, 1992; Duda, et al.,
1989).
Although studies on meanings of behavior or smoking are striking, they
predominantly focus on adolescents. College students, particularly Asian Americans, are
understudied. The present study extends previous investigations on meanings of smoking
79
to the underserved Asian American college student community, and examines 1)
meanings of smoking frequently endorsed by Asian American college students, 2) the
associations between these students’ actual smoking behaviors and the meanings that they
endorse, and 3) gender differences, if any, in meanings of smoking.

Method
Sample
Research indicates that smoking patterns may differ according to college types
(Wechsler et al., 1998). In order to investigate these differences, we recruited students
from two types of postsecondary institutions from southern California: a 4-year private
university, and a 2-year community college. Six hundred and twenty third college
students from the two postsecondary institutions participated in this study.
Data collection strategies
There were two strategies to collect the survey data. The first strategy was to create
a web-based survey and distributed it to students through internet. This strategy was
selected because using the web-based survey to collect data was speedy and
cost-effective. Compared to paper-and-pencil survey, participants were more likely to
80
self-disclose and less likely to respond in a socially desirable way due to the feeling of
safety and the sense of distance associated with responding in the internet (Daley, et al.,
2003; Erdman, et al., 1995). In addition, evidence shows that the web-based surveys
produced higher response rates than the mailed hardcopy surveys among undergraduate
students with respect to cigarettes smoking and illicit drug use (McCabe, 2004).
However, considering 1) some students may have difficulty to access the internet,
and 2) some student organizations that were keys to recruitment of Asian American
students for this study did not have their own webpage which made the strategy of
recruiting people via internet difficult; our secondary strategy was going to the
classrooms and administrating paper-and-pencil surveys. This strategy was time and labor
consuming, but high response rates were assured (Herrero & Meneses, 2006).
Student Recruitment
The main approach to recruit students was to use electronic invitations. The
researcher posted invitations for participation on the websites of student organizations in
which a large proportion of their members were Asian Americans, such as East Asian
Languages and Cultures Undergraduate Association, Asian-Pacific Students Outreach
Program, Chinese American Student Association, Accounting Society, Taiwanese,
81
Japanese, and Korean Student Associations, and etc. The researcher also distributed
flyers to students walking on campus, and posted flyers on the bulletin boards in libraries
and in schools. Students who were interested in participation could either complete the
survey online or request a hard-copy survey.
In the 4-year private college, the information technology services providing
networking, computing, and telecommunication resources are fully developed. The student
organizations are active. Most of them have their own web sites and send out newsletters
to their members frequently. In addition, most of the schools have student list serves and
strong computer networks connecting students and schools. All of these made our
recruitment easier and more efficient. The targeted number of subjects (300) was reached
without going to the classrooms to administer paper-and-pencil surveys.
We experienced difficulty recruiting students using the internet in the 2-year
community college because their student organizations frequently did not have their own
on-line webpage. Moreover, although accesses to students’ school email addresses was
provided, these school addresses were not their primary email addresses, making it
inefficient to contact students by using their school email addresses. Thus, in addition to

82
distributing electronic invitations and flyers, we approached individual professors for
permission to distribute our survey and collect data during class.
The selection criteria of participants included 1) 18 years or older and 2) enrolled as
undergraduate students at one or both of the two participating institutions during the
period of data collection. This study was fully approved by the Institutional Review
Board at both the private and the community colleges.
Procedures
Web-based survey: Students who were interested in participation can click or type
the link provided on the electronic invitations or the flyers. After clicking the link, the
Information Sheet came up on the screen, which served as Consent Form. Students were
asked to read the Information Sheet thoroughly and they were encouraged to ask questions
by emailing or making phone calls to the researcher before they proceeded with the survey.
Another link was provided in the bottom of the Information Sheet. Clicking the link led
students to the main page of the survey. Completion and return of the survey constituted
consent to participate in this research project. The information sheet clearly stated that
participants were required to be 18 or over, and no informed consent procedure was in

83
place for parents of minors. Therefore, data from respondents under the age of 18 was
deleted.
Paper-and-pencil survey: A trained data collector distributed the surveys. Students
were instructed to read the Information Sheet and ask questions before filling out the
survey. Participants could withdraw at any time without consequences of any kind.
Both the on-line and paper-and-pencil surveys took about 25 to 30 minutes. All
subjects were requested to provide email addresses. In order to reward the participants and
enhance the response rates, a lottery was held. Cash prizes from $150 to $5 were given to
86 winners whom were chosen by computer random selections. Each winner was informed
individually via email. Winners had 30-day period to claim their prizes. The unclaimed
funds were returned to the funding agency (Tobacco-Related Disease Research Program,
TRDRP). Data was collected during a 45-day period from February to March 2005 in both
schools.
In the private university, the electronic invitations and flyers were sent to 608 and
31 students respectively. A total of 322 students completed the web-based surveys and
resulted in a 50.4% return rate. In the community college, the response rates of the
web-based and the paper-and-pencil survey were 28.7% and 97.3% respectively. This
84
resulted in a 49.0% response rate in the community college and 49.7% for the entire
sample. Compared with our previous work using mailed surveys in college populations
(Hsia & Spruijt-Metz, 2004), the combination of web-based survey plus classroom
administration produced a higher response rate (49.7%) than a U.S. mail survey (15.3%).
Measures
Meanings of Smoking: A 24-item Meanings of Smoking Index for Asian American
College Students (MSI-AACS) was developed to assess meanings that were attached by
Asian American college students to cigarette smoking. Items were written as 24
statements, each beginning with a stem "I smoke (or might smoke) because: " followed
by a meaning (e.g., "it helps me to fit in") Students were asked to “check all that apply”.
The development of this scale has been described earlier (Hsia & Spruijt-Metz, under
review).
Smoking behaviors: Three self-report questions derived from a large survey (USC
Student Survey, Institute for Prevention Research, 2002) were used to assess student’s
smoking status. Students were asked about their lifetime use of cigarettes, “Have you
ever tried cigarette smoking, even one or two puffs?” lifetime use of 100 cigarettes,
“Have you smoked at least 100 cigarettes in your life?” and Past-month frequency of
85
cigarette use “Think about the last 30 days. On how many of these days did you smoke
cigarettes?” Biochemical research examining discrepancies between self-reported
smoking status and measured serum cotinine (a biochemical marker of tobacco
consumption) levels in adult populations provides evidence that self-reported smoking
status accurately reflects actual smoking behaviors (Caraballo, et al., 2001). Thus, we
have confidence that this self-report approach accurately assessed respondents’ smoking
status.
Ethnicity: A self-report question, “What is your ethnicity?” using “check all that
apply” strategy. The response options included subgroups of Asian Americans, such as
Chinese and Vietnamese Americans; subgroups of Hispanic Americans, such as Mexican
and South Americans; Pacific Islanders, Caucasians/European Americans, and African
Americans. Due to the aims of the overall study and the corresponding recruitment
strategies, samples of Hispanic Americans and African Americans, Pacific Islanders were
small, and were omitted from the present analyses.
Marital status: Students’ marital status was measured with the question “Your
present marital status?” Response options includes 1) Single, 2) Married, 3) Living with a
partner, 4) Widowed, 5) Separated, and 6) Divorced.
86
Parents’ educational levels: Parents’ educational levels were used to be an
indicator of student’s Socioeconomic Status (SES). As indicated by Hagedorn and her
colleagues (2002), SES is very difficult to measure for college students and merely
asking student’s annual income may yield inaccurate information because students,
particularly for those who are living with parents and receive financial support from
parents, may not know their parents’ income. This would tend to underestimate SES in
those cases when students who earn modest income but come from affluent family. In
contrast, when students coming from poor families are earning larger salaries, or when
students are the head of household for extended families, SES tends to be overestimated.
Therefore, using income to assess SES for college students is imperfect (Hagedorn, et al.,
2002). Given that Asian cultures in which parents encourage their children to pay fully
attention on study with no distraction from employment, and research findings that of the
indicators of SES, parents’ educational levels had large and significant association to
health outcomes throughout the life span (Gellis, et al., 2005; Bradley & Corwyn, 2002),
we used parents’ educational levels as the indicator of SES for college students.

87
Data Analysis
Chi-square Tests and Analysis of Variance (ANOVA) were performed to evaluate
differences across colleges in age, gender, smoking status, marital status, and parents’
educational levels. To understand the relationship between meanings of smoking and
smoking behaviors, the paradigm developed by Spruijt-Metz et al (2004) to understand
salient versus relevant meanings of smoking was utilized.
Salient meanings of smoking refer to meanings that are frequently attached by
students to smoking (Spruijt-Metz, et al., 2004). To determine the relative salience of the
meanings to the respondents, the rates of endorsement for each of the 24 meanings
(percent of the sample who endorsed each meaning) were counted first. The 24 meanings
were then rank-ordered from the highest to the lowest rates of endorsement. Each of the
participating colleges and genders were examined separately. Spearman’s correlation
analysis (a nonparametric measure used to calculate the correlation between two sets of
rank orderings of the same variable and is often used to compare items ranked by
different judges or according to different heuristics) was run to compare the rank
orderings between colleges and genders.

88
Relevant meanings of smoking refer to meanings that have significant
relationship with actual smoking behaviors (Spruijt-Metz, et al., 2004). To determine the
relative relevance of the meanings to lifetime smoking, odds ratios were calculated for
each of the 24 meaning first. These meanings were then ranked according to the
magnitudes of the odds ratios. Spearman’s correlation analysis was again computed to
make group comparisons in terms of the rank orderings by relevance.

Results
The characteristics of the participants and descriptive statistics for the 4-year private
University and 2-year community college are shown in Table 8. A total of 623 students
completed the survey. 163 of the original sample were excluded from data analyses
because they were under 18 years old (N= 1), they were not from the targeted institutions
(N= 5), they were identified as repeat users (N= 8), or they were not of Asian origin (N=
149). Our final sample included 460 Asian college students, with 232 from the 4-year
University and 228 from the 2-year community college. There were no significant
differences across colleges in the distribution of age, gender, and marital status.

89
Compared with students from the community college, parents of students from the
private had significantly higher educational levels.
Table 8. Demographic Characteristics of Respondents and Descriptive Statistics across Colleges
Total
(N= 460)
A 4-year
University
(N=232)
A 2-year
college
(N= 228)
ANOV A
test
χ2
test
Age [Mean (SD)] 20.9 (2.6) 21.0 (2.9) 20.7 (2.3) F= 2.1
P= 0.1
-
Gender (%)
Females/Males
55.5/44.5 51.1/48.9 59.9/41.1 - χ2= 3.6
P= 0.06
Marital status (%)
Single
Married/living with a partner
Separated/Divorced/Widowed

96.3
3.5
0.2

98.3
1.7
0

94.3
5.3
0.4

-

χ2= 5.3
P= 0.07
Parents’ educational levels
[Mean (SD)]

6.0 (2.9)

7.1 (2.8)

4.9 (2.7)

F= 70.5
P= 0.00

-
Lifetime smoking (%) 45.2 49.1 41.4 - χ2= 2.8
P= 0.10
100-cigarette smoking (%) 17.6 17.5 17.7 - χ2=0 .001
P= 0.97
30-day smoking (%) 18.7 16.8 20.6 - χ2= 1.1
P= 0.30
In contrast with previous studies (Wechsler, et al., 1998) showing that smoking rates
at public schools were higher than the private schools, the present study demonstrated a
parallel smoking rates between different types of schools ( χ
2
= 2.8, P= 0.1), with no
significant differences in lifetime smoking between the 4-year private college and the
2-year community college (49.1%, 41.4%, respectively). In addition, the rank orderings
90
of meanings of smoking between the two colleges were comparable (See Table 9).
Because both smoking rates and rank orderings of meanings of smoking were similar
across colleges, it was justified to run the subsequent analyses by combining data from
the two colleges without expecting to encounter cross-level interactions.
Table 9. Comparisons Between Rankings of Meanings of Smoking across Colleges (Spearman’s Correlation
Coefficients)
Salience Relevance
The 4-year private university vs. the 2-year
community college
.83** .52**
** r
s
is significant at the .01 level (two-tailed).
Table 10 shows descriptive statistics and the characteristics of the combined sample
from the two colleges. The final sample was 460 college students, with 255 female and
205 male students. The Mean age for the total sample was 20.9. The majority of the
student (96.3%) was single. ANOV A and Chi square tests demonstrated gender
differences on 100-cigarette and 30-day smoking rates, with females smoking
significantly less than males.
Table 10. Demographic Characteristics of Respondents and Descriptive Statistics across Genders.
Total
Sample
(N= 460)
Females
(N= 255)
Males
(N= 205)
ANOVA
test
χ2
test
Age [Mean (SD)] 20.9 (2.7) 20.8 (2.3) 21.0 (2.8) F= 0.05
P= 0.83
-

91
Table 10: Continued
Total
Sample
(N= 460)
Females
(N= 255)
Males
(N= 205)
ANOV A
test
χ2
test
Marital status (%)
Single
Married/living with a partner
Separated/Divorced/Widowed

96.3
3.5
0.2

96.5
3.1
0.4

96.1
3.9
0.0

-

χ2= 1.4
P= 0.60
Parents’ educational levels
[Mean (SD)]
6.0 (2.9) 5.6 (3.0) 6.4 (2.8) F= 8.1
P= 0.00
-
Lifetime smoking (%) 45.2 43.4 47.5 - χ2= 0.8
P= 0.37
100-cigarette smoking (%) 17.6 24.5 17.6- - χ2= 11.9
P= 0.00
30-day smoking (%) 18.7 14.8 23.5- - χ2= 5.6
P= 0.02
Salient meanings of smoking
The rates of endorsement of the 24 meanings of smoking are presented in Table 11.
Spearman’s correlation coefficient between meanings rank-ordered by salience was .90
with p=.00, indicating that the rank orderings of salient meanings were quite stable across
genders (Table 12). The top three meanings of smoking which were most frequently
endorsed by both female and male students were “It helps me to deal with stress,” “It
makes me feel more comfortable in social situations,” and “It helps me to relax.” Each of
these meanings had more than 17% endorsement rates.

92
Table 11. Meanings of Smoking by Genders According to Salience (Rates of Endorsement %)
I smoke (or might smoke) because: Total Females Males
1. It helps me deal with stress 28.4 29.4 27.3
2. It makes me feel comfortable being in social situations 21.9 18.0 26.8
3. It helps me to relax 19.3 17.3 22.0
4. It helps me calm down 16.7 14.5 20.0
5. It helps me deal with anger 14.1 13.3 15.1
6. It gives me something to do or it keeps me from being bored 14.1 12.2 16.6
7. It refreshes my mind 13.2 12.5 14.1
8. It helps me to fit in 11.7 12.2 11.2
9. It puts me in a better mood 10.8 10.2 11.7
10. It makes me feel more relaxed or high about things in
combination with drinking alcohol
10.2 11.0 9.3
11. It helps me to forget my problems 8.2 9.0 7.3
12. It helps me bond with others 8.0 7.5 8.8
13. It helps me keep my weight down 8.0 11.8 3.4
14. I don’t want to make another person smoke alone 7.8 7.5 8.3
15. It makes me more cheerful, joyful, and happier 7.3 5.9 9.3
16. It makes me feel I’m part of a group 5.6 6.7 4.4
17. It makes me feel I’m not stupid or weird 5.6 5.1 6.3
18. It makes me look good or cool 5.0 5.5 4.4
19. It makes me more popular 3.0 3.1 2.9
20. It makes me feel like I’m socially accepted 3.0 2.0 4.4
21. It makes me feel closer to my boy- or girlfriend 2.0 2.4 0.5
22. It helps me concentrate 5.4 4.0 7.3
23. I’m rebelling against authorities 5.9 6.7 4.9
24. It makes me feel like I’m making a good impression 2.0 1.2 2.0

Table 12. Association Between Group Rank Orderings of Meanings of Smoking (Spearman’s Correlation
Coefficients)
**r
s
is significant at the .01 level (two-tailed); *r
s
is significant at the .05 level (two-tailed).
Salience Relevance
Males vs. Females .90** .35
93
Relevant meanings of smoking
The meanings of smoking which have significant associations (odds ratios) to
lifetime smoking are presented in Table 13. Of the 24 salient meanings, 14 meanings
were significantly related to lifetime smoking for females and 16 meanings for males.
Table 13. Meanings of Smoking by Genders According to Relevance (Odds Ratios)
I smoke (or might smoke) because: Total Females Males
1. It gives me something to do or keeps me from being bored 25.7** 25.1** 25.8**
2. It makes me feel high about things in combination with drinking alcohol 6.0** 9.7** 3.4*
3. It makes me feel I’m not stupid or weird 5.5** 7.8** 4.0*
4. I don’t want to make another person smoke alone 4.7** 4.0** 5.8**
5. It makes me more popular 4.6*
6. It puts me in a better mood 4.5** 5.0** 3.8**
7. It helps me bond with others 4.2** 4.0** 4.3*
8. It helps me calm down 4.1** 3.2** 5.1**
9. It makes me feel more comfortable in social situations 3.8** 2.6** 5.5**
10. It makes me more cheerful, joyful, and happier 3.7** 3.9* 3.4*
11. It makes me feel like I’m part of a group 3.5** 3.4*
12. It helps me deal with anger 3.5** 3.2** 3.9**
13. It helps me to fit in 3.0** 6.3**
14. It helps me to relax 2.9** 2.4** 3.6**
15. It helps me forget my problems 2.9** 2.7* 3.3*
16. It makes me look good or cool 2.9*
17. It helps me deal with stress 2.8** 2.0** 4.5**
18. It helps me concentrate 2.7* 4.9*
19. I’m rebelling against authorities 2.6*
20. It refreshes my mind 2.8*
** is significant at the .01 level (two-tailed); * is significant at the .05 level (two-tailed).
Table 13 shows that meanings which are significantly related to smoking may differ
across genders. For instance, “It helps me to fit in” and “It helps me to concentrate” had
94
significant odds ratios to smoking for males but not for females. Odds of being a
smoker were at least 4.9-fold higher among male students who imbued smoking with
these meanings compared with those who did not. Similarly, “it makes me feel like I’m
part of a group” was significantly related to smoking for females but not for males.
Female students who attached this meanings to smoking were 3.4 time more likely to
become lifetime smokers than those who did not.
The rank ordering of meanings by relevance between genders were not correlated
(r
s
= .35, p > 0.05, Table 12). For examples, the top three relevant meanings for females
were related to dealing with boredom, alcohol use, and feeling of stupid or weird. The top
three relevant meanings of smoking for males had to do with dealing with boredom,
fitting in, and being unwilling to make another person smoke alone.
The rank orderings by relevance were different from the rank orderings by salience
for both males and females. Among females, the top three salient meanings were related
to the feeling of stress (29.4%), the feeling of comfort in social situations (18%), and the
feeling of relaxation (17.3%). When meanings were ranked according to the magnitude of
their association to lifetime smoking, a different picture emerged. The feeling of stress
and comfort in social situations were no longer the top meanings. Instead, the top three
95
meanings were “I smoke (or might smoke) because it gives me something to do and
keeps me from being bored” (OR= 25.1, p < 0.01), “I smoke (or might smoke) because it
makes me feel more relaxed or high about things in combination with drinking alcohol”
(OR= 9.7, p < 0.01), and “It makes me feel stupid or weird” (OR= 7.8, p < 0.01).
Among males, the top three salient meanings, identical with those for females, were
related to the feeling of stress (27.3%), the feeling of comfort in social situations (26.8%),
and relaxation (22.0%). However, the top three relevant meaning for males were once
again totally different from the top ranked salient meanings for males. The top three
relevant meanings had to do with the feeling of boredom, feeling of fitting in, and being
unwilling to make another person smoke alone. Odds of being a smoker were at least
5.8-fold higher among males who imbued smoking with any of these meanings compared
with those who did not. The Spearman’s coefficients between rankings of salience and
relevance was -.57 (p < .05) for females and 0.32 (p > .05) for males, indicating that what
students think drive their behaviors might not really drive their behaviors (Table 14).
Table 14. Association Between Rank Orderings According to Salience and Relevance across Genders
(Spearman’s Correlation Coefficients)
Salience vs. Relevance P value
Females -0.57* 0.03
Males 0.32 0.20
*r
s
is significant at the .05 level (two-tailed).
96
Discussion
Our results showed that intra-personal or personal meanings, such as “smoking
gives me something to do and keeps me from being bored,” and “smoking makes me feel
more relaxed or high about things in combination with drinking alcohol,” were highly
endorsed and strongly associated with smoking. The feeling of boredom was the top one
risk factor for smoking in our data (OR= 25.7, p < 0.01). An in-depth exploration of the
feeling of boredom experienced by Asian students is recommended and strategies which
help students to deal with boredom should be developed to prevent smoking for
Asian-American college students.
Meanings of smoking were also strongly associated with the desire for cigarettes
while drinking alcohol, and the longing for the feeling of enhanced high with concurrent
use of alcohol and tobacco. Students in our sample who attached such meanings to
smoking were 6 times more likely to smoke than those who did not. The association was
stronger among females than among males (OR= 9.7, and 3.4, respectively). As reported
by students, when they drank alcohol, they were very likely to smoke and increase
cigarette consumptions. Our findings corresponded with prior studies demonstrating the
high correlation between alcohol consumptions and increased rates of smoking (Field, et
97
al., 2005; Fenzel, 2005; McKee, et al., 2004; Wong, et al., 2004), and further point out
that the determinants of smoking may not only relate to the drug effects of alcohol but
also relate to the psychological longing for the enhanced high experienced by concurrent
use of alcohol and cigarettes. Given the strong associations between cigarette and alcohol
use in college populations, smoking prevention programs need to address the
co-occurrence of drinking and smoking behaviors as well in Asian college populations for
both genders.
In addition to personal meanings, socially relevant meanings, such as “smoking
makes me feel I’m not stupid or weird,” “I don’t want to make another person smoke
alone,” and “It makes me feel more comfortable in social situations,” also had high
endorsement rates and strong associations with smoking. The results of the study
correspond with a previous suggestion that socially relevant meanings may be decisive
for smoking among Asian men (Hsia & Spruijt-Metz, 2003). Male students in our sample
who attached social meanings to cigarette use were at least 3.8 times more likely to try
cigarette smoke than those who did not. For the first few years on college campuses
students have to adapt themselves to the new environment and learn new systems. One of
the important things that they have to learn is how to find social support, meet new
98
people, and make friends. Moreover, Asians tend to be collectivistically oriented
(Triandis, 1995; Hwang, et al., 2003) and are susceptible to influenced by their culture
which traditionally encourages people, particularly males, to smoke for social reasons
(Hsia & Spruijt-Metz, 2003; Weiss, et al., 2006). Therefore, under the influences of the
new college environment and Asian cultures, smoking and “fitting in with a new group of
people,” “bonding with others,” or “feeling comfortable in social situations” may be
easily and tightly connected. This may explain the high associations between smoking
and social meanings among Asian American students.
In contrast to Asian males, Asian females were expected to be less likely to endorse
social meanings, as observed in several previous studies (Hsia & Spruijt-Metz, 2003;
Weiss, et al., 2006). According to these studies, Asian women feel that Asian society
welcomes smoking among men but discourages smoking among women. However, the
present study does not appear to support these suggestions. The associations between
social meanings and smoking among women were significant, indicating that social
meanings are important determinants for smoking among Asian women. For instance,
female students in our sample who experienced smoking as a way to bond with friends
had a risk of becoming a lifetime smoker four times greater than those for whom smoking
99
did not represent a way to bond with friends. The inconsistence between prior studies
and our findings may be attributed to the different social norms between the U.S. and
Asian cultures. Participants in previous studies were pure Chinese or new immigrants
(Hsia & Spruijt-Metz, 2003; Weiss, et al., 2006). They probably retain many of their
traditional values in which smoking is welcome among men but discouraged among
women. However, subjects in our study include Asians from various Asian countries and
with different acculturation styles. Many of them were born in the US or have lived in the
US for a long time. Their health behaviors may be more acculturated to the American
culture, where female smokers are not viewed unfavorably, as they are in Asian countries.
Perhaps because the social norm of smoking in the U.S. does not differ for men and
women, we found no differential impact of social meanings on smoking across genders.
This inconsistence may also reflect a fact that Asian society may be a protective
agent against smoking. In contrast, the American culture may be a risk factor for smoking
for Asian women. Future smoking research may benefit from continuing the exploration
of whether and how acculturation to the U.S. cultures increases the risk for smoking
among Asian women.

100
Consistent with a prior work in adolescents (Spruijt-Metz, et al., 2004), we
found that meanings with high endorsement rates may not have strong or significant
associations with cigarette use. Taken together with these findings, it is suggested that for
both middle school and college students, the meanings of smoking perceived by them
may not fully reflect the actual determinants of their smoking behaviors. The implication
drawn from these findings is that health professionals can best consider both the
perceived and actual determinants of smoking when designing a smoking prevention
curriculum for youth. Although focusing on the actual determinants of smoking seems to
be more effective, as indicated by Spruijt-Metz and her colleagues (2004), ignoring
perceived determinants of smoking “may jeopardize buy-in by the targeted population”
(Spruijt-Metz, et al., 2004, p. 204) and threaten the effectiveness of smoking prevention
programs. Therefore, smoking prevention programs may be more successful if they
highlight both meanings endorsed by students and meanings significantly associated with
actual smoking behaviors.
Given the results of this study showing rank-orderings of meanings according to
relevance differed between genders, and that meanings which had strong influences on
smoking for one gender may have limited influence on smoking for the other gender, it is
101
suggested that the determinants of smoking may vary across genders. Health
professionals may have to consider the development of a gender-tailored smoking
prevention program in order to effectively reduce cigarette consumptions in Asian
populations.
The limitations of the present study include the restricted generalizability of the
study. The samples were drawn from a 4-year private and a 2-year community colleges in
Southern California. Findings from this study may not generalize broadly to the overall
college populations.
In addition, data analysis is correlational in nature. The predictive ability of the
meanings of smoking on smoking behaviors should be examined in future research by
using more rigorous designs, such as using a longitudinal study design to identify the
causal relationships between the meanings attached to smoking and smoking behaviors.
Another limitation is that we clustered meanings with personal, social and
functional meanings to assist our analyses on group differences on meanings of smoking.
Although it is confirmed that personal, social, and functional meanings are the three of
the underlying factors of the meanings of smoking for middle school children, whether
meanings of smoking can be clustered with the three broad categories for college students
102
has not been tested. To better understand the impact of meanings of smoking on
college smoking behaviors, further evaluation of the factorial structure of meanings of
smoking for college population is necessary.
We had very large odds ratios exceeding 20 in some cases. This may be because we
encountered small cell size in the 2 x 2 table while estimating odds ratios. The small cell
size may limit the precision of the estimates of odds ratios.
Questions also arise about how we identify repeat users. We asked participants to
provide their e-mail addresses in the surveys. When identical or similar e-mail addresses
were found, and when the responses of some items were clearly identical (i.e., age,
gender, school, ethnicity, etc.), we removed the repeated cases from the database.
However, this strategy can not identify such an individual who provided multiple email
addresses which were entirely different. More thorough strategies, such as simultaneously
using e-mail addresses and cookies (a small text file written to users’ computers each
time they visited the survey) to screen repeat users, is highly recommended in future
research using web-based surveys.
A final limitation concerns the response rates. We calculated the ratio of the total
number of students whom we invited by emails or flyers for participation to the total
103
number of the completed surveys. Although we could accurately calculate the number
of students to whom we invited, we were unable to accurately identify how many
students who we invited by emailing did receive or had a chance to check our emails.
This may lead to a bias for the response rates.
Despite these limitations, the present study guided by the Theory of Meanings of
Behavior opens new avenues for research on the determinants of smoking. This study
provides important information on the association between meanings of smoking and
smoking behaviors, and revealed that dealing with boredom, alcohol use, and some
socially relevant meanings were highly endorsed by students and most strongly related to
student’s cigarette use. It points out the needs to address both the perceived and actual
determinants of smoking. It also points out the need for the design of a gender-tailored
smoking prevention program. Moreover, the present study developed the MSI-AACS for
the underserved Asian American college populations. This will assist future research to
asses the impact of the meanings of smoking on smoking behaviors in Asian populations.
This study not only advances our current understanding on the determinants of smoking
but also contributes to future smoking research and intervention efforts for Asian
American populations.
104
CHAPTER 4
FACTOR STRUCTURE AND INV ARIANCE OF THE MEANINGS OF SMOKING
INDEX FOR ASIAN AMERICAN COLLEGE STUDENTS (MSI-AACS)

Introduction
Several school-based smoking prevention programs have been developed and
implemented during the past three decades. These educational programs were proven to
have a significant short-term effect on smoking reduction but these effects tended to fade
over time. Generally the effect lasted for about 1 to 4 years (Institute of Medicine, 1994).
The high prevalence of have ever smoked (52.8%) among 12
th
graders (Johnston, et al.,
2005a) and the increase in smoking behaviors on college campuses (Wetter, et al., 2004;
Johnston, et al., 2005b) reflect the limits of elementary and secondary school based
prevention efforts
Two possible factors may account for the limited effectiveness of current school
based smoking prevention programs. First, the majority of these programs have tended to
be based on cognitively oriented models which assume that human beings are primarily
rational and act primarily according to knowledge or rational judgment. In other words
cognition is assumed to play a major role in determining human behaviors. Spruijt-Metz
105
(1995) criticized the overemphasis on the association between cognitions and
behaviors. She argued that for several groups of people or people in several
developmental stages, cognition is not the only or the most decisive factor in human
behaviors. Her recent work in adolescent populations demonstrated that affection and
emotion often take precedence over cognition to determine health-related behaviors
(Spruijt-Metz, et la., 2004, 2005). She points out that cognitive models of behavior may
fail to reflect the most critical determinants of smoking for youth (Spruijt-Metz, 1995).
A second factor that may account for the limited effectiveness of current school
based smoking prevention programs is that large-scale smoking prevention curriculums
implemented in schools overwhelmingly targeted the mainstream White cultures
(Johnson, et al., 2005), and overlook the fact that student populations in schools are
becoming increasingly ethic diverse in many areas of the U.S. According to California
Department of Education (2005), the student populations in California are comprised of
45% Hispanic, 33.7% White, 8.7% Asian and Pacific Islander, and 8.3% African
American students. Given that cultural factors of smoking, such as social norms, cultural
values, and acceptability of smoking vary across different cultural and ethnic groups, and
the fact that cultural factors play a strong role in the prevention or development of
106
smoking behaviors (Spruijt-Metz, et al., 2004, 2005; Johnson, et al., 2005; Hsia &
Spruijt-Metz, 2003), it follows that programs developed without a cultural focus may not
suitable for a population with diverse cultures.
Because cognitive models may fail to reflect the most critical determinants of
smoking and current smoking prevention programs frequently fail to address
multicultural issues, substantial modifications of cognitively oriented models and the
incorporation of multicultural issues into prevention curriculum are necessary to reduce
cigarette consumptions for students (Spruijt-Metz, et al., 2005; Johnson, et al., 2005).
In the light of the notion that affect and cultural factors may play the major roles in
youth’s health behaviors, Spruijt-Metz (1995) developed The Theory of Meanings of
Behaviors (TMB) to provide an alternative perspective on the determinants of smoking
and other health-related behaviors. According to this theory, the meanings of behavior are
defined as what a specific behavior symbolizes for (or means to) an individual. For
instance, eating sweets or fried foods may represent “relief from worries.” The meanings
adhering to a behavior are personal and affective in nature. Unlike cognitive theories
assuming that cognitions determine behaviors, the TMB contends that people,
particularly adolescents or young adults, do not always behave rationally according to
107
their cognitions or scientific facts which they learn or have been told from outside
resources (Spruijt-Metz, et al, 2004, 2005). When there is a conflict between their
personal meanings and cognitions, meanings often take precedence over cognition to
determine a behavior. An example is that students continue smoking because it makes
them feel like they ‘belong’ or ‘fit in’ even if they understand that smoking is highly
related to lung cancer (Hsia & Spruijt-Metz, 2003).
The TMB also highlights the important influence of culture and ethnicity on the
development of personal meanings. People from different cultures and ethnicities may
develop different meanings on health-related behaviors, which in turn, motivate their
behaviors. For example, young adults from several Asian countries tend to imbue
smoking with “survival” (Hsia & Spruijt-Metz, 2003). They have to conform to the rule
or custom of the large society—smoke when your boss and your colleagues smoke.
Otherwise, they risk being viewed as uncooperative, which can threaten their successful
negotiation in work and society. Personal meanings in this context are tinged with
collectivist values. On the other hand, young adults from western countries may not
connect smoking with survival, probably because their culture is individualistic and
applauds individual preference and choice (Hsia & Spruijt-Metz, 2003). Recognizing the
108
impact of culture and the fact that America is an ethnically pluralistic society, the
TMB addresses the important associations between culture, meanings, and behaviors. It is
essential to understand different cultures and incorporate meanings into health promotion
programs in order to effectively reduce risk behaviors for adolescents and young adults.
The emerging literature on meanings of smoking and smoking behaviors shows
promise in helping us to understand ethnic differences in smoking behaviors. In order to
analyze and compare meanings across ethnic groups, Hsia and Spruijt-Metz (2003)
conceptually derived three types of meanings of smoking from college interviewing data:
personal, social, and functional meanings. Personal meanings, originally conceptualized
by Tappe (1992) and re-conceptualized by Spruijt-Metz (1995), deal with intra-personal
relations, such as comforting oneself, dealing with bad moods or being independent.
These meanings help the regulation of inner affective states. Social meanings (U.S.
Department of Health and Human Service, 1994) deal with interpersonal relation, such as
peer-group acceptance and culturally proper behavior. Functional meanings,
conceptualized by Perry (1999), are primarily conceptualized as having to do with
achieving physical or environmental goals, such as enhancing concentration.

109
Building on these studies, Spruijt-Metz and her colleagues (2005) examined the
factor structure of meanings of smoking among middle school students. Analyses of
middle school data introduced an additional factor, weight concern, into the structure of
meanings of smoking and found that personal and functional meanings were highly
predictive of lifetime smoking (have ever tried cigarette smoking even one or two puffs
in their lifetime). Hsia and Spruijt-Metz (2006, under review) further developed the
24-item Meanings of Smoking Index for Asian American college students (MSI-AACS).
They clustered meanings with personal, social, and functional meanings, and used these
three clusters as a template to analyze data. Although many interesting and important
insights have been reported, the conceptually derived three-factor structure of meanings
of smoking has not been empirically tested in college populations. Therefore, it is
important to evaluate the factorial structure of meanings of smoking because this
information will advance our knowledge on college smoking as well as guide future
intervention efforts.
The purpose of this study was to evaluate whether and how the factor structure of
meanings of smoking corresponded with the previously proposed 3-factor structure in
Asian American college populations. The 24-item Meanings of Smoking Index for Asian
110
American college students (MSI-AACS), developed by Hsia and Spruijt-Metz (under
review) was used. We divided the 24 meanings in MSI-AACS into the three categories-
personal, social, and functional meanings- according to the strategies delineated by Hsia
and Spruijt-Metz (2003). The theoretically hypothesized 3-factor model is presented in
Figure 1. It was expected that of the 24 meanings in MSI-AACS, 10 meanings clustered
with the personal factor, 11 meanings clustered with the social factor, and 3 meanings
clustered with the functional meanings (see Figure 1). Gender differences on the factor
structure of meanings of smoking were also explored.
111
Figure 1. The Proposed Theoretical 3-factor Structure of Meanings of Smoking.
It helps me deal with stress
It helps me deal with anger
It puts me in a better mood
It helps me to forget my problems
It gives me something to do or keeps me from being bored
It helps me calm down
It makes me feel more relaxed or high about things in
combination with drinking alcohol
It makes me more cheerful, joyful, and happier
I’m rebelling against authorities
It helps me to relax

It helps me bond with others
It helps me to fit in
It makes me more popular
It makes me feel I’m part of a group
It makes me feel like I’m socially accepted
It makes me feel closer to my boy- or girlfriend
It makes me feel comfortable being in social situations
It makes me feel I’m not stupid or weird
I don’t want to make another person smoke alone
It makes me look good or cool
It makes me feel I’m making a good impression

It refreshes my mind
It helps me concentrate
It helps me keep my weight down

Personal
Meanings
Social
Meanings
Functional
Meanings
112
Method
Sample
Participants in this study were 623 college students from two postsecondary
institutions in Southern California. The institutions were selected because they were
culturally diverse and contained large proportions of Asian American students. In
addition, research indicates that smoking patterns may differ according to college types
(Wechsler et al., 1998). In order to investigate the difference, two types of postsecondary
institutions were selected: a 4-year private university, and a 2-year community college.
The Format of the Survey
The survey distributed to students had two formats: one was web-based survey, and
the other one was conventional paper-and-pencil survey. The content of the surveys were
exactly the same. The web-based survey mode was used to collect data because it offers
speedy, cost-effective, asynchronous communication (ie., responding at the convenience
of the participants). Participants may be more likely to self-disclose and/or less likely to
respond in a socially desirable way due to the sense of distance associated with
responding in the internet, (Daley, et al., 2003; Erdman, Klein, & Greist, 1995). In
addition, a teacher’s instructional time and student’s learning time are not jeopardized.
113
More importantly, the similarity of the results derived from the web-based and
paper-based survey modes on drug use among undergraduate students was affirmative
(McCabe, 2004).
However, considering 1) some students may have difficulty to access the internet, 2)
some student organizations that were key to recruitment of Asian American students for
this study did not have their own webpage which made the strategy of recruiting people
via internet difficult. our secondary strategy was going to the classrooms and
administrating paper-and-pencil surveys.
Student Recruitment
In the 4-year private college, the information technology services providing
networking, computing, and telecommunication resources are fully developed. The student
organizations are active. Most of them have their own web sites and send out newsletters
to their members frequently. In addition, most of the schools have student list serves and
strong computer networks connecting students and schools. All of these make our
recruitment easier and more efficient. The recruitment strategies in the private college
included 1) posting invitations for participation on the websites of student organizations of
which a large proportion members are Asian American, such as East Asian Languages and
114
Cultures Undergraduate Association, Asian-Pacific Students Outreach Program,
Chinese American Student Association, Accounting Society, Taiwanese, Japanese, and
Korean Student Associations, and etc.; 2) distributing flyers to students walking on campus,
and posting flyers on bulletin boards in libraries and individual schools. In this way,
students who were not members of any of the student associations and who may have
difficulty accessing the internet could also be recruited. Students who were interested in
participation could either complete the survey online or request a hard copy of the survey.
The targeted number of subjects (300) was reached without going to the classrooms to
administer paper-and-pencil surveys.
Recruiting students through internet was difficult in the 2-year community college
because their student organizations frequently did not have their own on-line webpage.
Moreover, although accesses to students’ school email addresses was provided, these
school addresses were not their primary email addresses, making it inefficient to contact
students by using their school email addresses. Therefore, we approached individual
professors for the permission to distribute our survey and collect data during class in
addition to recruitment of students by sending electronic invitations to student
organizations and distributing flyers to students walking on campus.
115
Procedures
All of the participants in the 4-year private college completed the surveys through the
internet. Students who were interested in participation can click or type the link provided
on the electronic invitations or the flyers. After clicking the link, the Information Sheet
came up on the screen, which served as Consent Form. Students were asked to read the
Information Sheet thoroughly and they were encouraged to ask questions by emailing or
making phone calls to the researcher before they preceded with the survey. Another link
was provided in the bottom of the Information Sheet. Clicking the link led students to the
main page of the survey. Completion and return of the survey constituted consent to
participate in this research project. The information sheet clearly stated that participants
were required to be 18 or over, and no informed consent procedure was in place for parents
of minors. Therefore, data from respondents under the age of 18 was deleted. This study
was fully approved by the Institutional Review Board at both the private and the
community colleges.
In the 2-year community college, about half of the respondents completed the
surveys through the internet. Another half of them completed the paper-and-pencil survey
during classes. Trained data collector distributed the surveys. Students were instructed to
116
read the Information Sheet and ask questions before filling out the survey.
Participants could withdraw at any time without consequences of any kind.
Both the on-line and paper-and-pencil surveys took about 25 to 30 minutes. All
subjects were requested to provide email addresses. In order to reward the participants and
enhance the response rates, a lottery game was held. Cash prizes from $150 to $5 were
given to 86 winners whom were chosen by computer random selections. Each winner was
informed individually via email. Winners had 30-day period to claim their prizes. The
unclaimed funds were returned to the funding agency (Tobacco-Related Disease Research
Program, TRDRP).
Data was collected during a 45-day period from February to March 2005 in both
schools. In the private university, the electronic invitations and flyers were sent to 608
and 31 students respectively. A total of 322 students, excluding identified repeat
respondents, completed the web-based surveys and resulted in a 50.4% return rate. In the
community college, the response rates of the web-based and the paper-and-pencil survey
were 28.7% and 97.3% respectively. This resulted in a 49.0% response rate in the
community college and 49.7% for the entire sample. Compared with our previous work
using mailed surveys in college populations (Hsia & Spruijt-Metz, 2004), the
117
combination of web-based survey plus classroom administration produced a higher
response rate (49.7%) than a U.S. mail survey (15.3%).
Measures
Meanings of Smoking: A 24-item Meanings of Smoking Index for Asian American
College Students (MSI-AACS) was developed to assess meanings that were attached by
Asian American college students to cigarette smoking. In order to generate items
appropriate for this population, we first conducted individual interviews at the two
participating colleges, with 50 Asian students including newly immigrant, 1
st
generation,
2
nd
generation, and 3
rd
generation Asian Americans and over. Second, in order to
triangulate the findings from the current individual interviews, we also used data from
Hsia and Spruijt-Metz’s (2003) group interviews for Chinese and Taiwanese Americans,
and reviewed existing literature on meanings of physical activity, drug, alcohol, and
cigarette use (Spruijt-Metz et al., 2004). Combining results from the individual interviews,
the group interviews, and literature review, 24 meanings were generated. By adapting the
format of Meanings of Smoking Index (MSI) developed by Spruijt-Metz et al. (2004), the
24 meanings were written as 24 statements, each beginning with a stem "I smoke (or

118
might smoke) because: " followed by a meaning (e.g., "it helps me to fit in") Students
were asked to “check all that apply”.
Smoking behaviors: Smoking behaviors were assessed using self-report questions
derived from a large survey (USC Student Survey, Institute for Prevention Research,
2002). Students were asked about their lifetime use of cigarettes, smoking initiation,
past-month frequency, daily cigarette consumption, quitting behaviors, and perceptions of
current college smoking rates. Examples of the questions are: “Have you ever tried
cigarette smoking, even one or two puffs? (Lifetime use of cigarette)” “How old were
you when you smoked your first whole cigarette? (Smoking initiation)” “Think about the
last 30 days. On how many of these days did you smoke cigarettes? (Past-month
frequency of cigarette use) Lifetime smokers were defined as students who had ever
smoked even one or two puffs, 100-cigarette smokers as those who had smoked at least
100 cigarettes in their lifetime, and 30-day smokers as those who smoked on one or more
days during the 30 days preceding the survey. Biochemical research has established the
validity of self-reported smoking status among adult respondents (Caraballo, et al., 2001).
Therefore, we are confident that self-report provides an accurate assessment of smoking
status.
119
Ethnicity: Ethnicity was assessed by a self-report question, “What is your
ethnicity?” using check all that apply strategy. There are 15 response options including
subgroups of Asian Americans, such as Chinese and Vietnamese Americans; subgroups of
Hispanic Americans, such as Mexican and South Americans; Pacific Islanders,
Caucasians/European Americans, and African Americans. Due to the aims of the overall
study and the corresponding recruitment strategies, samples of Hispanic Americans and
African Americans, Pacific Islanders were small, and were omitted from the present
analyses.
Marital status: Students’ marital status was measured with the question “Your
present marital status?” Response options includes 1) Single, 2) Married, 3) Living with a
partner, 4) Widowed, 5) Separated, and 6) Divorced.
Socioeconomic Status (SES): As indicated by Hagedorn and her colleagues (2002),
SES is very difficult to measure for college students because SES is compounded by a
jumble of factors such as whether students are still living with parents or other family
members, whether their parents or relatives fully or partially support student’s tuition and
living cost, whether a student receives financial assistance from schools (e.g., loan) or
other sources, or whether a student is the head of entire household. Researchers
120
(Hagedorn, et. Al., 2002) indicated that merely asking student’s annual income may
yield inaccurate information because students, particularly those who are living with
parents and receive financial support from parents, may not know their parents’ income.
This would tend to underestimate SES in those cases where students earn a modest
income but come from affluent family. In contrast, when students coming from poor
families are earning larger salaries, or when students are the head of household for
extended families, SES tends to be overestimated. Therefore, using income to assess SES
for college students is imperfect (Hagedorn, et al., 2002). Considering 1) Asian cultures
in which parents encourage their children to pay full attention to their studies with no
distraction from employment, and 2) research findings that of the indicators of SES,
parents’ educational levels had large and significant association to health outcomes
throughout the life span (Gellis, et al., 2005; Bradley & Corwyn, 2002), we used parents’
educational levels as the indicator of SES for college students.
Data analysis
Descriptive statistics were generated using SPSS for windows version 12.0 (SPSS
Inc., 2003). Chi-square Tests and Analysis of Variance (ANOV A) were used to evaluate

121
differences across colleges in age, gender, smoking status, marital status, and parents’
educational levels.
Confirmatory factor analyses. To evaluate the hypothesized 3-factor model, a
confirmative factor analysis (CFA) using Mplus software version 3.13 (Muthen &
Muthen, 2004) was conducted. Missing data was handled using listwise deletion by
which any case with one or more missing data points is omitted entirely from analysis.
Weighted least-square parameter estimates using diagonal weight matrix with standard
errors and mean- and variance-adjusted chi-square test statistic (WLSMV) was used. This
estimator was appropriate for models with categorical or binary dependent variables
(Muthen, DuToit, & Spisic, 1997).
Even though a non-significant chi-square is a desired result for not rejecting a
model, it is suggested that chi-square fit statistic is overly strict and sensitive for models
with numerous variables and subjects (Bentler, 1990; Newcomb, 1994). Therefore, in this
study we evaluated model fit on the basis of multiple indices, including the Comparative
Fit Index (CFI; Bentler, 1990), the Tucker-Lewis Index (TLI; Tucker & Lewis, 1973), the
Root-Mean-Square Error of Approximation (RMSEA), and the ratio of Chi-square to the
degree of freedom (Carmines & McIver, 1981; Bollen, 1993). CFI and TLI values near or
122
greater than .95 are considered indicative of a good fitting model, with .90 as a lower
limit (Vandenberg & Lance, 2000; Bollen, 1989). The RMSEA is a measure of the size of
the average error involved in the estimating model. Values lower than .08, .05, and equal
to 0 are indicative of a reasonable, close, and perfect fitting model respectively (Browne
& Cudeck, 1993). Models with a Chi-square /degree of freedom ratio lower than 2 were
interpreted as having a good fit and models in which this ratio was lower then 3 were
interpreted as exhibiting a moderate fit (Bollen, 1989).
Model modification in Mplus was based on Modification Index (MI) and
substantive reasoning. Items were removed when their parameter estimates were not
significant. Factor loadings on multiple latent variables were also removed. When MI
values were larger than 10 and theoretically meaningful, correlated measurement errors
were added to the model.
Invariance analyses across genders. In order to compare the factorial structure
across genders, invariance testing was conducted using Mplus software version 3.13
(Muthen & Muthen, 2004). Following recommendations by Ballon (1989), Vandenberg et
al. (2000), and Widaman et al. (1997), we tested a series of nested models in an orderly
sequence of analytic steps. The first step was to evaluate whether an unconstrained model
123
described the data in both male and female samples. Once acceptable results were
yielded (a good match to the data with the Chi-square /degree of freedom ratio lower than
3, and GFIs greater than .95, and RMSEA lower than 0.08), the invariance of the
unconstrained model held in both genders and we moved to the second step.
In this step, factor loadings and thresholds (the measurement part of the model)
were constrained. A Chi-square difference test was run to determine whether the
constrained model resulted in a significant increase in Chi-square. When the increase in
Chi-square was nonsignificant (p <0.05), the same relation of indicators to latent variable
held in both groups, and we moved to a more restrictive step for invariance testing.
Constraints of the variances and covariances of the factors (the structural part of the
model) were added to the model in step three. A gain, a Chi-square difference test was
run. When changes in Chi-square were not significant, the similarity of the structural
parameters across genders held which led us to the final step.
The test for invariance in the final step was the most restrictive one. Full invariance
across genders was assumed by adding constraints of residual variances. When the model
with all three parameter matrices constrained matched the data, the assumption that the
same model operates in both groups was supported.
124
Results
Of the original sample of 623 participants, 163 were excluded from data analyses
because they were under 18 years old, they were not from the targeted institutions, they
were identified as repeat users, or they were not Asian of origin. This resulted in 460
Asian college students in the final sample, with 232 from the 4-year University and 228
from the 2-year community college. There were no significant differences across colleges
in the distribution of age, gender, and marital status. Students from the private university
tended to have parents with higher educational levels compared to those from the
community college (7.1, 4.9, respectively). The characteristics of study participants and
descriptive statistics for the two different types of schools were shown in Table 15.
The lifetime smoking rates in the two participating colleges were equivalent ( χ
2
=
2.8, P= 0.1), with 49.1% at the 4-year private college and 41.4% at the 2-year community
college. This finding was inconsistent with a prior study (Wechsler, et al., 1998) showing
that different types of colleges had different smoking rates. Due to lack of significant
differences across colleges in smoking patterns, we felt justified in running the
subsequent analyses by combining data from the two colleges without expecting to
encounter cross-level interactions.
125
Table 15. Demographic Characteristics of Respondents and Descriptive Statistics across Colleges
Total
(N= 460)
A 4-year
University
(N=232)
A 2-year
college
(N= 228)
ANOV A
test
χ2
test
Age [Mean (SD)] 20.9 (2.6) 21.0 (2.9) 20.7 (2.3) F= 2.1
P= 0.1
-
Gender (%)
Females/Males
55.5/44.5 51.1/48.9 59.9/41.1 - χ2= 3.6
P= 0.06
Marital status (%)
Single
Married/living with a partner
Separated/Divorced/Widowed

96.3
3.5
0.2

98.3
1.7
0

94.3
5.3
0.4

-

χ2= 5.3
P= 0.07
Parents’ educational levels
[Mean (SD)]

6.0 (2.9)

7.1 (2.8)

4.9 (2.7)

F= 70.5
P= 0.00

-
Lifetime smoking (%) 45.2 49.1 41.4 - χ2= 2.8
P= 0.10
100-cigarette smoking (%) 17.6 17.5 17.7 - χ2=0 .001
P= 0.97
30-day smoking (%) 18.7 16.8 20.6 - χ2= 1.1
P= 0.30
Table 16 shows descriptive statistics and the characteristics of the entire sample.
The final sample was 460 college students, with 255 female and 205 male students. The
Mean age for the total sample was 20.9. The majority of the student (96.3%) was single.
ANOV A and Chi square tests demonstrated gender differences on 100-cigarette and
30-day smoking rates, with females smoking less than males.

126
Table 16. Demographic Characteristics of Respondents and Descriptive Statistics across Genders.
Total
Sample
(N= 460)
Females
(N= 255)
Males
(N= 205)
ANOVA
test
χ2
test
Age [Mean (SD)] 20.9 (2.7) 20.8 (2.3) 21.0 (2.8) F= 0.05
P= 0.83
-
Marital status (%)
Single
Married/living with a partner
Separated/Divorced/Widowed

96.3
3.5
0.2

96.5
3.1
0.4

96.1
3.9
0.0

-

χ2= 1.4
P= 0.60
Parents’ educational levels
[Mean (SD)]
6.0 (2.9) 5.6 (3.0) 6.4 (2.8) F= 8.1
P= 0.00
-
Lifetime smoking (%) 45.2 43.4 47.5 - χ2= 0.8
P= 0.37
100-cigarette smoking (%) 17.6 24.5 17.6- - χ2= 11.9
P= 0.00
30-day smoking (%) 18.7 14.8 23.5- - χ2= 5.6
P= 0.02
Confirmative factor analyses
The proposed three-factor structure with 24 items of the meanings of smoking
yielded a moderate fit [ χ
2
(47) =83.68, p=.00, CFI=.94, TLI=.95, RMSEA=.04].
According to the modification index, the item “It makes me feel I’m making a good
impression” was double loaded and therefore this item was excluded from subsequent
analysis. After removing this item, the model represented a very good fit [ χ
2
(51) =86.39
p=.00, CFI=.95, TLI=.97, RMSEA=.04]. Factor loadings of the final model for males and
females students are presented in Table 17.

127
Table 17. Factor Loadings of the Retained Items for Male and Female Students
Factors and Items Males Females
A. Personal meanings
1. It helps me deal with stress .77 .89
2. It helps me deal with anger .92 .78
3. It puts me in a better mood .90 .89
4. It helps me to forget my problems .91 .80
5. It gives me something to do or keeps me from being bored .76 .69
6. It helps me calm down .82 .78
7. It makes me feel more relaxed or high about things in combination with
drinking alcohol
.74 .75
8. It makes me more cheerful, joyful, and happier .62 .71
9. I’m rebelling against authorities .42 .43
10. It helps me to relax .84 .77
B. Socially relevant meanings
11. It helps me bond with others .83 .77
12. It helps me to fit in .85 .77
13. It makes me more popular .89 .76
14. It makes me feel I’m part of a group .90 .90
15. It makes me feel like I’m socially accepted .52 .93
16. It makes me feel closer to my boy- or girlfriend .77 .81
17. It makes me feel comfortable being in social situations .76 .67
18. It makes me feel I’m not stupid or weird .76 .84
19. I don’t want to make another person smoke alone .70 .72
20. It makes me look good or cool .56 .80
C. Functional meanings
21. It refreshes my mind 1.03 .77
22. It helps me concentrate .86 .92
23. It helps me keep my weight down .54 .43
Note. All loadings are significant at P < 0.05.

128
Invariance analyses across gender
Results of invariance testing for males and females were shown in Table 18. The
unconstrained two-group model demonstrated a good fit to the data [ χ
2
(51) =81.10,
P=.005, CFI=.95, TLI=.96, RMSEA=.051]. The model fit remained satisfactory
throughout the increasingly restrictive tests for invariance. The Chi-square difference
tests were non-significant in each hypothesis testing, indicating the equivalence of the
construct in both Asian males and females.
Table 18. Results of Invariance Testing Between Genders

Competing models χ
2
(df) CFI TLI RMSEA △ χ
2
(df) p
1. No cross-group constraints 81.10(51) 0.95 0.96 0.051 - 0.005
2. Factor loadings & item thresholds
invariant
88.09(62) 0.95 0.96 0.048 6.99(11) 0.799
3. Factor loadings, item thresholds,
variances, & covariances invariant
89.45(64) 0.96 0.97 0.045 1.36(2) 0.503
4. Factor loadings, item thresholds,
variances, covariances & residual
variances invariant
115.29(80) 0.96 0.97 0.046 25.84(16) 0.056

Discussion
The goal of the present study was to examine whether and how the factor structure
of meanings of smoking corresponded with the proposed theoretical 3-factor structure in
Asian American college populations. The conceptually driven three-factor construct was
129
established by confirmatory factor analysis. The three-factor construct consisted of 23
of the original 24 items. The item, “It makes me feel like I’m making a good impression”
was dropped from the model. Ten items clustered with personal meanings, another ten
items clustered with socially relevant meanings, and the rest 3 items clustered with
functional meanings.
As expected, ten items loaded on the personal meanings. This cluster is similar to
our prior work in Chinese and Taiwanese immigrants at college level (Hsia &
Spruijt-Metz, 2003), in which the meaning “smoking gives me something to do or keeps
me from being bored” was categorized to the personal meanings. However, when
comparing with a study in middle school populations (Spruijt-Metz, et al., 2005), this
meaning did not cluster with the personal factor but with the functional factor. Personal
meanings conceptualized in the present study had to do with intrapersonal
relationship—the self’s relationship to the self. In order to keep balanced and pleasant
relationships with the self, people would act in some way, such as smoking cigarette and
binge eating, to regulate inner affective states. Moreover, from our interview data,
feelings of boredom seemed to be unpleasant feelings that threatened the relationships
with the self. As reported by interviewees, they felt bored when they were waiting for
130
someone or had breaks from classes or when studying. Just standing still and doing
nothing during the waiting or breaks made them feel uncomfortable. Exhaling and
inhaling the smoke at this time made them feel like they were doing something, and then
the feelings of boredom went away. From the above observations, dealing with boredom
for college students was related to the inner relationships with self, and therefore could be
clustered with personal factor. The mechanism of smoking for adolescents and young
adults may be different.
Due to the significant double loading, the meanings “It makes me feel like I’m
making a good impression” which was hypothesized to cluster with the social factor was
removed from the entire model. Therefore, the social factor was measure by ten items.
This cluster is similar to previous findings among both the college and middle school
students (Hsia & Spruijt-Metz, 2003, Spruijt-Metz, et al., 2004, 2005).
The results of our CFA supported the hypothesis that the three items, “smoking
refreshes my mind,” smoking helps me concentrate on school work,” and “smoking helps
to keep my weight down,” loaded on the functional factor. This cluster was similar to our
previous work among Asian college immigrants (Hsia & Spruijt-Metz. 2003), where
body image-related meanings (i.e., smoking helps to keep my weight down) do not form
131
a separate factor but clustered with the functional meaning. This findings differ from
findings in middle school students (Spruijt-Metz, et al., 2005), where body-image related
meanings, such as smoking helps controlling appetite and makes one’s clothes fit, do not
cluster with the functional meaning, but form a separate factor. One possible explanation
may be that adolescence is a time in which children search self-identity and develop
self-esteem. Self image, particularly body image is a major issue for children at this
developmental stage. The meanings related to body image or weight control formed a
separate factor may reflect the importance of the developmental issue. In order words,
smoking may rarely mean a way for adolescents to reduce weight but may also mean a
way to create personal identity or enhance self-esteem on their appearance. Self-identity
and self-esteem for young adults are more stable than for adolescents. Thus searching
self-identity or enhancing body image in college populations may be not as important as
that in adolescents. Therefore, meanings related to body image did not form an
independent factor. Future research may benefit from exploring the association between
the developmental issues and meanings of smoking between adolescents and young
adults.

132
Results of a series of invariance tests across genders demonstrated that the
3-factor model had a good fit to the data even when all three parameter matrices were
simultaneously constrained, indicating the equivalence in the factorial structure of
meanings for both Asian male and female college students.
There were some limitations that must be noted while interpreting the results and
drawing implications from the present study. First, samples were drawn from a 4-year
private and a 2-year community colleges in Southern California. Findings from this study
may not generalize broadly to the overall college populations. Second, the strategy we
used to identify repeat users was to ask participants to provide their e-mail addresses in
the surveys. When identical or similar e-mail addresses were found and these participants
had supplied responses of some items that were clearly identical (i.e., age, gender, school,
ethnicity, etc.), we removed the repeated cases from the database. However, this strategy
can not identify such an individual who took the survey multiple times and provided
entirely different email addresses. More thorough strategies, such as simultaneously using
e-mail addresses and cookies (a small text file written to users’ computers each time they
visited the survey) to screen repeat users, is highly recommended in future research using
web-based surveys. Finally, we calculated the response rates by the ratio of the total
133
number of students whom we invited by emails or flyers for participation to the total
number of the completed surveys. Although we could accurately calculate the number of
students to whom we invited, we were unable to accurately identify how many students
who we invited by emailing did receive or have a chance to check our emails. This may
lead to a bias for the response rates.
The present study supports previous suggestions that there are three underlying
factors (personal, social, and functional) for the meanings of smoking among Asian
college students. It also adds new information to the field of smoking research that the
three factor structure of the meanings of smoking is equivalent for both Asian males and
females. The confirmed three-factor structure could be used as a template for future
interpretations of smoking behaviors and could be used as a guide for the design of
smoking prevention programs for the underserved Asian American populations. Future
research should examine the predictive ability of the personal, social, and functional
meanings on smoking among Asian America college students, and explore possible
moderation effects brought by genders.
134
CHAPTER 5
DISCUSSION
The main objectives of this dissertation were to explore the meanings that Asian
America college students attached to cigarette smoking, and the associations between
these meanings and their smoking behaviors. Due to the wealth of information from the
data, this dissertation was divided into three papers. The first paper examined the
meanings of smoking across acculturation patterns and ethnicities. The second paper
examined the meanings of smoking across genders, and the last paper evaluated the
factorial structure of meanings of smoking. The current chapter summarizes the findings
of this study, and concludes with a discussion of the limitations of the study and
contributions that this study makes to the field of public health.
Discussion of the meanings of smoking
When examining meanings of smoking across acculturation patterns and ethnicities,
we found that the meanings dealing with intra-personal relationships were highly
endorsed by students and strongly associated with actual smoking behaviors. Among
these personal meanings, “the feelings of boredom” and “the feeling of enhanced high
with concurrent use of alcohol and tobacco” were the most risky factors for cigarette use
135

for Separation, Integration, and White students. Students who endorsed either of these
two meanings were at least 5.9 times more likely to become a lifetime smoker.
When examining meanings of smoking across Asian men and women, the personal
meanings related to boredom and alcohol use were once again highly endorsed by
students and strongly associated with smoking consumptions. Taken together, personal
meanings, particularly “the feelings of boredom” and “the feeling of enhanced high with
concurrent use of alcohol and tobacco,” were the most decisive determinants of smoking
for Asian America college students in this study, no matter what their acculturation
patterns or their genders. Smoking prevention programs should generate strategies that
help students to deal with the feelings of boredom and address the co-occurrence of
drinking and smoking behaviors for Asian as well as White students.
In addition to the personal meanings, the social meanings which had to do with
inter-personal relationships were another important determinants of smoking. Comparing
with Integration and White students, Separation students seemed to be more likely to
smoke for social reasons. In most Asian countries, smoking for social reasons is
encouraged. People, particularly males, smoke for social reasons are believed to have
136
pleasant characteristics by Asian society. Thus, when Asian immigrants retain most of
their home culture and avoid contacting the host American cultures, they tend to smoke.
This finding supports a prior suggestion that newly immigrated Asian Americans are
more likely to smoke than those who have been living in the U.S. for a long time (Hsia &
Spruijt-Metz, 2003).
It is worth noting that the entire Asian society encourages smoking for men but not
for women. The larger Asian society discourages smoking among women, so several
studies indicate that Asian women are less likely to smoke for social reasons than Asian
men (Hsia & Spruijt-Metz, 2003; Weiss, et al., 2006). However, the present study does
not support these findings. Our results showed that both men and women may smoke in
many social situations. This inconsistence may be due to the differences on acculturation
styles of participants. Participants in our previous study were new immigrants (Hsia &
Spruijt-Metz, 2003). They moved to the US no more than 2 years prior to the study, and
therefore may have retained many of their traditional values or customs. Subjects in our
sample included Asians from different acculturation levels. Many of them were born in
the US or have lived in the US for a long time. Their health behaviors seem to be
influenced by the American culture which is more accepting of smoking in females than
137
are Asian countries. Therefore, in this study social factors do not seem to play a
protective agent against smoking for Asian females. Future research is recommended to
explore the moderation effects of gender on the association between acculturation and
social meanings of smoking.
When examining rank orderings by salience and relevance, it was found that the
rank orderings by salience were similar across acculturation patterns and genders.
Nevertheless, the rank orderings by relevance were different across acculturation and
genders. This is similar to previous findings and indicates that the meanings of smoking
perceived by college students may not fully reflect the actual determinants of their
smoking behaviors. Although focusing on the actual determinants of smoking seems to be
more appropriate for an effective smoking prevention program, as indicated by
Spruijt-Metz and her colleagues (2004), ignoring perceived determinants of smoking
“may jeopardize buy-in by the targeted population” (Spruijt-Metz, et al., 2004, p. 204)
and negatively influence the effectiveness of smoking prevention efforts. Therefore,
health professionals should consider both the perceived and actual determinants of
smoking when designing a smoking prevention curriculum for youth.

138
Discussion of the factorial structure of the meanings of smoking
The proposed theoretical three-factor structure for the 24 items of MSI-AACS was
established by confirmatory factor analysis. As expected, ten items loaded on the personal
meaning, 3 items loaded on the functional meanings, and 10 of the original proposed 11
items loaded on the social factor. The meaning “It makes me feel like I’m making a good
impression” was removed from the entire model because of a significant double loading.
These results correspond with previous suggestions in college and middle school
students for the most part, except the following:
The item “smoking gives me something to do or keeps me from being bored”
loaded on the personal factor in our college sample. However, this item loaded on the
functional meanings in a prior research focusing on middle school students (Spruijt-Metz,
et al., 2005). Personal meanings conceptualized in the present study is associated with
intrapersonal relationships—the self’s relationships to the self. Keeping balanced and
pleasant relationships with the self is the ultimate goal for human well being. Once the
balanced inner affective state is threatened, people would act in some way to regulate it.
In this study, it is apparent that for college students the feeling of boredom is unpleasant
and uncomfortable that disturbs their inner affective state. As reported by our
139
interviewees, they tended to feel bored, for example, when they went outside the
library for a break, or when they stayed somewhere to wait for their dates or friends. The
way they dispelled unpleasant feelings of boredom was to smoke, because exhaling and
inhaling the smoke at that moment made them feel like they were doing something, and
then the feelings of boredom dissipated. Both our interview data and statistical results
support that “the feeling of boredom” should cluster with the personal meanings. The
reason why this meaning loaded on functional factor but not on personal factor among
middle school students may be due to the different physical or psychological mechanisms
of smoking between adolescents and young adults.
In addition, the item “Smoking helps me keep my weight down” loaded on the
functional meanings in this study and did not form a separate factor as what researchers
found in adolescents (Spruijt-Metz, et al., 2005). One possible explanation may be that
adolescence is a time in which children search self-identity and develop self-esteem. Self
image, particularly body image is a major issue for children at this developmental stage.
The fact that the meanings related to body image or weight control formed a separate
factor for adolescents may reflect the importance of these developmental issues. In order
words, smoking may mean a way for adolescents to reduce weight but may also mean a
140
form of identity development or enhancement of self-esteem and appearance.
Self-identity and self-esteem for young adults are more stable than for adolescents.
Searching for self-identity or enhancing body image in college populations may be not as
important as these things are for adolescents, and therefore, meanings related to body
image did not form an independent factor. Future research may benefit from exploring the
association between these developmental issues and meanings of smoking between
adolescents and young adults.
The three-factor model was evaluated by a series of invariance tests to investigate
possible gender differences. Our results demonstrated that the model fit the data very well
for both Asian men and women, indicating the three-factor model could be applied to
both Asian male and female college students.
Limitations and suggestions for future research
Some limitations should be noted when interpreting the results of this study. First,
meanings of smoking items were generated for Asian American college students. The
validity may be unstable when using these items in other ethnic populations or age
groups.

141
Second, samples were drawn from 4-year private and 2-year community
colleges in Southern California. As a consequence, the ability to generalize results from
this study to all Asian college students is severely limited.
Third, the strategy we used to identify repeat users was to ask participants to
provide their e-mail addresses in the surveys. When identical or similar e-mail addresses
were found, and when the responses of some items were clearly identical (i.e., age,
gender, school, ethnicity, etc.), we removed the repeated cases from the database.
However, this strategy can not identify such an individual who took the survey multiple
times and provided entirely different email addresses. More thorough strategies, such as
simultaneously using e-mail addresses and cookies (a small text file written to users’
computers each time they visited the survey) to screen repeat users, is highly
recommended in future research using web-based surveys.
Moreover, concerning the response rates, we calculated the ratio of the total number
of students whom we invited by emails or flyers for participation to the total number of
the completed surveys. Although we could accurately calculate the number of students to
whom we invited, we were unable to accurately identify how many students who we

142
invited by emailing did receive or have a chance to check our emails. This may make
the response rates skewed up or down.
Furthermore, in some instances, we had very large odds ratios exceeding 20. This
may be because we encountered small cell size in the 2 x 2 table while estimating odds
ratios. The small cell size may limit the precision of the estimates of odds ratios.
Finally, the design of the study was neither experimental nor longitudinal. As a
result, the cause-and-effect relationships between meanings of smoking and smoking
behaviors could not be established. Interpretations must be made carefully. In order to
further determine whether and how meanings predict smoking, a longitudinal
experimental design is necessary.
Unique contributions of the study
The meanings or the determinants of cigarette consumptions are very complicated.
It is not simply because psychological, physical, and social factors all contribute to
tobacco use, but because the U.S. is a ethnically pluralistic society in which immigrants
come from many different countries, and thus bring different cultural norms, values,
health behaviors and attitudes with them. With contacting with the mainstream American
cultures over time, immigrants’ values, customs, health behaviors and attitudes would be
143
changed gradually. Thus, when examining meanings or determinants of smoking
behaviors, researchers can not ignore cultural issues and the impact of acculturation on
smoking behaviors. Current research on meanings or determinants of smoking
overwhelmingly focused on adolescents. Very few studies conducted in college
populations. In addition, although many studies explore ethnic differences on the
meanings of smoking, the association between acculturation and meanings of smoking
has not been evaluated. The present dissertation focused on college students and took
different acculturation patterns and genders into account. Thus, the results of the current
dissertation added new information to the meanings of smoking research by exploring
similarities and differences between acculturation patterns and genders among
Asian-America college students.
This dissertation also developed a measurement tool-the MSI-AACS- for the
under-representative Asian-America college students by a carefully design qualitative and
quantitative methodologies. This could assist future study to asses the impact of the
meanings attached to smoking on smoking behaviors in Asian populations. Although a
previous study (Hsia & Spruijt-Metz, 2003) suggested three underlying factors for
meanings of college smoking and obtained striking findings using the three broad
144
categories of meanings of smoking to interpret their data, their conceptually derived
factorial structure has not been tested. This dissertation was also designed to evaluate the
proposed theoretical three-factor model for the meanings of college smoking. Our
Confirmatory Factor Analysis confirmed this structure. This result could provide a useful
template for future interpretation of the meanings of smoking, and guide the development
of culturally appropriate smoking prevention curriculum for Asian-America college
students.
Conclusions
In conclusion, the present study grounded on the Theory of Meanings of Behavior
to examine the determinants of smoking between acculturation patterns and genders. The
results of the study showed that personal and socially relevant meanings were highly
endorsed and most strongly related to cigarette smoking across all cultural and gender
groups. The results also showed that the reasons why college students think that they
smoke may not relate to their smoking behaviors, indicating the difference between the
perceived and actual determinants of smoking. In order to enhance the effectiveness of a
smoking prevention program, it is recommended to take both the perceived and actual
determinants of smoking into account. Important differences on smoking between
145
acculturation patterns and ethnicities were revealed, indicating the differences on the
major determinants of smoking across cultural groups, and a need for a cultural-tailored
smoking prevention curriculum.
To further understand the predictive power of meanings for smoking behaviors,
future research should involve longitudinal or experimental designs in order to provide
more conclusive evidence for the relationships suggested by the current dissertation. The
moderation effects of genders on the association between acculturation patterns and
smoking is worthy to be explored because this information is essential to the design of a
smoking prevention curriculum targeting different genders.
146
REFERENCES

Abernathy, T. J., Massed, L., & Romano-Dwyer, L. (1995). The relationship
between smoking and self-esteem. Adolescence, 30 (120), 899-907.

Adolphs, D., Tranel, D., & Damasio, A. R. (2003). Dissociable neural systems for
recognizing emotions. Brain and Cognition, 52, 61–69.

Alvan, S. L., Belgrave, F. Z., & Zea, M. C. (1996). Stress, social support, and college
adjustment among Latino students. Cultural Diversity & Mental Health, 2 (3), 193-203.

American Legacy Foundation (2001). New Smoking Prevalence Data [On-line].
Available: http://www.americanlegacy.org/

American Legacy Foundation (2004). Cigarette Smoking among Youth; Results from
the 2002 National Youth Tobacco Survey [On-line]. Available:
http://www.americanlegacy.org/americanlegacy/skins/alf/home.aspx

American Lung Association (2004). Smoking and Asian Americans/ Pacific Islanders
Fact Sheet [On-line]. Available:
http://www.lungusa.org/site/apps/s/content.asp?c=dvLUK9O0E&b=34706&ct=66706

Barnes, J. S. & Bennett, C. E. (2002). The Asian Population: 2000. U.S. Department
of Commerce, Economics and Statistics, U.S. Census Bureau.

Barry, J. W. (2003). Conceptual Approaches to Acculturation. In K. M. Chun, P. B.
Organista, & G. Marin (Eds.), Acculturation: Advances in Theory, Measurement, and
Applied Research (pp. 17-37). Washington, DC: American psychological Association.

Bentler, P. M. & Chou, C. P. (1987). Practical issues in structural modeling.
Sociological Methods and Research, 16 (1), 78-117.

Bentler, P. M. (1990). Comparative fix indexes in structural models. Psychological
Bulletin, 107, 238–246.

147
Bhui, K., Stansfeld, S., Head, J., Haines, M., Hillier, S., Taylor, S., Viner, R., &
Booy, R. (2005). Cultural identity, acculturation, and mental health among adolescents in
east London's multiethnic community. Journal of Epidemiology & Community Health, 59
(4), 296-302.

Bollen, K. A. (1989). Structural Equations with Latent Variables. New York: John
Wiley & Sons, Inc.

Bollen, K. A., & Long, J. S. (1993). Testing structural equation models. Newbury
Park: Sage Publications.

Bradley, R. H. & Corwyn, R. F. (2002). Socioeconomic status and child development.
Annual Review of Psychology, 53, 371-399.

Broussard, D. L. & Magnus, J. H. (2004). Risk assessment and screening for low
bone mineral density in a multi-ethnic population of women and men: does one approach
fit all? Osteoporosis International, 15 (5):349-60.

Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K.
A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136-162).
Newbury Park: Sage Publications.

Caetano, R., Ramisetty-Mikler, S., & McGrath, C. (2004). Acculturation, drinking,
and intimate partner violence among Hispanic couples in the United States: A
longitudinal study. Hispanic Journal of Behavioral Sciences, 26 (1), 60-78.

California Department of Education. (2005). Enrollment by Ethnicity 1981-82
through 2002-03 [On-line]. Available: http://www.cde.ca.gov/ds/sd/cb/enreth.asp.

Caraballo, R. S.; Giovino, G. A., Pechacek, T. F., & Mowery, P. D. (2001). Factors
associated with discrepancies between self-reports on cigarette smoking and measured
serum cotinine levels among persons aged 17 years or older - Third National Health and
Nutrition Examination Survey, 1988–1994. American Journal of Epidemiology, 153 (8),
807-814.

148
Carmines, E. G. & McIver, J. P. (1981) "Analyzing Models with Unobserved
Variables: Analysis of Covariance Structures." In G. W. Bohrnstedt & E. F. Borgatta
(Eds.), Social Measurement: Current Issues. Beverly Hills, CA: Sage Publications,
65-115.

Centers for Disease Control and Prevention. (2002). Cigarette smoking among
adults-United States, 2000. MMWR, 51 (29), 642-645.

Centers for Disease Control and Prevention. (2004). Tobacco Use and the Health of
Young People [On-line]. Available:
http://www.cdc.gov/HealthyYouth/tobacco/pdf/facts.pdf.

Centers for Disease Control and Prevention. (2005). Cigarette smoking among
adults-United States, 2003. MMWR, 54 (20), 509-513.

Chen, X., Unger, J. B., Cruz, T., & Johnson, C. A. (1999a). Smoking patterns of
Asian-American youth in California and their relationship with acculturation. Journal of
Adolescent Health, 24 (5), 321-328.

Chen, X., Unger, J. B., & Johnson, C. A. (1999b). Is acculturation a risk factor for
early smoking initiation among Chinese American minors? A comparative perspective.
Tobacco Control, 8, 402-410.

Clark, M. R. (2005). Negotiating the freshman year: Challenging and strategies
among first-year college students. Journal of College Student Development, 46 (3),
296-316.

Crump, R. L., Lillie-Blanton, M., Anthony, J. C. (1997). The influence of
self-esteem on smoking among African-American school children. Journal of Drug
Education, 27 (3), 277-291.

Daley, E. M., McDermott, R. J., McCormack Brown, K. R., & Kittleson, M. J.
(2003). Conducting web-based survey research: A lesson in internet designs. American
Journal of Health Behavior, 27 (2), 116-124.

149
Day, J. C. & Jamieson, A. (2003). School Enrollment: 2000. U.S. Census Bureau,
Washington, D.C. [On-line]. Available:
http://www.census.gov/prod/2003pubs/c2kbr-26.pdf

Duda, J. L., Smart, A. E., Tappe, M. K. (1989). Predictors of adherence in the
rehabilitation of athletic injuries: An application of personal investment theory. Journal of
Sport & Exercise Psychology, 11, 367-381.

Emmons, K. M., Wechsler, H., & Abraham, M. (1998). Predictors of smoking
among US college students, American journal of Public Health, 88(1), 104-107.

Epistein, J. A., Botvin, G. J., & Diaz, T. (2001). Linguistic acculturation associated
with higher marijuana and polydrug use among Hispanic adolescents. Substance use and
misuse, 36, 477-499.

Erdman, H., Klein, M., & Greist, J. (1995). The reliability of a computer interview for
drug use/abuse information. Behavior Research Methods and Instrumentation.
15(1):66-68.

Fenzel, L. M. (2005). Multivariate analyses of predictors of heavy episodic drinking
and drinking-related problems among college students. Journal of College Student
Development, 46 (2), 126-140.

Field, M., Mogg, K., & Bradley, B. P. (2005). Alcohol increases cognitive biases for
smoking cues in smokers. Psychopharmacology, 180, 63–72.

Freeman, D., Brucks, M., & Wallendorf, M. (2005). Young children’s understanding
of cigarette smoking. Addiction, 100, 1537-1545.

Fu, S. S., Ma, G. X., Tu, X. M., Siu, P. T., & Metlay, J. P. (2003). Cigarette smoking
among Chinese Americans and the influence of linguistic acculturation. Nicotine &
Tobacco Research, 5 (6), 803–811.

150
Gellis, L. A., Lichstein, K. L., Scarinci, I. C., Durrence, H. H., Taylor, D. J., &
Bush, A. J. (2005). Socioeconomic Status and Insomnia. Journal of Abnormal
Psychology, 114 (1), 111–118.

Goetz, J. P. & LeCompte, M. D. (1984). Ethnography and Qualitative Design in
Educational Research. Orlando, Florida: Academic Press.

Gulliver, S. B., Rohsenow, D. J., Colby, S. M., & Dey, A. N. (1995).
Interrelationship of smoking and alcohol dependence, use and urges to use. Journal of
Studies on Alcohol, 56 (2), 202-206.

Hagedorn, L., Maxwell, W., Chen, A., Cypers, S., and Moon, H. S. (2002). A
Community College Model of Student Immigration, Language, GPA, and Course
Completion [On-line]. Available:
http://www.usc.edu/dept/education/truccs/Papers/Immigration%20Paper2.pdf

Hahm, H. C., Lahiff, M., & Guterman, N. B. (2003). Acculturation and parental
attachment in Asian-American adolescents' alcohol use. Journal of Adolescent Health, 33
(2), 119-129.

Herrero, J. & Meneses, J. (2006). Short Web-based versions of the perceived stress
(PSS) and Center for Epidemiological Studies-Depression (CESD) Scales: a comparison
to pencil and paper responses among Internet users. Computers in Human Behavior, 22,
830–846.

Hsia, F. –N. & Spruijt-Metz, D. (2003). The meanings of smoking among Chinese
American and Taiwanese American college students. Nicotine and Tobacco Research, 5
(6), 837-849.

Hsia, F. –N. & Spruijt-Metz, D. (2004, November). Self-esteem and cigarette
smoking among college students. Paper presented at the American Public Health
Association 132
nd
Annual Meeting, Washington, DC.

Hsia, F. -N. & Spruijt-Metz, D. (2006). [The determinants of college smoking].
Unpublished data.
151
Hubert, H. B.; Snider, J., & Winkleby, M. A. (2005). Health status, health
behaviors, and acculturation factors associated with overweight and obesity in Latinos
from a community and agricultural labor camp survey. Preventive Medicine: An
International Journal Devoted to Practice and Theory, 40 (6), 642-651.

Hwang, A., Francesco, A. M., Kessler, E. (2003). The relationship between
individualism-collectivism, face, and feedback and learning processes in Hong Kong.
Singapore, and the United States. Journal of Cross Cultural Psychology, 34 (1), 72-91.

Ikard, F. F. & Tomkins, S. (1973). The experience of affect as a determinant of
smoking behavior: A series of validity studies. Journal of Abnormal Psychology, 81,
172–181.

Institute for Prevention Research. (2002). USC Student Survey. (Available from
University of Southern California, Institute for Prevention Research, 1000 South Fremont
Ave. Alhambra, CA, 91803.

Institute of Medicine. (1994). Growing up Tobacco Free: Preventing Nicotine
Addiction in Children and Youth. Washington DC: National Academy Press.

Jennings, P. S., Forbes, D., McDermott, B. J. S., & Hulse, G. (2005). Acculturation
and eating disorders in Asian and Caucasian Australian adolescent girls. Psychiatry and
Clinical Neurosciences, 59(1), 56-61.

Jessor, R. (1984). Adolescent development and behavioral health. In Matarazzo, J.
D., Weiss, S. M., Herd, J. A., & Miller, N. E. (Eds), Behavioral Health: A Handbook of
Health Enhancement and Disease Prevention. New York: Wiley.

Johnson, C. A., Unger, J. B., Ritt-Olson, A., Palmer, P. H., Cen, S. Y., Gallaher, P.,
& Chou, C-P. (2005). Smoking prevention for ethnically diverse adolescents: 2-year
outcomes of a multicultural, school-based smoking prevention curriculum in Southern
California. Preventive Medicine, 40, 842– 852.

152
Johnston, L. D., O'Malley, P. M., Bachman, J. G., & Schulenberg, J. E. (2005a).
Monitoring the Future national survey results on drug use, 1975-2004. Volume I:
Secondary school students (NIH Publication No. 05-5727). Bethesda, MD: National
Institute on Drug Abuse.

Johnston, L. D., O'Malley, P. M., Bachman, J. G., & Schulenberg, J. E. (2005b).
Monitoring the Future national survey results on drug use, 1975-2004. Volume II:
College students and adults ages 19-45 (NIH Publication No. 05-5728). Bethesda, MD:
National Institute on Drug Abuse.

Kadison, R. & DiGeronimo, T. F. (2004). College of the Overwhelmed: The Campus
Mental Health Crisis and What to Do About It. San Francisco: Jossey-Bass.

Kaplan, C. P., Ericson, P. I., & Juarez-Reyes, M. (2002). Acculturation, gender role
orientation, and reproductive risk-taking behavior among Latina adolescent family
planning clients, Journal of Adolescent Research, 17, 103-121.

Katz, S. K. & Lavack, A. M. (2002). Tobacco-related bar promotions: insights from
tobacco industry documents. Tobacco Control, 11 (suppl. I), i92–i101.

Kim, B. S. K., & Omizo, M. M. (2005). Asian and European American cultural
values, collective Self-Esteem, acculturative stress, cognitive flexibility, and general
self-efficacy among Asian American college students. Journal of Counseling Psychology,
52 (3), 412–419.

King, A. C. & Epstein, A. M. (2005). Alcohol Dose-Dependent Increases in
Smoking Urge in Light Smokers. Alcoholism: Clinical and Experimental Research, 29
(4), 547-552.

Koval, J. J., Pederson, L. L., Mills, C. A., McGrady, G. A., & Carvajal, S. C. (2000).
Models of the relationship of stress, depression, and other psychosocial factors to
smoking behavior: A comparison of a cohort of students in grade 6 and 8. Preventive
Medicine, 30 (6), 463-477.

153
Krohn, F. B., & Goetz, K. M. (2005). The Goetz plan: a practical smoking
cessation program for college students. College Student Journal, 39 (2), 260-269.

Lai, M. K., Ho, S. Y., & Lam, T. H. (2004). Perceived peer smoking prevalence and
its association with smoking behaviours and intentions in Hong Kong Chinese
adolescents. Addiction, 99 (9), 1195-1205.

Lantz, P. M., Jacobson, P. D., Warner, K. E., Wasserman, J., Pollack, H. A., Berson,
J., & Ahlstrom, A. (2000). Investing in youth tobacco control: A review of smoking
prevention and control strategies. Tobacco Control, 9, 47-63.

Lau, E.M.C., Lee, P., Lynn, H., Sham, A., & Woo, J. (2003). The epidemiology of
cigarette smoking in Hong Kong Chinese women. Preventive Medicine, 37, 383–388.

Lee, S-K, Sobal, J., & Frongillo, E. A. (2000). Acculturation and health on Korean
Americans. Social Science & Medicine, 51, 159-173.

Lowery, S. E., Kurpius, S. E. R., Befort, C., Blanks, E. H., Sollenberger, S., Nicpon,
M. F., & Huser, L. (2005). Body image, self-esteem, and health-related behaviors among
male and female first year college students. Journal of College Student Development, 46
(6), 612-623.

Luke, D., Esmundo, E., & Bloom, Y. (2000). Smoke signs: Patterns of tobacco
billboard advertising in a metropolitan region. Tobacco Control, 9, 15-23.

Lynn, L. L., Kang, K. J., & Ludman, E. K. (1999). Korean elderly: Diet, food beliefs,
and acculturation. Journal of Nutrition for the Elderly, 19, 1-16.

Ma, G. X., Tan, Y., Toubbeh, J. I; Su, X., Shive, S. E., & Lan, Y. (2004).
Acculturation and smoking behavior in Asian-American populations. Health Education
Research, 19 (6), 615-625.

Madison-Colmore, O. (2003). Substance use among Taiwanese female college
students. International Journal for the Advancement of Counseling, 25 (1), 43-51.

154
Marin, G. & Gamba, R. J. (2003). Acculturation and Changes in Cultural Values.
In K. M. Chun, P. B. Organista, & G. Marin (Eds.), Acculturation: Advances in Theory,
Measurement, and Applied Research (pp. 83-93). Washington, DC: American
psychological Association.

SPSS Inc. (2003). Statistical Package for the Social Sciences version 12.0
[Computer software]. Chicago, IL: SPSS Inc.

McCabe, S. E. (2004). Comparison of web and mail surveys in collecting illicit drug
use data: A randomized experiment. Journal of Drug Education, 34 (1), 61-72.

McKee, S. A., Hinson, R., Rounsaville, D., & Petrelli, P. (2004). Survey of
subjective effects of smoking while drinking among college students. Nicotine &
Tobacco Research, 6 (1), 111–117.

Morgan, D. L. (1997). Focus Group as Qualitative Research. (2
nd
ed.). Newbury
Park, London, New Delhi: Sage Publications.

Muggli, M. E., Pollay, R. W., Lew, R., & Joseph, A. M. (2002). Targeting of Asian
Americans and Pacific Islanders by the tobacco industry: results from the Minnesota
Tobacco Document Depository. Tobacco Control, 11, 201–209.

Muthen, B. O., du Toit, S. H. C., & Spisic, D. (1997). Robust inference using
weighted least squares and quadratic estimating equations in latent variable modeling
with categorical and continuous outcomes. Unpublished manuscript.

Muthen, L. K. & Muthe, B. O. (2004). Mplus version 3.13 [computer software]. Los
Angeles, CA: Muthen & Muthen.

Nader P. R., Sellers D. E., Johnson C. C.,& Perry C. L. (1996). The effect of adult
participation in a school-based family intervention to improve children's diet and physical
activity: The child and adolescent trial for cardiovascular health. Preventive Medicine: An
International Devoted to Practice & Theory, 25,455-464.

155
Newcomb, M. D. (1994). Drug use and intimate relationships among women and
men: Separating specific from general effects in prospective data using structural
equation models. Journal of Consulting and Clinical Psychology, 62, 463–476.

Ouarasse, O. A. & van de Vijver, F. J. R. (2005). The role of demographic variables
and acculturation attitudes in predicting sociocultural and psychological adaptation in
Moroccans in the Netherlands. International Journal of Intercultural Relations, 29 (3),
251-272.

Panksepp, J. (1998). Affective Neuroscience: The Foundations of Human and
Animal Emotions. New York: Oxford University Press.

Panksepp, J (2003). At the interface of the affective, behavioral, and cognitive
neurosciences: Decoding the emotional feelings of the brain. Brain & Cognition, 52,
4–14.

Patterson, F., Lerman; C., Kaufmann; V. G., Neuner; G. A., and Audrain-McGovern,
J. (2004). Cigarette smoking practices among American college students: review and
future directions. Journal of American College Health, 52 (5), 203-211.

Perry, C. L. & Kelder, S. H. (1992). Models for effective prevention. Journal of
Adolescent Health, 13, 355–363.

Perry, C. L. (1999). Creating Health Behavior Change (Vol. 43). Thousand Oaks,
London, New Delhi: Sage Publications, Inc.

Peterson, A. V. Jr., Kealey, K. A., Mann, S. L., Marek, P. M., & Sarason, I. G.
(2000). Hutchinson smoking prevention project: Long-term randomized trial in
school-based tobacco use prevention—results on smoking. Journal of the National
Cancer Institute, 92 (24), 1971-1991.

Phinney, J. S. (2003). Ethnic Identity and Acculturation. In K. M. Chun, P. B.
Organista, & G. Marin (Eds.), Acculturation: Advances in Theory, Measurement, and
Applied Research (pp. 63-81). Washington, DC: American psychological Association.

156
Price, R. K., Risk, N. K., Wong, M. M., & Klingle, R. S. (2002). Substance use
and abuse by Asian Americans and Pacific Islanders: Preliminary results from four
national epidemiologic studies. Public health Reports, 117 (Suppl. 1), S39-S50.

Redfield, R., Linton, R., & Herskovits, M. (1936). Memorandum on the study of
acculturation. American Anthropologist, 38, 149-152.

Rigotti, N. A., Lee, J. E., & Wechsler, H. (2000). US college students’ use of tobacco
products: Results of a national survey. Journal of the American Medical Association, 284
(6), 699-705.

Rigotti, N. A., Moran, S. E., & Wechsler, H. (2005). US college students’ exposure
to tobacco promotions: Prevalence and association with tobacco use. American Journal of
Public Health, 95 (1), 138-144.

Robbins, A. S., Chao, S. Y ., Fonseca, V . P., Snedecor, M. R., & Knapik, J. J. (2001).
Predictors of low physical fitness in a cohort of active-duty U.S. Air Force members.
American Journal of Preventive Medicine, 20 (2):90-96.

Satia, J. A., Patterson, R. E., Kristal, A. R., Hislop, T. G., Yasui, Y., & Taylor, V. M.
(2001). Development of scales to measure dietary acculturation among
Chinese-Americans and Chinese-Canadians. Journal of the American Dietetic
Association, 101, 548-553.

Shor, R. E. (1981). Beliefs of smokers and never smokers about the motives that
underlie tobacco smoking. Addictive Behaviors, 6 (4), 317-324.

Silverstein, M. & Chen, X. (1999). The impact of acculturation in Mexican
American families on the quality of adult grandchild-grandparent relationships. Journal
of Marriage & the Family, 61 (1), 188-198.

Slovic, P. (2001). Cigarette smokers: Rational actors or rational fools? In P. Slovic
(Ed.), Smoking: Risk, Perception and Policy (pp. 97-124). Thousand Oaks, CA: Sage.

157
Song, Y. J., Hofstetter, C. R., Hovell, M. F., Paik, H. Y., Park, H. R., Lee, J., &
Irvin, V. (2004). Acculturation and health risk behaviors among Californians of Korean
descent. Preventive Medicine: An International Journal Devoted to Practice and Theory,
39 (1), 147-156.

Spruijt-Metz, D. (1995). Personal incentives as determinants of adolescent health
behavior: The meaning of behavior. Health Education Research: Theory and practice, 10,
355-364.

Spruijt-Metz, D. (1999). Adolescence, affect, and Health. London: Psychology
Press.

Spruijt-Metz, D., Gallaher, P., Unger, J. B., and Johnson, C. A. (2004). Meanings of
smoking and adolescent smoking across ethnicities. Journal of Adolescent Health, 35,
197-205.

Spruijt-Metz, D., Gallaher, P., Unger, J. B., and Johnson, C. A. (2005). Unique
contributions of meanings of smoking and outcome expectancies to understanding
smoking initiation in middle school. Annals of Behavioral Medicine, 30 (2), 104-111.

SPSS Inc. (2003). Statistical Package for the Social Sciences version 12.0
[Computer software]. Chicago, IL: SPSS Inc.

Sussman, S. (2001). School-based tobacco use prevention and cessation: Where are
we going? America journal of Health Behavior, 25(3), 191-199.

Tappe, M. K. (1992). The model of personal investment; A theoretical approach for
explaining and predicting adolescent health behavior. Health Education Research, 7,
277-300.

Tonstad, S. & Svendsen, M. (2005). Premature coronary heart disease, cigarette
smoking, and the metabolic syndrome. American Journal of Cardiology, 96 (12),
1681-1685.

158
Triandis, H. C. (1995). Individualism and Collectivism. Boulder, CO: Westview.

Trimble, J. E. (2003). Introduction: Social change and acculturation. In M. Chun, P.
B. Organista, & G. Marin (Eds.), Acculturation: Advances in Theory, Measurement, and
Applied Research. Washington, DC: American Psychology Association.

Tucker, L. R., & Lewis, C. (1973) A reliability coefficient for maximum likelihood
factor analysis. Psychometrika, 38, 1-10.

Unger, J. B., Cruz, T. B., Ribisl, K. M., Baezconde-Garbanati, L., Chen, X., &
Johnson, C. A. (2000). English language use as a risk factor for smoking initiation
among Hispanic and Asian American Adolescents: Evidence for mediation by
tobacco-related beliefs and social norms. Health psychology, 19 (5), 403-410.

Unger, J. B., Gallaher, P., Shakib, S., Ritt-Olson, A., Palmer, P. H., & Johnson, C. A.
(2002). The AHIMSA acculturation scale: A new measure of acculturation for adolescents
in a multicultural society. Journal of Early Adolescence, 22 (3), 225-251.

Unger, J. B., Reynolds, K., Shakib, S., Spruijt-Metz, D., Sun, P., & Johnson, C. A.
(2004). Acculturation, physical activity, and fast-food consumption among
Asian-American and Hispanic adolescents. Journal of Community Health, 29 (6),
467-481.

U.S. Department of Health and Human Services. (1994). Preventing Tobacco Use
among Young People: A Report of the Surgeon General. Atlanta, Georgia: U.S.
Department of Health and Human Services, Public Health Service, Centers for Disease
Control and Prevention, National Center for Chronic Disease Prevention and Health
Promotion, Office on Smoking and Health.

U.S. Department of Health and Human Services. (1998). Tobacco Use Among U.S.
Racial/Ethnic Minority Groups- African Americans, American Indians and Alaska
Natives, Asian Americans and Pacific Islanders, and Hispanics: A Report of the Surgeon
General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease
Control and Prevention.

159
US Department of Health and Human Services (2000). Healthy People 2010:
Understanding and Improving Health. Washington, DC: US Department of Health and
Human Services [On-line]. Available: http://www.healthypeople.gov

Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the
measurement invariance literature: Suggestions, practices, and recommendations for
organizational research. Organizational Research Methods, 3, 4-69.

VERBI Software (2001). MAX Qualitative Data Analysis version 1.0 [Computer
software]. Berline, Germany: VERBI Software.

Wechsler, H., Rigotti, N. A., Gledhill-Hoyt, J., & Lee, H. (1998). Increased levels of
cigarette use among college students: A cause for national concern. Journal of the
American Medical Association, 289, 1673-1678.

Weiss, J. W., Spruijt-Metz, D., Palmer, P. H., Chou, C. P., and Johnson, C. A. (2006).
Smoking among adolescents in China: An analysis based upon the Meanings of Smoking
theory. American Journal of Health Promote, 20 (3), 171-178.

Wen, C. P., Tsai, S. P., Chen, C. J., & Cheng, T. Y . (2004). The mortality risks of
smokers in Taiwan: Part I: cause-specific mortality. Preventive Medicine, 39 (3):528-35.

Wetter, D. W., Kenford, S. L., Welsch, S. K., Smith, S. S, Fouladi, R. T., Fiore, M. C.,
and Baker, T. B. (2004). Prevalence and predictors of Transitions in smoking behavior
among college students. Health Psychology, 23 (2), 168-177.

Whalen, C. K., Jamner, L. D., Henker, B., & Delfino, R. J. (2001). Smoking and
moods in adolescents with depressive and aggressive dispositions: Evidence from surveys
and electronic diaries. Health Psychology, 20 (2), 99-111.

Widaman, K. F., & Reise, S. P. (1997). Exploring the measurement invariance of
psychological instruments: Applications in the substance use domain. In K. J. Bryant, M.
Windle, & S. G. West (Eds.), The science of prevention: Methodological advances from
alcohol and substance abuse research (pp. 281-324). Washington, DC: American
Psychological Association.
160
Wong, M. M., Klingle, R. S., & Price, R. K. (2004). Alcohol, tobacco, and other
drug use among Asian American and Pacific Islander Adolescents in California and
Hawaii. Addictive Behaviors, 29, 127-141.

Yang, Y., Zhang, Z., Mukherjee, A. B., & Linnoila, R. I. (2004). Increased
susceptibility of mice lacking clara cell 10-kDa protein to lung tumorigenesis by
4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, a potent carcinogen in cigarette smoke.
The Journal of Biological Chemistry, 279, 29336–29340.
161
APPENDICES
APPENDIX A: DEMOGRPHIC ITEMS
A1. How old are you?
______________ years old

A2. What is your gender?
1. #Female 2. #Male

A3. Year in college or university?
1. #1
st
year (Freshman) 5. #More than 4 years
2. #2
nd
year (Sophomore) 6. #Master level
3. #3
rd
year (Junior) 7. #Doctoral level
4. #4
th
year (Senior) 8. #Other

A4. If you are currently an international student (holding F1 status), how long have you lived in
the United States?
1. #I’m not an international student 5. #3 to 4 years
2. #Less than 1 year 6. #4 to 5 years
3. #1 to 2 years 7. #More than 5 years
4. #2 to 3 years

A5. What is your approximate (cumulative) grade point average during undergraduate year(s)?
1. #A+, A 6. #C+
2. #A- 7. #C
3. #B+ 8. #C -
4. #B 9. #D or lower
5. #B-

A6. In what country was your mother born?
1. #United States
2. #Other ________
3. #Not sure
162
A7. In what country was your father born?
1. #United States
2. #Other ________
3. #Not sure

A8. Were any of your grandparents born in the United States?
1. #Yes
2. #No
3. #Not sure

A9. Were you born in the United States?
1. #Yes
2. #No
3. #Not sure

A10. What is your ethnicity? (Check all that apply)
1. #Chinese or Chinese American
2. #Taiwanese or Taiwanese American
3. #Japanese or Japanese American
4. #Korean or Korean American
5. #Vietnamese or Vietnamese American
6. #Filipino or Filippine American
7. #Pacific Islander (Hawaiian, Tongan, Samoan, or Guamanian)
8. # Mexican or Mexican American
9. # Central American
10. # South American
11. #Other Hispanic
12. # White, Caucasian, Anglo, European American (not Hispanic)
13. # American Indian/Native American
14. # Black or African American
15. #Other _______________________

163
A11. Your present marital status?
1. #Single
2. #Married
3. #Living with a partner
4. #Widowed
5. #Separated
6. #Divorced

A12. What is the highest level of formal education that your parents obtained either in the U.S. or
in another country?
Father Mother
1. # # 6th grade or less
2. # # Junior high or middle school
3. # # Some high school
4. # # Finished high school or GED
5. # # Some community college
6. # # Completed community college
7. # # Completed trade school
8. # # Some 4-year college
9. # # Completed 4-year college
10. # # Advanced degree (Master’s degree, M.A., M.D., Ph.D., J.D., etc)
11. # # I don’t know
164
APPENDIX B: ACCULTURATION SCALE
FOR ASIAN-AMERICAN COLLEGE STUDENTS
Many people in the United States have ancestors who came from another country. When you
think about the country (or countries) that your family came from...

B1. My favorite music is from:
1. #The United States
2. #The country my family is from
3. #Both
4. #Neither

B2. My favorite TV shows/series are from:
1. #The United States
2. #The country my family is from
3. #Both
4. #Neither

B3. The holidays I celebrate are from:
1. #The United States
2. #The country my family is from
3. #Both
4. #Neither

B4. The food I eat at home are from:
1. #The United States
2. #The country my family is from
3. #Both
4. #Neither

B5. The way I do things and the way I think about things are from:
1. #The United States
2. #The country my family is from
3. #Both
4. #Neither
165
B6. Cultural values that I know pretty well are from:
1. #The United States (e.g., freedom, independence, individualism.)
2. #The country my family is from (e.g., respect parents and elders)
3. #Both
4. #Neither

B7. I feel MOST comfortable being with people who are: (Check all that apply)
1. #Chinese or Chinese American
2. #Taiwanese or Taiwanese American
3. #Japanese or Japanese American
4. #Korean or Korean American
5. #Vietnamese or Vietnamese American
6. #Filipino or Filippine American
7. #Pacific Islander (Hawaiian, Tongan, Samoan, or Guamanian)
8. # Mexican or Mexican American
9. # Central American
10. # South American
11. #Other Hispanic
12. # White, Caucasian, Anglo, European American (not Hispanic)
13. # American Indian/Native American
14. # Black or African American
15. #Other _______________________

B8. My BEST friend(s) is: (Check all that apply)
1. #Chinese or Chinese American
2. #Taiwanese or Taiwanese American
3. #Japanese or Japanese American
4. #Korean or Korean American
5. #Vietnamese or Vietnamese American
6. #Filipino or Filippine American
7. #Pacific Islander (Hawaiian, Tongan, Samoan, or Guamanian)
8. # Mexican or Mexican American
9. # Central American
10. # South American
166
11. #Other Hispanic
12. # White, Caucasian, Anglo, European American (not Hispanic)
13. # American Indian/Native American
14. # Black or African American
15. #Other _______________________

B9. The people I fit in with BEST are: (Check all that apply)
1. #Chinese or Chinese American
2. #Taiwanese or Taiwanese American
3. #Japanese or Japanese American
4. #Korean or Korean American
5. #Vietnamese or Vietnamese American
6. #Filipino or Filippine American
7. #Pacific Islander (Hawaiian, Tongan, Samoan, or Guamanian)
8. # Mexican or Mexican American
9. # Central American
10. # South American
11. #Other Hispanic
12. # White, Caucasian, Anglo, European American (not Hispanic)
13. # American Indian/Native American
14. # Black or African American
15. #Other _______________________
167
APPENDIX C: ITEMS FOR SMOKING BEHA VIORS
C1. Have you ever tried cigarette smoking, even one or two puffs?
1. #Yes
2. #No

C2. Have you ever smoked a whole cigarette?
1. #Yes
2. #No

C3. Have you smoked at least 100 cigarettes in your life? (about 5 packs)
1. #Yes
2. #No

C4. Think about the last 30 days. On how many of these days did you smoke cigarettes?
1. #I don’t smoked
2. #0 days
3. #1 or 2 days
4. #3 to 5 days
5. #6 to 10 days
6. #11 to 19 days
7. #20 to 28 days
8. #29 or 30 days

C5. Think about the last 30 days. On the days that you did smoke, what was the average
number of cigarettes you smoked per day?
1. #I don’t smoke
2. #1 or less
3. #2-5
4. #6-10
5. #11-20
6. #More than 20 cigarettes per day

168
C6. Have your parents/ guardians directly or indirectly told you that smoking is a bad habit?
1. #Yes
2. #No
3. #Not sure

C7. How old were you when you smoked your first whole cigarette?
1. #I have never smoked a whole cigarette
2. #6 years or younger
3. #7-11 years old
4. #12-14 years old
5. #22 or 23 years old
6. #18 or 19 years old
7. #20 or 21 years old
8. #15-17 years old
9. #24 years or older

C8. Describe your current use of cigarettes:
1. #I have never smoked a whole cigarette
2. #I stopped smoking completely
3. #I smoke only on special occasions
4. #I smoke monthly
5. #I smoke weekly
6. #I smoke daily

C9. How old were you when you first tried to quit smoking?
1. #I have never smoked a whole cigarette
2. #I have never considered to quit smoking
3. #12 years old or younger
4. #13-17 years old
5. #18 or 19 years old
6. #20 or 21 years old
7. #22 or 23 years old
8. #24 years or older

169
C10. Are you seriously considering stopping smoking within the next 12 months?
1. #I don’t smoke
2. #Not at all
3. #A little
4. #Some what
5. #Extremely serious

C11. Are you planning to stop smoking the next 12 months?
1. #I don’t smoke
2. #Yes
3. #No

C12. Do any of your parents, step-parents, or guardians currently smoke cigarettes?
1. #Yes
2. #No
3. #Not sure

C13. Out of every 100 students at your college, how many do you think smoke cigarettes once a
month or more?
Perceived rates for Perceived rates for
male students female students
1. #0% #0%
2. #10% #10%
3. #20% #20%
4. #30% #30%
5. #40% #40%
6. #50% #50%
7. #60% #60%
8. #70% #70%
9. #80% #80%
10. #90% #90%
11. #100% #100%

170
C14. Do most people at your age think it is okay to smoke cigarettes once in a while?
1. #Yes, definitely
2. #Yes, probably
3. #No, probably not
4. #No, definitely not

C15. Do your parents or guardians think it is okay for you to smoke cigarette?
1. #Yes, definitely
2. #Yes, probably
3. #No, probably not
4. #No, definitely not

C16. How upset would your parents or guardians be if they found out that you smoked?
1. #Not at all
2. #A little
3. #Some what
4. #Extremely upset

C17. How upset would your parents or guardians be if they found out that you drank alcohol?
1. #Not at all
2. #A little
3. #Some what
4. #Extremely upset

C18. How upset would your parents or guardians be if they found out that you used drugs? (e.g.,
Marijuana, cocaine, amphetamines, etc.)
1. #Not at all
2. #A little
3. #Some what
4. #Extremely upset
171
APPENDIX D: THE MEANINGS OF SMOKING INDEX
FOR ASIAN-AMERICAN COLLEGE STUDENTS (MSI-AACS)
I smoke (or might smoke) because: (Check all that apply)
1. #It helps me bond with others
2. #It helps me to fit in
3. #It makes me more popular
4. #It makes me feel like I am part of a group
5. #It makes me feel like I’m socially accepted
6. #It makes feel closer to my boyfriend or girlfriend
7. #It makes me feel more comfortable in social situations
8. #I don’t want to make another person smoke alone
9. #It makes me feel I’m not stupid or weird
10. #It helps me deal with stress
11. #It helps me deal with anger
12. #It puts me in a better mood
13. #It helps me forget my problems
14. #It makes me feel more relaxed or high about things in combination with drinking
alcohol
15. #It makes me more cheerful
16. #It gives me something to do
17. #It makes me look good
18. #It keeps me from being bored
19. #It refreshes my mind
20. #It helps me to relax
21. #It helps me concentrate
22. #It helps me keep my weight down
23. #It helps me calm down
24. #I’m rebelling against authorities
25. #Other (explain) __________________________________________

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Motivation and the meanings of health behavior as factors associated with eating behavior in Latino youth

PDF

The relationship of model minority stereotype, Asian cultural values, and acculturation to goal orientation, academic self-efficacy, and academic achievement in Asian American college students

PDF

Objectively measured physical activity and body fat distribution in overweight Hispanic and African American adolescents

PDF

Relationship of gender role conflict and acculturation to willingness to seek psychological help among Asian American and European American men

PDF

Are acculturation and parenting styles related to academic achievement among Latino students?

PDF

Sex-role orientation, gender role attitudes, and acculturation as predictors of psychological well-being

PDF

The relationship of teachers' parenting styles and Asian American students' reading motivation

PDF

Psychosocial and behavioral ractors associated with emotional eating in adolescents

PDF

Who do we tell? School violence and reporting behavior among native Hawaiian high school students

PDF

Mixed methods investigation of user engagement with a smoking cessation app

PDF

The effect of diversity courses on international students from China and Hong Kong: a focus on intergroup peer relationships

PDF

Marginalization in acculturation is related to objectively measured physical activity in Latina adolecents

PDF

The relationship among the nurturance and monitoring dimensions of parenting, academic self-concept, and acculturation, in the academic achievement of Latino college students

PDF

College choice of natural science students: the factors and sources that influence enrollment decisions

PDF

The relationship between parenting styles, career decision self-efficacy, and career maturity of Asian American college students

PDF

Relationship of teacher's parenting style to instructional strategies and student achievement

PDF

Comparing the effectiveness of online and face-to-face classes among California community college students

PDF

Physical aggression in higher education: student-athletes’ perceptions and reporting behaviors

PDF

Exploring the relationship between academic achievement and classroom behavior of Asian American elementary school students

PDF

A study of student affairs administration professional preparation in Chinese higher education

Asset Metadata

Creator Hsia, Fan-Ni (author)

Core Title Acculturation styles, gender differences, and meanings of smoking in Asian-American college students

School Rossier School of Education

Degree Doctor of Education

Degree Program Learning and Instruction

Publication Date 11/30/2006

Defense Date 08/07/2006

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag acculturation,Asian Americans,College students,OAI-PMH Harvest,Smoking

Language English

Advisor Hitchcock, Maurice (committee chair), Spruijt-Metz, Donna (committee member), Sundt, Melora A. (committee member)

Creator Email hsia.fanni@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m203

Unique identifier UC1460614

Identifier etd-Hsia-20061130 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-159847 (legacy record id),usctheses-m203 (legacy record id)

Legacy Identifier etd-Hsia-20061130.pdf

Dmrecord 159847

Document Type Dissertation

Rights Hsia, Fan-Ni

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu