{ll

l

  

W

W

 

HI“

 

fig WWW“NWWW

 

{b.5335

f‘

uO‘: O

 

LIBRARY
Michigan State
University

 

This is to certify that the
thesis entitled

INTERCULTURAL FRIENDSHIP FORMATION AND
DEVELOPMENT: A SURVEY STUDY BASED ON AN
ALL-CHINESE SAMPLE ABOUT THE ROLE OF
FACEBOOK IN THE FRIENDSHIP BETWEEN CHINESE

M.A.

AND AMERICAN STUDENTS

presented by

Jian Rui

has been accepted towards fulfillment
of the requirements for the

degree in Telecommunication,
Information Studies, and Media

 

 

Major Professor’s Signature

biz. lilac?

Date

 

 

MSU is an Affirmative Action/Equal Opportunity Employer

 

 

 

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

 

DATE DUE

DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:fProlecc&Pres/ClRC/DateDue indd

INTERCULTURAL FRIENDSHIP FORMATION AND DEVELOPMENT: A
SURVEY STUDY BASED ON AN ALL-CHINESE SAMPLE ABOUT THE ROLE
OF FACEBOOK IN THE FRIENDSHIP BETWEEN CHINESE AND AMERICAN

STUDENTS

By

J ian Rui

A THESIS

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

MASTER OF ARTS

Telecommunication, Information Studies and Media

2009

ABSTRACT

INTERCULTURAL FRIENDSHIP FORMATION AND DEVELOPMENT: A
SURVEY STUDY BASED ON AN ALL-CHINESE SAMPLE ABOUT THE ROLE
OF FACEBOOK IN THE FRIENDSHIP BETWEEN CHINESE AND AMERICAN

STUDENTS

By

J ian Rui

The current research studies how Chinese students use F acebook to develop
friendship with Americans, and how F acebook impacts their friendship with
Americans. Factors that affect friending decisions, the level of uncertainty reduction
about American friends, and the amount of bonding social capital gained from
Americans via Facebook, and the specific usage of Facebook were studied. The result
shows a neutral attitude by Chinese students towards Facebook’s ability to develop
intercultural friendship. Respondents tend to make friends with those they have
previous offline contact with and those they share similarity to. In addition,
friend-related Facebook use pattern positively predicts the extent of uncertainty
reduction while self-related F acebook use pattern positively predicts the amount of

bonding social capital obtained from Americans.

ACKNOWLEDGEMENTS

I would like to express my sincere and whole-hearted appreciation to my
committee chair, Dr. Nicole Ellison, who has persistently supported me with her great
patience, helpful guidance, and invaluable recommendations. Dr. Ellison led me to the
right path when I first entered this field, and has provided utmost support to me since
then.

I also wish to thank Dr. Clifford Lampe and Dr. Wei Peng for their precious
suggestions to my research, especially Dr. Peng, who still attended my oral defense
during her maternal break.

Special thanks should be given to my friends and parents, who have
encouraged and supported me in the whole process of this study. I will never forget
my friends’ efforts in data collection and my parents’ encouragement whenever
necessary. Finally, I would like to say to my grandfather that the only sorrow in

completing this research is that I cannot share the joy with you.

iii

TABLE OF CONTENT

LIST OF TABLES ............................................................................... V
LIST OF SYMBOLS OR ABBREVIATIONS ............................................. VI
INTRODUCTION ................................................................................ 1
LITERATURE REVIEW

The meaning of friendship ...................................................................... 2
Online friendship formation ..................................................................... 6
Online friendship development ............................................................... 16
Research questions .............................................................................. 25
METHOD

Sample and procedure .......................................................................... 29
Measures .......................................................................................... 30
RESULT

Descriptive findings ............................................................................ 33
Research questions .............................................................................. 35
DISCUSSION

Conclusions ...................................................................................... 41
Limitation and implication ..................................................................... 51
APPENDICES ................................................................................... 55
REFERENCES .................................................................................. 63

LIST OF TABLES

Table 1: Perceived effectiveness of Facebook in developing cross-cultural
friendships ....................................................................................... 62

Table 2: Social capital people can obtain from Americans and Chinese respectively by

using Facebook .................................................................................. 62
Table 3: Uncertainty reduction scale ......................................................... 63
Table 4: Factors that affect friending decisions ............................................. 64
Table 5: Facebook intensity .................................................................... 65
Table 6: Facebook use pattern ................................................................. 66

Table 7: Demographic information, and the use of the Internet and Facebook. . . . . . ...67

Table 8: Paired sample T-test of factors that affect friending decisions on
Facebook ......................................................................................... 67

Table 9: Regression predicting the extent of uncertainty reduction ...................... 68

Table 10: Regression predicting the amount of American bonding social capital. . ....68

LIST OF SYMBOLS OR ABBREVIATION
CMC: Computer-mediated communication
SIP: Social information processing theory
SNS: Social Network Sites
SPT: Social Penetration Theory
UR: Uncertainty reduction

URT: Uncertainty reduction theory

vi

Intercultural friendship formation and development: A survey study based on an
all-Chinese about the role of Facebook in the friendship between Chinese and

American students

Introduction

The popularity of F acebook has spurred an abundance of research on this most
popular social network site (SNS) in US. Many researchers studied the interpersonal
interaction that occurred through Facebook. An important field researchers focus on is
how Facebook helps form and develop new relationships, what these “Facebook new
relationships” are and what benefit they can give rise to. According to their research
findings, people tended to represent their offline new relationships online and move
those relationships forward with the relevant information offered by Facebook
(Andon, 2007; Ellison, Steinfield, & Lampe, 2007; Lampe, Ellison, & Steinfield,
2006; Walther, Van Der Heide, Kim, Westerman & Tong, 2007). In addition to
developing newly formed relationships, people also used F acebook to keep in touch
with their friends and hence maintain existing social ties (Dimicco, Millen, Geyer,
Dugan, Brownholtz, & Muller, 2006; Ellison et al., 2007; Lampe et al., 2006; Joinson,
2008). By helping people connect their old friends and new friends, Facebook thus
provided its users with social capital. Researchers generally found that Facebook was
an effective approach to both maintaining and producing social capital (Ellison et al.,
2007; Steinfield, Ellison, & Lampe, 2008; Valenzuela, Park, & Kee, 2008).

Research on these topics has produced great understanding of the interpersonal

communication conducted through Facebook, but they all used American students as
participants and focused solely on intracultural communication. However, the huge
success of F acebook does not only lie in its predominance in the United States but
also its great international influence. Since October 2005, Facebook has expanded to
UK, Mexico, Puerto Rico, US Virgin Islands, Germany, Israel and many other places
in the world. Therefore, in addition to intracultural communication, Facebook users
also make intercultural interaction on this popular SNS. Furthermore, as Facebook has
become an important tool of interpersonal communication in America, especially on
American campus, international students are thus likely to follow this new trend in
their daily interaction. As a result, it is necessary to study cross-cultural interpersonal
communication on Facebook, which is lacking research now.

This research thus intends to fill out the vacancy by studying the formation
and development of close friendship through Facebook between Chinese and
Americans. Specifically, this research will focus on how Facebook helps form and
develop close friendship with Americans, what mechanisms affect the formation of
this friendship, how this friendship develops, and what Chinese can benefit from this
friendship. It is hoped that the research will be able to produce some implications to

cross-cultural communication research and practices.

Literature Review
The meaning of friendship

Conceptualization of friendship.

In spite of a great deal of research on friendship, none of them has given a
clear definition of this concept due to two difficulties, which Boyd (2006)
summarized after reviewing the previous literature on conceptualization and
operationalization of friendship. First, research diverged on the scope of friendship.
Some research thought that friendship could also refer to more intense relationships
like the ones between family members and sexual partners. Some research, however,
ruled that type of relationship out of friendship. The other difficulty was that
hierarchies existed within fiiendship. As Boyd (2006) mentioned, phrases such as
“best friends” and “bestest friends” suggested that there were different types of
friendship in reality.

As a result of the dual difficulties, researchers have found a great variety of
components of friendship, which made difficult conceptualizing this term. For
instance, Bukowski, Hoza, and Boivin (1994) argued that friendship included five
components: companionship, conflict, help, security, and closeness. Patterson, Bettini,
and Nussbaum (1993) summarized nine components of friendship, including devotion,
commnality, reciprocity, relational stratification, fiequent contact, positive attributions,
positive impact, understanding, and familial comparison. Adams, Blieszner, and de
Vries (2000) defined fiiendship from five angles and summarized 17 elements as
follow: behavioral processes (self-disclosure, sociability, assistance, and shared
activities), cognitive process (loyalty/commitment, trust, shared interests/values,
acceptance, empathy, and appreciation/respect), affective process (compatibility, and

care), structural characteristics (solidarity, and homogeneity), and proxy measures of

process (frequency of contact, length of acquaintance, and duration of contacts).

In addition to conceptualizing friendship by examining its components,
researchers indicated another approach to conceptualizing this term. Friendship is a
type of social ties characterized by its great intensity and strength. Therefore, it was
reasonable to conceptualize this concept from the perspective of the strength of social
ties. Granovetter (1973) summarized four elements in defining the strength of a social
tie, including “the amount of time, the emotional intensity, the intimacy (mutual
confiding), and the reciprocal services” (p. 1361). Therefore, based on Granovetter’s
argument (1973), friendship should be a relationship in which people interact
frequently, have intense emotions towards each other, confide and help each other.

The present research focuses on close fiiendship. According to this
conceptualization, there are several implications about close friendship, which would
be explained in detail in the following sections. First, as friends were supposed to
confide to each other, close friends should have mutual knowledge about each other.
In other words, the level of uncertainty should below in close friendship. Second, as
fiiendship required people to help each other, people in close friendship should gain

social capital from others when necessary.

Friendship in the online context.
Many SNS allow people to publicly articulate their relationships with others,
but the only word people can use to do so is “Friend” on Facebook and some other

SNS (Boyd, & Ellison, 2007). This single word covers all different types of human

relations, among which friendship is merely one (Boyd, 2004; Fono, &
Raynes-Goldie, 2006).

Fono and Raynes-Goldie (2006)’s study on LiveJournal, a free blogging and
social networking service, found that there lacked a fixed understanding of being
“Friends” among the users of that online social networking service. Their result found
seven kinds of “Friendship” on LiveJournal. Only two of them, facilitating offline
relationships and interacting with other users, were related to interpersonal
communication. The other had nothing to do with fiiendship. For example, some
people made Friends with those “who [they] like[d] to read” (Friendship as Content
section, para. 2), and this relationship was called Friendship as content. Similar to
Friendship as content, some people admitted that their Friends were merely those they
“trust to read [their] private journal” (Friendship as Trust section, para. 1). Friendship
was also used to declare the relationships with particular users, to show courtesy, to
avoid troubles, or even to mean nothing. Fono and Raynes-Goldie (2006) thus argued
that due to the multiple meanings of “Friendship” on LiveJoumal it was difficult to
decide whether two people who called each other “Friends” meant the same thing.

With particular regard to Facebook, researchers did not make in-depth study
on the definition of “Friendship” on this most popular SNS. Like many other online
social network services, users also only rely on the single word “Friend” to describe
all their connections on F acebook. As a result, the definition of “Friendship” on
Facebook seemed ambiguous as well. However, what made “Friendship” on

F acebook less complicated was that relations on Facebook generally represented

users’ offline connections (Ellison etal., 2007). As the online space and the offline
world are not exclusive, online relationship rarely stays online. Research has found
that people used Facebook to maintain contact with their existing friends and develop
their newly formed relations to more intense ones (Ellison et al., 2007; J oinson, 2008;
Lampe et al., 2006). In other words, people extended their offline interactions to

F acebook, where they displayed those offline relations. As a result, friendship must be
among those multiple human relations on Facebook. In addition, Facebook allows
users to develop a superficial “Friendship” to a close friendship, which would be
explained in detail later. Therefore, it was safe to argue that there was friendship on
Facebook that could be conceptualized, as defined earlier, by a relationship in which
people interact frequently, have intense emotions towards each other, and confide and

help each other.

Online friendship formation

The motivation for online “Friendship " formation.

Although the present research focuses on close friendship, it was still
necessary to study the formation of “Friendship”. “Friendship” on Facebook can
include at least two categories. The first is real friendship while the second is more
superficial. For the former type, the mechanisms that affect “Friendship” formation
should be those that impact the formation of close friendship. For the latter type, the
formation of close friendship serves the first step of a long process of developing a

superficial “Friendship” to a closer relationship. As a result, the mechanisms that

affect “Friendship” formation should affect close relationship that will come out in the
future. Therefore, no matter which situation, it is necessary to examine the factors that
influence online “Friendship” formation.

In online social network services, there is a word used particularly to refer to
becoming “Friends” in those contexts: “friending”. As mentioned above, there were a
great variety of understandings of “Friend” on those online social networking services.
As the definition of “Friend” was closely related to the motivation for friending, there
should also be multiple motivations for friending on those sites.

Fono and Raynes-Goldie’s research (2006) produced seven definitions of
“Friendship” on LiveJournal, which suggested seven types of motivations for
friending. People made “Friends” to facilitate offline relationships, to interact with
other community members, to read others’ private content, to declare their
relationships with someone else, to display politeness, to show trust towards someone
else, and even for nothing. These factors that affected fiiending decisions were related
to interpersonal communication, self-presentation, politeness rules, and the specific
interaction context.

Boyd’s (2006) study on the reasons for friending on MySpace and Friendster,
two other popular SNS in the US, produced more results. She found a list of friending
motivations on the two sites.

1) Actual friends

2) Acquaintances, family members, colleagues

3) It would be socially inappropriate to say no because you know them

4) Having lots of Friends makes you look popular

5) It’s a way of indicating that you are a fan (of that person, band, product,
etc.)

6) Your list of Friends reveals who you are

7) Their Profile is cool so being Friends makes you look cool

8) Collecting Friends lets you see more people (F riendster)

9) It’s the only way to see a private Profile (MySpace)

10) Being Friends lets you see someone’s bulletins and their Friends-only blog
posts (MySpace)

11) You want them to see your bulletins, private Profile, private blog
(MySpace)

12) You can use your Friends list to find someone later

13) It’s easier to say yes than no (To Friend or not to Friend section, para. 3).

Among all the motivations above, only first two were about friendship or

interpersonal interaction while others reflected a number of other friending purposes.

For instance, friending was used as a strategy of self-presentation, to show that users

were popular (4) or cool (7), or to declare the users’ identity (5, 6). Motivations found

in Fono and Raynes-Goldie (2006) were also found here, such as for the purpose of

content (9, 10, 11), or for courtesy (3).

In spite of the complicated mechanisms that affected online friendship

formation, the case with Facebook might not be that difficult. As Facebook tended to

represent the offline human relations, some mechanisms found in Boyd (2006) and

Fono and Raynes-Goldie (2006) might not be found on Facebook. The close
connection between offline and online relationship in the context of Facebook
suggests that some mechanisms that affected offline friendship formation can remain
the same in online “Friendship” formation. Because of this close connection, the
following sections concerning online friendship formation would mainly focus on

research about offline relationship formation.

Uncertainty reduction theory (URT).

Interpersonal communication and computer-mediated communication (CMC)
research has produced a number of theories and models about online relationship
formation. Among them Berger and Calabrese’s URT (1975) discussed how peOple
communicated with strangers during the initial interaction, offering compelling
explanations of how people formed human relations.

URT began with the assumption that the goal of interpersonal communication
was to understand other people and make predictions of what they would say and do.
However, uncertainty, the lack of knowledge about others, existed at the initial
encounter and prevented understanding and predicting. Thus, it was important to
reduce uncertainty. Otherwise, uncertainty would prevent or retard relationship
development. Therefore, interpersonal communication was a process of information
collection and uncertainty reduction, and during this process human relations would
move forward.

URT stated that mutual knowledge about each other was the foremost drive of

relationship formation and development. Thus, mutual self-disclosure was an
important strategy to form any human relations and move them forward. Berger and
Calabrese (1975) believed that people chose information to disclose based on their
relationships. In the initial stage, where the uncertainty level was the highest, it was
the exchange of demographic information that dominated personal communication.
When it went to the personal phase, the information exchanged became “central
attitudinal issues, personal problems, and basic values” (p. 326); people knew each
other more as unique persons. The final stage was the exit phase, in which people
decided whether or not to continue the relationship in the future.

Additionally, Berger and Calabrese then summarized 7 axioms, discussing
both causes and consequences of uncertainty. As they pointed out, the limited amount
of verbal communication (a1), low degree of nonverbal affiliative expressiveness (a2),
and low similarity (36) caused uncertainty. Moreover, a high level of uncertainty
caused an increase in information seeking behavior (a3), a decrease in the intimacy
level of communication content (a4), a high rate of reciprocity (a5) and a decrease in
liking (a7). Based on those axioms, Berger and Calabrese further deduced 21
theorems about the relationship among those causes and consequences.

To sum up, URT assumed uncertainty prevented people making correct
predictions about each other, and so people were trying to decrease it as much as
possible. However, uncertainty reduction needed to be processed gradually, and
people would choose the information they exchanged in accordance with the

relationships they were in. The increasingly deepening information exchange pushed

10

human relations forward and made them smooth. Another implication of URT was
that people involved in a relationship were all motivated to decrease uncertainty, so
uncertainty reduction was reciprocal. Great uncertainty did not deter people but
motivated them to find out more information about each other. Information flow was

not simply from one person to another but bilateral and reciprocal.

Previous oflline contact.

The core argument of URT was that people must know something about each
other before friendship was formed. It is almost impossible to start any meaninng
human relations without mutual knowledge in face-to-face communication, but the
Intemet’s capability to connect people physically faraway enables people to first
become “Friends” before experiencing the process of uncertainty reduction, especially
in online chat rooms and virtual communities. However, the technological affordance
of multiple online services varies from one site to another, so it is dangerous and
simplistic to generalize the argument related to online chat rooms and virtual
communities to Facebook.

Lampe et a1. (2006) studied college students’ use of Facebook, and found
F acebook enabled its users to “track other members of their community” (p. 167).
Researchers then classified this function of surveillance into two types: “social
searching” and “social browsing”.iThe former referred to finding out those on
Facebook that people had already known offline while the latter referred to seeking

those on Facebook that people did not know. Researchers compared these two

11

patterns of Facebook usage and found out that social searching was much more
common than social browsing among F acebook users. This result indicated that
people tended to make “Friends” on Facebook with those they had already known
offline.

The finding by Lampe and his colleagues (2006) gained support from other
studies. Ellison et al. (2007) proposed that F acebook represented a communication
mode characterized by “offline to online”. The relations displayed on F acebook
usually reflected users’ ofiline contacts. J oinson (2008) found that the primary
motivation for using Facebook was to contact those people that users may not “get to
see very often” (p. 1030), maintain relationships with their previous contacts, or
reconnect with those they had lost contact with. Thus, according to Joinson (2008),
people should have already had contact with those “Friends” on Facebook before their
Facebook friendship started. This result supported the argument by Lampe et al. (2006)
and Ellison et al. (2007) that people tended to make “Friends” with those they had
already known offline on Facebook.

This “offline to online” communication mode implied that having previous
offline contact might be an important factor that affected “Friendship” formation on
F acebook. This argument resulted fiom not only the conclusion produced by previous
research (Ellison etal., 2007; Joinson, 2008; Lampe et al., 2006) but also the
argumentation of URT. Having offline contact indicated that people had somewhat
knowledge about each other. According to URT, people could not form any human

relations unless they had mutual knowledge about each other because uncertainty

12

would prevent them making predictions about each other and conducting further
interactions. The knowledge accumulated in the process of previous offline interaction
had actually set a foundation for further online interaction in the future. Although

F acebook provides many channels to uncertainty reduction, as previous research has
found, many users set their profiles private, so a person usually could not look at
another person’s profile to start the process of uncertainty reduction unless they first
made “Friends” with each other. Therefore, having previous offline contact should be

an important factor that affected Friendship formation on Facebook.

Similarity.

One factor proposed in URT that affected uncertainty reduction was similarity.
Similarity was one of the factors summarized by Berger and Calabrese (1975) that
would impact uncertainty. Axiom 6 stated that “similarities between persons reduce
uncertainty, while dissimilarities produced increases in uncertainty” (p. 333). Berger
and Calabrese (1975) thought uncertainty would be a barrier to interpersonal
relationship development, so it would be easier to build friendships if two people
shared more similarities. As a result, people tended to make friends with those that
were more similar to them.

Interpersonal communication research found that people preferred those
similar to them. McPherson, Smith-Lovin, and Cook (2001) named this preference
“homophily”, and defined “homophily” as “the principle that a contact between

similar people occurs at a higher rate than among dissimilar people” (p. 416).

13

McPherson and his colleagues categorized homophily into two types: baseline
homophily and inbreeding homophily. The former referred to the homophily “created
by the demography of the potential tie pool” (p. 419), such as gender, ethnicity, and
age, whereas the latter referred to the homophily “measured as explicitly over and
above the opportunity set” (ibid), such as value, education, and social status. They
then reviewed previous literature on the impact of homophily on interpersonal
interaction, and found homophily generally existed in all kinds of human relations.
For example, there was a high level of homophily in race/ethnicity in all types of
human relations; people liked to interact with those of the same race/ethnicity.
Therefore, the tendency to homophily supported the argument concerning similarity
proposed in URT.

Centola, Gonzalez-Avella, Eguiluz, and San Miguel (2007) summarized the
reason for the tendency to homophily. First, people felt justified in their opinions
when surrounded by those sharing similar beliefs to them. Second, people felt more
comfortable when interacting with those sharing a similar cultural background to
them. Finally, people would become more similar as they interacted with each other
more, which then rendered people more likely to choose those that shared similar
beliefs and a similar cultural background.

The tendency to homophily also applies to cross-cultural communication.
Rogers and Steinfatt (1999) argued that the great cultural gap between two people
from different cultural background leads to a high level of uncertainty. Since

uncertainty prevents prediction and interpersonal communication (Berger, &

l4

Calabrese, 1975), cross-cultural communication should be quite hard.

Additionally, Berger and Calabrese (1975) argued that similarity was
positively correlated with attraction. Axiom 7 stated that “increases in uncertainty
level produce decreases in liking; decreases in uncertainty level produce increases in
liking” (p.333). Theorem 21 stated that “similarity and liking are positively related”
(p.336). As uncertainty would prevent further interpersonal interaction, similarity,
characterized by a low level of uncertainty, should cause more attraction between
people.

Academic research has found that similarity was important in cross-cultural
interpersonal communication. Gudykunst (1985) found that cultural similarity
strongly predicted attraction and the degree of shared social network in intercultural
interaction. People tended to prefer those who came from the similar cultural
background and who had a higher degree of shared social network with them. Lee and
Boster (1991) studied how similarity was correlated with interpersonal cross-cultural
communication. Researchers found that “participants perceived intracultural partners
as much more similar to themselves” (p. 201) and “more attractive than intercultural
partners” (ibid). They thus concluded that “the effect of cultural similarity on the
uncertainty reduction variables, such as perceptions of similarity, attributional
confidence and interpersonal attraction, [wa]s robust” (p. 204). Bochner, McLeod, and
Lin (1977) studied the friendship network of foreign students at an American
university and their result showed foreign students preferred co-nationals as friends.

Due to its limited sample size, Fumham and Alibhai (1985) replicated the previous

15

study with an expanded sample size at a British university. Their data showed a strong
preference for co-national fiiends in a friendship network as well as among the
preferred companions.

Studies reviewed above all indicated the importance of similarity in
cross-cultural interpersonal communication. Similarity decreased the difficulty of
interpersonal communication as people could judge others based on their own
experiences. Similarity also made people comfortable in interpersonal cormnunication.
In short, people liked to interact with those similar to them. Therefore, similarity

should be another important factor that affected friendship formation on F acebook.

Friendship development on Facebook

Friendship contained four components according to Granovetter (1973). The
present research would pick out two of them to focus on to study the development of
friendship between Chinese and Americans on Facebook. The first was the level of
uncertainty reduction, which paralleled with confiding to each other. The second was

the amount of bonding social capital, which paralleled with exchanging mutual help.

Uncertainty reduction in friendship development.

URT provides a helpful approach to understanding close friendship
development. As close friendship is a special kind of human relations that contains
strong affection and deep feeling, it should demand mutual knowledge of a relatively

high level about each other. Since it is impossible to know everything about other

16

people, uncertainty reduction remains necessary as a human relation moves forward
from one stage to another. Hence, uncertainty reduction should continue to play an
indispensible role in close friendship development. Although URT was originally
restricted to the initial interaction by Berger and Calabrese (1975), more and more
research has found that URT can also be applied to subsequent communication. The
three stages of interpersonal relationship development in URT, as mentioned above,
clarified the intercorrelation between uncertainty reduction and relationship
development, thereby suggesting the legitimacy of applying URT to subsequent
interaction.

Other studies also indicated that the continuation of uncertainty reduction
played an important role in relationship development. For example, social penetration
theory (SPT) by Altman and Taylor (1973) described interpersonal communication as
a process in which the information exchanged became increasingly personal and
uncertainty between each other diminished gradually. Altman and Taylor (1973)
divided interpersonal communication into four stages, defined by the depth and
breadth of self-disclosure. At the first stage, people shared little personal information.
As the communication moved forward, self-disclosure increased in depth and breadth,
covering a greater number of topics and exchanging more personal information.
According to SPT, the degree of self-disclosure indicated the level of a relationship.
In other words, SPT considered uncertainty reduction both as the outcome of a
relationship and as an indicator of the quality of a relationship. Increased

self-disclosure gave rise to more personal information exchanged and more

17

uncertainty reduced, which then developed relationships to a higher level. A closer
relationship further rendered the content of self-disclosure to reach greater depth and
breadth.

URT and SPT both indicated that uncertainty reduction took an important part
not only in the stage of friendship formation but also in the following stages of
friendship development. Granovetter (1973) suggested that fiiends were supposed to
confide to each other. Therefore, friendship should display a high level of uncertainty

reduction while uncertainty reduction could help develop fiiendship firrther.

Uncertainty reduction in CMC.

The Internet explored a new space for interpersonal communication. The
emergence of more and more online communication tools and services has led to a
great number of studies on uncertainty reduction in CMC. Although traditional
research had a negative comment on the effectiveness of online interpersonal
communication, more and more studies now have held a positive position on this
issue.

Social information processing theory (SIP; Walther, 1992; Walther, & Tidwell,
1995) stated that people could adapt to the lack of nonverbal cues in CMC. Once
people became adapted, they would make full use of the remaining cues for
interaction. Although the adaptation process increased the time of online relationship
formation, people online would enjoy similar quality of communication and

relationships to the offline world. As a result, people could exchange information

18

online and develop personal ties with others under the condition of enough time. In
another research by Walther (1996), he put forward the “hyperpersonal
communication model”. He discussed how people could take advantage of the limited
cues available in CMC to make interpersonal communication better. For example,
people could make selective self-presentations, avoid showing their defects, which
would not be able to be avoided in face-to-face communication, and polished their
reply before sending it out.

These models and theories directed a great deal of research to find out that
uncertainty reduction in CMC could be effective as well. Tidewell and Walther (2002)
argued that the lack of communication cues in CMC could turn into “some sheltering
effects” (p. 325) which rendered mediated interactants to feel safe and comfortable to
disclose more and intimate information. Merkle and Richardson (2000) found a higher
level of self-disclosure in CMC and attributed it to relative anonymity. McKenna,
Green, and Gleason (2002) summarized four reasons for the effectiveness of CMC to
reduce uncertainty, especially to encourage self-disclosure. First, the relative
anonymity of CMC reduced the risk of self-disclosure, especially about intimate
information, “because one can share one ’s inner beliefs and emotional reactions with
much less fear of disapproval and sanction” (p. 10). Second, CMC avoided
face-to-face meeting, thereby preventing “people who are less physically attractive or
socially skilled fi'om developing relationships to the stage at which disclosure of
intimate information could begin” (p. 11). Thirdly, CMC was usually not once only;

interactants would keep in touch online or offline. The self-disclosed information

19

could become the foundation for a continuing close relationship, so the expectation of
future interaction would encourage self-disclosure. Finally, it was easier to find those
sharing common and specialized interests in CMC, which also encouraged
self-disclosure.

Westerman (2008) studied information seeking in CMC in the perspective of
different communication channels. Nine common communication channels were
classified into identifiable, unidentifiable and mixed types based on the degree to
which they allowed an individual to seek information about a target person. SNS was
classified as a type of mixed channel that combined mass and interpersonal media
qualities, and would be more likely to be used “for learning information about both
lesser and better-known targets” (ibid).

Facebook actually does provide many channels to uncertainty reduction. The
profiles provide a great deal of information about users, not only simple demographic
information but also more personal information such as their interest of music, movies,
and activities. There are also many more services on Facebook that could help reduce
uncertainty. For instance, users can update their status to let their friends know what
has recently happened to them, or upload photos to show what their recent life is like.
Users can also know what their fiiends are doing from News Feed as soon as they log
on to F acebook. In addition, people use Facebook mainly to interact with those they
have already known or at least have already met (Ellison et al., 2007). This familiarity
can make users willing to disclose credible personal information. Hence, Facebook

should be useful to reduce uncertainty from the perspective of its services.

20

Furthermore, unlike traditional CMC services, F acebook users cannot control
all the information presented on their profiles. Their “Friends” can leave comments on
their Wall, and these comments will also provide information for others who want to
know more about them. In addition, not all information on Facebook is text-based;
users can also obtain some information about the profile’s owner via photos and
videos. Compared with text, photos and videos are less malleable and “cannot be
manipulated as easily” (Ramirez & Walther, 2009; p.70). The multisource and
multimedia nature of F acebook indicates that Facebook users cannot manipulate all
the information flowed on this site about them. The information provided by the
third-party source can let information seekers know more about the target person.
Moreover, research has found that on the Internet people tended to trust information
provided by the third-party source. The high credibility of this kind of information
resulted from the difficulty to manipulate the information by the target person
(Ramirez & Walther, 2009). Therefore, the third-party information can be a
supplement to the information provided by the profile’s owners and thus can help
information seekers reduce uncertainty about targets.

However, there is one potential problem concerning the impact of the
information flowed on Facebook on uncertainty reduction. The information provided
by the third-party source may differ from that disclosed by the profile’s owners.
Moreover, people may make different interpretations about the information channeled
through text, videos, and photos. The contradicting information may cause confusion

about the knowledge of the profile owners. As a result, the uncertainty level may not

21

decrease but increase for information seekers.

In spite of this potential disadvantage, research generally acknowledged the
positive impact of F acebook use on uncertainty reduction. Lampe et al. (2006) and
Ellison et al. (2007) both found that information-seeking was a common activity for
F acebook users, and that result indicated Facebook was a frequently used tool of
information-seeking. Walther et al. (2007) found that third-party source had a
significant impact on how others evaluate target persons. For instance, male profile
users were regarded as more attractive if their “friends posted negatively valenced
messages about moral behaviors (e. g. drinking)” (p. 76). Although the third-party
source might provide contradicting information that easily confuses information
seekers, their information can be a supplement to a whole picture of a profile owner in
many cases. Andon (2007) tested the effectiveness of Facebook in forming romantic
relationships. The author found that people could obtain a greater amount of
knowledge by examining a user’s Facebook profile than by meeting face-to-face.
Thus, Facebook was very effective to reduce uncertainty for people seeking romantic
relationships. The researcher also found that heavy users of Facebook were more
likely to understand their partners, find similarities with their partners and form
romantic relationships. Considering the research mentioned above, Facebook users

should be able to benefit a decrease of uncertainty about their friends.

Social capital generated by Facebook.

Granovetter’s summary of the component of friendship (1973) included

22

mutual help in this relationship. Studies on the component of friendship also always
regarded help, such as informational, instrumental, and emotional support, as an
indispensable component of fiiendship (Bukowski, Hoza, & Boivin, 1994; Fehr, 2004;
Adams, Blieszner, & de Vries, 2000). The present research thus found it necessary to
study social capital, a source of this help.

Social capital generally referred to the resources people could obtain from
human relations. Bourdieu and Wacquant (1992) defined social capital as “the sum of
the resources, actual or virtual, that accrue to an individual or a group by virtue of
possessing a durable network of more or less institutionalized relationships of mutual
acquaintance and recognition” (as cited in Ellison et al., 2007, p. 1145). Researchers
have found that social capital could produce help for people who own it. Paxton (1999)
said social capital could facilitate “the production of some good” (p. 93), like mutual
help at the individual level or civic engagement at the community level.

Putnam (2000) distinguished two kinds of social capital: bonding social capital
and bridging social capital. The former existed between people who were already
close, such as family members and friends, while the latter existed between people
who were not closely connected, such as acquaintances. The difference between
bonding and bridging social capital did not only lie in the strength of the relationships
where they existed but also in their functions. Bridging social capital could introduce
people to what they were not familiar with and thus offered people access to the
information and social networks usually unavailable to them. In contrast, bonding

social capital usually did not have this “bridging” function but served to bring

23

resources to a closely-knit relationship. Granovetter’s research on strong ties and
weak ties (1973, 1983) also generated similar conclusions. Although not all weak ties
could work as a “bridge”, weak ties were more likely to become bridging social
capital than strong ties.

For a long time, there has been a debate about whether the Internet increased
social capital or not. Putnam (2000) expressed his concern that social capital had been
declining in the United States. Among all the causes of this decline, he mentioned the
technological development, especially the Internet, was increasingly “privatizing” or
“individualizing” people’s life and thus decreasing the opportunities to produce social
capital. In spite of Putnam’s negative opinion (2000), relevant research generally
found positive results (Bargh, & McKenna, 2004; Quan-Haase, & Wellman, 2004;
Wellman, Haase, Witte, & Hampton, 2001). Kraut, Patterson, Lundmark, Kiesler,
Mukopadhyay, and Scherlis (1998) found the Internet use was associated with an
increase in depression and loneliness, but a follow-up study in 2002 (Kraut, Kiesler,
Boneva, Cummings, Helgeson, & Crawford, 2002) reported that the Internet use was
associated with positive psychological and social outcomes. Wellman et al. (2001)
argued that the greatest advantage of the Internet was its ubiquitous connectivity,
which could make more contacts and more information available to people. This
result indicates that the Internet could help produce more social capital.

As for F acebook particularly, researchers found that it was effective to the
maintenance and growth of social capital (Lampe et al., 2006; Ellison et al., 2007;

Steinfield et al., 2008; Valenzuela et al., 2008). Ellison and her colleagues did a

24

longitudinal study about college students’ use and effect of Facebook. They found that
Facebook was effective in maintaining existing social ties and fostering both bridging
social capital and bonding social capital. As for the specific type of social capital,

F acebook impacted bridging social capital more than bonding social capital.
Researchers further explained the reason that the affordance of Facebook service
made it easier to “lower barriers to participation and... encourage the formation of
weak ties” (p. 1163, Ellison et al., 2007) than “creat[ing] the close kinds of
relationships that are associated with bonding social capital” (ibid). Valenzuela and
his colleagues (2008) found that Facebook use had an impact on social trust, life
satisfaction, civic engagement, and political participation. One important point that
has drawn researchers’ attention is that the impact of SNS on generating social capital
should depend on the way it is used. For instance, Nyland, Marvez, and Beck (2007)
found that very heavy users of MySpace felt less involved in the community than
lighter users. This result indicated an excessive use of SNS could prevent obtaining
social capital. Based on previous research, it was still reasonable to argue that people

can obtain social capital from their real friends through using Facebook.

Research questions

The current research intends to study the formation and development of
fiiendship between Chinese and American with the help of Facebook. The researcher
chose Facebook particularly because of its popular among Americans and

international students in the US, and its effectiveness in helping social interactions.

25

Second, international students have become a rapidly growing group in the States
nowadays, especially in many public universities. Thus, it is important to make a
diversified campus more cohesive from the perspective of a university. Given
F acebook’s potential in helping social interactions, it is necessary to study this issue.

In addition, F acebook is also a good place to study relationship formation and
development. The research would study relationship formation by examining the
factors that affect fiiending decisions, and relationship development by studying the
degree of uncertainty reduction and the amount of social capital its users can enjoy.

Further, the reason why Chinese were chosen to study is that there have been
more and more Chinese people nowadays in the United States. Chinese students have
also become a huge ethnic group at many American colleges. Therefore, studying the
communication between Americans and Chinese can produce some practical
implications for improving interpersonal interactions on American campuses
especially between different ethnic groups and thereby making college campuses
more cohesive.

The first question asked how the fiiendship between Americans and Chinese
forms on Facebook.

RQl: What factors affect Chinese’ friending decisions with Americans on
Facebook?

Two specific issues were discussed in the. present research regarding the effect
of Facebook on relationship development, uncertainty reduction and social capital.

According to Granovetter (1973), uncertainty reduction and social capital were two

26

components of fiiendship. The present research chose the two components to examine
the quality of intercultural friendship between Chinese and Americans.

Previous research has already studied the impact of F acebook on these issues
(Andon, 2007; Ellison et al., 2007; Steinfield et al., 2008; Valenzuela et al., 2008).
Researchers predicted the impact of Facebook usage, measured by Facebook intensity,
on uncertainty reduction and social capital. Facebook intensity includes three
variables, the total number of friends on Facebook, the time spent on Facebook, and a
six-item attitudinal scale that measures users’ attitude towards Facebook. However,
this scale was not enough to reflect the usage of Facebook completely and predict the
social capital generated from F acebook in that it only measured the attitude and the
quantity aspect of Facebook usage without touching on how people use Facebook.
Uses and gratification theory proposed by Katz, Blumler, and Gurevitch (1973)
provided theoretical support to examining the impact of specific usage of F acebook
on the result people could gain. Uses and gratification theory argued that based on
expectations of different kinds of media people chose to use a particular kind of media
to meet their specific demands; the use of media would result in gratifications of those
needs and other unintended consequences. Therefore, there is a direct association
between media usage and the outcome of using media. In fact, what specific activities
people make on SNS has drawn researchers’ attention and become a trend in current
SNS research. For example, Papacharissi and Mendelson (in press) made a scale that
measured what people did on Facebook. Researchers have been trying to find out the

impact of specific Facebook usage on uncertainty reduction and social capital.

27

This thinking led to introducing a new concept into the present research:
Facebook use pattern, which refers to the specific activity people make on Facebook.
The researcher further argued that F acebook use pattern along with F acebook
intensity predicted the outcome of using Facebook.

RQ2: How do F acebook intensity and Facebook use pattern impact uncertainty
reduction between Chinese and Americans?

In terms of social capital, the researcher argued that uncertainty reduction
should also influence the amount of social capital in addition to Facebook usage.
Albrecht and Adelrnan (1984) argued that the key point of whether to be able to
obtain social support was “whether uncertainty toward the other [wa]s reduced to an
acceptable level” (p.19). Researchers argued that the degree of uncertainty reduction
included the knowledge of a person in need of social support, such as what kind of
help this person needed, what was an appropriate way to help him or her. Therefore,
only when people had enough information a target person could he or she receive
social capital from them. In other words, the more people knew about each other, the
more probably that they could provide social capital to each other. Therefore, a
positive relationship should exist between the degree of uncertainty reduction and the
amount of social capital.

One point that needed to be mentioned was the type of social capital studied in
the present research. Although fiiends could also provide bridging social capital, it
was weak ties that were more likely to produce bridging social capital (Granovetter,

1973). In other words, compared with bridging social capital, friendship was more

28

likely to give rise to bonding social capital. Since the present research focuses on
close friendship, only bonding social capital generated from Facebook use would be
studied.

RQ3: How do F acebook intensity, Facebook use pattern and uncertainty
reduction impact bonding social capital that Chinese can obtain from their American
friends?

The final research question would intend to find out whether Chinese thought
Facebook helped or hindered their ability to interact with Americans.

RQ4: Do Chinese F acebook users think F acebook helps or hinders their ability

to interact with Americans?

Method

Sample and Procedure

A snowball sampling was used in the present research. The snowball sampling
is a non-probability sampling, and that becomes a weakness in the present study.
However, given the situation that the researcher did not have access to the complete
sampling frame and the exploratory nature of the research on Chinese students
specifically, the snowball sampling was the best choice. In terms of Chinese, anyone
that claimed him or her to be Chinese, no matter whether he or she had already
obtained the American permanent residency or American citizenship, was counted as
Chinese.

The survey was hosted on Zoomerang (http://www.zoomerangcom), an online

29

survey hosting site. An invitation to participate the survey, which contained a link to
the survey, was posted on a listserve that belongs to an association of Chinese
students at Michigan State University (http://list.msu.edu/archives/cssamsu.html). In
the meantime, the researcher also used F acebook and renren, an SNS many Chinese
students are using, to make the survey known to Chinese as many as possible.
Specifically, the researcher first wrote a note on Facebook and renren, which
announced the beginning of the survey, explained it, and urged participants to invite
friends of their own to take part in the survey. The link to the survey was posted in the
status field of his profile on these two SNS. The survey started in October 2009 and
lasted two weeks. During the two weeks when the survey was open, four reminder

posts were announced on the listserve. The survey ended with 69 respondents.

Measures

The instrument used in this research included following parts: (1) the
perceived effectiveness of F acebook in interaction with Americans, (2) the bonding
social capital Chinese obtained from both Americans and Chinese respectively by
using F acebook, (3) the degree of uncertainty reduction Chinese perceived about their
American friends through using F acebook, (4) the factors that affected Chinese’s
fiiending decisions with Americans, and (5) respondents’ demographic information
and their usage of F acebook. The researcher also added an open-ended question in the
end to collect qualitative data about respondents’ opinion about the role of Facebook

played in interaction with Americans. The question stated as follows, “Do you think

30

F acebook helps or hinders your ability to develop friendships with Americans? Why?

How?”

The perceived eflectiveness of F acebook in interaction with Americans.

This scale intended to find out the general attitude of Chinese towards
F acebook’s ability of helping interactions with Americans. Made by the researcher,
the scale contained three items (Cronbach’s alpha=0.85), including the perceived
effectiveness of Facebook in developing relationships with new American fiiends, in
helping learn information about Americans, and in helping get closer to American

acquaintances (See Table 1).

Measures of Bonding Social Capital.

Bonding social capital was assessed by using nine items from Williams’
instrument of online social capital (2006). Responses were reported on a five-point
Likert scale. Respondents were requested to think of F acebook fiiends who were
Americans and regarded as real friends before answering the questions. In order to see
if there was any difierence between social capital obtained from Chinese and
Americans, social capital was measured twice with the same scale related to Chinese
(Cronbach’s alpha=0.8) and Americans (Cronbach’s alpha=0.88) respectively (See

Table 2).

Measure of uncertainty reduction.

31

Andon’s (2007) scale of uncertainty reduction was used to measure
uncertainty reduction (Cronbach’s alpha=0.87) with some minor changes to adapt to
the present research. There were six items in total and all were on a five-point Likert

scale (See Table 3).

Measure of factors that aflectfriending decisions.

Based on Boyd’s list (2006) of different motivations for seeking fiiendships on
SNS, a l6-item scale (Cronach’s alpha=0.85) was created to measure factors affecting
Chinese friending decisions with Americans. Those items were all five-point Likert

questions (See Table 4).

Measures of F acebook intensity.

Facebook intensity scale was adapted from Ellison et al (2007). The first
question asked how many fiiends Chinese have on Facebook, and was a close-ended
question. The second asked how much time Chinese spend on Facebook, and was an
open-ended question. Finally, a six-item attitudinal scale was developed, using a
five-point Likert scale (Cronbach’s alpha=0.91).

Following Steinfield et al. (2008), the researcher made changes to the data.
First, the researcher replaced the ordinal value of the number of Facebook friends
with the mid point of the range. For example, if a respondent’s number of fiiends was
between 51 and 100, this was converted to 75.5; if a respondent’s number of fiiends

was between 101 and 150, this was converted to 125.5. Second, the researcher took

32

the log of the time spent on Facebook and the converted number of Facebook friends.
After making these changes, the three parts of data were combined into F acebook

intensity scale (See Table 5).

Measures of F acebook use pattern.

A l4-item scale was developed from Papacharissi and Mendelson (in press) to
measure F acebook use pattern, which was what specific activities people made on
Facebook. Respondents were asked about the frequency of doing the listed activities
on a five-point Likert scale while they were using Facebook (Cronbach’s alpha=0.93;

see Table 6).

Results

Descriptive findings

The researcher would first present the descriptive findings about Facebook
users (See Table 7), their use of the Internet and F acebook, and their perceptions of
the effectiveness of Facebook in developing friendship with Americans. All the
respondents of this survey were ethnically Chinese. Except one person who did not
want to disclose, 65 were born in China (94.2%), 1 was Chinese with US citizenship
or permanent residency (1.54%), and 1 was Chinese with other country’s citizenship
or permanent residency (1.54%). 27 were male (39.13%) while 42 were female
(60.87%). They had been in US for 1.66 years on average. There were 10 freshmen

(14.49%), 18 sophomore (26.09%), 1 junior (1.45%), 1 senior (1.45%), and 39

33

graduate students (56.52%). Thus, there were 30 undergraduate students (43.48%) and
39 graduate students (56.52%).

In addition, respondents spent 267.22 minutes, approximately 4.45 hours, on
average using the Internet every day. As for Facebook use particularly, they spent
27.78 minutes on average using Facebook every day.

Among all the 69 respondents, 33 had fiiends on Facebook with the number
less than 50 (47.83%), 19 between 51 and 100 (27.54%), 7 between 101 and 150
(10.14%), 4 between 151 and 200 (5.8%), 4 between 201 and 250 (5.8%), 1 between
351 and 400 (1.45%), and 1 more than 500 (1.45%). The researcher then assigned a
number to each range, with 1 to the least (less than 50), 11 to the most (more than
500), and the rest in between according to the numerical order. The average number of
F acebook friends fell in the range between 51 and 100 (M =2. I 3). The same statistic
was replicated to the question that asked the percent of Americans on Facebook. The
statistical analysis turned out 68 complete responses for this question. 40 respondents
reported the percent of Americans on their Facebook profiles was less than 20%
(58.82%), 8 between 20% and 39% (11.76%), 11 between 40% and 59% (16.18%), 6
between 60% and 79% (8.7%), and 3 between 80% and 100% (4.35%). The average
percent of Americans on Facebook fell in the range of less than 20% (M =1 .86).

The Facebook intensity scale showed that Chinese students used F acebook
less intensely compared with other populations (M =2. 63, SD=0. 95). In terms of the
frequency of making some specific activities on Facebook, the frequency was not

high, somewhat between seldom and sometimes (M =2. 52, SD=0. 72).

34

Besides, Chinese Facebook users had a neutral cormnent on F acebook’s
effectiveness to develop cross-cultural friendships with Americans (M =2. 99,

SD=I. 03). Specifically, they evaluated Facebook most effective in helping obtain
personal information about Americans (M=3. I9, SD=I.26), a bit less effective in
developing relationships with new American friends (M =2. 91 , SD=I . I ), and the least
effective in helping get closer to Americans (M =2.8 7, SD=1. I 6).

Moreover, the researcher obtained a series of findings regarding the result of
using Facebook. Chinese students had a neutral comment on the result of their using
Facebook to reduce uncertainty (M =3. 09, SD=0. 87). In terms of obtaining social
capital from Americans and Chinese with the help of Facebook respectively,
respondents reported more bonding social capital obtained from Chinese (M =3. 29,
SD=0. 83) than Americans (M =2. 53, SD=0. 7). The difference in the amount of the
bonding social capital obtained from Chinese and Americans was statistically
significant (p<0.001).

Finally, graduate students and undergraduate students did not difier from each
other in terms of F acebook use and their opinions regarding F acebook’s ability of
relationship formation and development. The only difference lies in the time spent on
Facebook. Undergraduate students (M =43. 28, SD=69.47) spent more time on

Facebook every day than graduate students (M =1 5. 95, SD=18. 9, p<0.05).

Research Questions

RQl intended to find out the factors that affected friending decisions with

35

Americans. A factor analysis was conducted and yielded three factors, with one item
left out (See Table 4). These factors included having previous offline contact
(Cronbach’s alpha=0.84, M =3. 1 4, SD=1. 05), fiiending anyone (Cronbach’s
alpha=0.82, M =2.5 6, SD=0. 99), and sharing similarity (Cronbach’s alpha=0.83,
M=2.83, SD=0.81).

To further understand the importance of these factors, the researcher
conducted paired-sample T—test, which turned out significant differences between all
three factors (p<0.05; See Table 8). Having previous offline contact was a more
significant factor than sharing similarity (mean difference=0.31, p<0.01), and
fiiending anyone (mean difference=0.59, p<0.01). Sharing similarity was also more
significant than friending anyone (mean difference=0.28, p<0.05). Thus, when
making friending decisions, Chinese students considered whether they had previous
offline contact or shared any similarity with a target person while making fiiending
decisions on Facebook.

RQ2 discussed the relationship between Facebook usage and uncertainty
reduction (See Table 9). Facebook usage was measured by Facebook intensity and
Facebook use pattern. A factor analysis was first conducted for the 14-item scale of
Facebook use pattern (See Table 6). With 2 items left out, the analysis produced two
factors, friend-related use pattern (Cronbach’s alpha=0.89, M =2.38, SD=0. 76) and
self-related use pattern (Cronbach’s alpha=0.79, M =2.5 7, SD=0. 75). F fiend-related
use pattern was defined as activities aimed at communicating with friends while

self-related use pattern was defined as activities aimed at making users’ own

36

situations known to their friends. A multiple linear regression was then run with
uncertainty reduction as the dependent variable, and F acebook intensity, friend-related
use pattern, and self-related use pattern as independent variables.

The researcher first controlled gender, undergraduate versus graduate, time in
US, and percent of Americans among F acebook fiiends. The length of time for which
a sojoumer has stayed in the US can affect his or her ability to communicate with
Americans and thus the result of reducing uncertainty about Americans and the
amount of bonding social capital they can obtain from Americans. Percent of
Americans among Facebook fiiends could affect uncertainty reduction and bonding
social capital. First, American friends can be a channel to understanding American
culture, so the more American fiiends a sojoumer has, the more likely that he or she
can understand American culture better. This enhanced cultural ability can help him or
her understand his or her American fiiends’ ideas and behaviors better. Second, the
size of a social network can affect the amount of social capital a person can obtain
(Gudykunst, 2005). Thus, the more American fiiends a sojoumer has, the more likely
that he or she can obtain bonding social capital.

Percent of Americans among Facebook fiiends was converted to continuous
variable by assigning 1 to the smallest range (less than 20%), 5 to the biggest range
(between 80% and 100%), and 2 to 4 in between according to their numerical orders.
Collinearity diagnostics showed little collinearity among predictor variables. The
adjusted R 2 for this model was 0.23. Percent of Americans among Facebook friends

(standardized coefficients=0.26, p<0.01) and friend-related Facebook use pattern

37

(standardized coefficients=0.74, p<0.01) were significant. Thus, friend-related
Facebook use pattern and percent of Americans among Facebook friends positively
predicted the degree of uncertainty reduction Chinese students benefited from using
Facebook.

RQ3 intended to explain the impact of uncertainty reduction and Facebook
usage on the amount of bonding social capital (See Table 10). The independent
variables included Facebook intensity, fiiend-related F acebook use pattern,
self-related F acebook use pattern, and uncertainty reduction. Gender, undergraduate
versus graduate, time in US, and percent of Americans among Facebook fiiends were
also controlled. Collinearity diagnostics again showed little collinearity among
predictor variables.

The similar procedure was replicated while analyzing this model. The adjusted
R2 for this model was 0.34. Uncertainty reduction (standardized coefficients=0.33,
p<0.001), percent of Americans among F acebook fiiends (standardized
coefficients=0. l9, p<0.05), self-related Facebook use pattern (standardized
coefficients=0.56, p<0.001), and friend-related Facebook use pattern (standardized
coefficients=-0.44, p<0.05) were significant. Therefore, on the one hand, uncertainty
reduction, self-related F acebook use pattern, and percent of Americans among
F acebook fiiends positively predicted the amount of bonding social capital Chinese
obtained from their American fiiends by using Facebook. On the other hand,
friend-related use pattern negatively predicted the bonding social capital Chinese

benefited from their American friends by using Facebook.

38

RQ4 asked if Chinese thought Facebook enhanced or hindered their ability to
develop relationships with Americans. 50 respondents answered this question. 37
(74%) acknowledged Facebook’s help in developing friendships with Americans, 9
(18%) denied, and the remaining 4 respondents (8%) expressed a neutral opinion on
the issue.

Among those who reported a positive comment on Facebook’s help in
developing friendships with Americans, they mainly thought that Facebook had the
following four advantages. First, they could know more people by using F acebook.
One respondent reported the reason was that Facebook was “so popular among
American students and they always build up network through it”. Due to the
popularity of Facebook, people who joined it later, to which Chinese belonged
obviously, could enjoy its great benefit of a largely expanded network by many
previous users. Facebook’s ability to develop and maintain weak ties was also
mentioned in some responses. One respondent regarded Facebook as “a
communication media between me and Americans who might contact me very
unusually”. In other word, the respondent believed Facebook could help him or her
reach out people he or she usually did not have access to. Another respondent said
“it's easier to maintain weak ties which I (as a Chinese student) share with most
American friends”. These responses lent support to Ellison et al. (2007)’s finding that
F acebook was more effective in developing weak ties. Second, respondents reported it
was easy to gather a great deal of personal information about Americans by using

Facebook. One respondent mentioned that he or she could “know them well” through

39

“read[ing] personal information”. This paralleled with findings generated from the
present study and many previous studies that Facebook provided a channel to
reducing uncertainty. This finding also indicated the ability of F acebook to help move
superficial relationship forward by promoting mutual knowledge between
acquaintances. Third, respondents reported that Facebook provided an easy way to
interact with people. One respondent said it was easier “to share information and say
hello to a friend on Facebook than by email”. Another mentioned the shielding effect
of Facebook, similar to many other CMC services, that he or she did not “chat with
some Americans face to face” but could “talk many things” on Facebook. Fourthly,
respondents thought that it was easy to keep in touch with Americans through

F acebook, especially when some Americans were not usually “physically around”.
Other comments included “funny and interesting”, and Facebook’s ability to “link
friends of many others”. Some respondents also mentioned F acebook’s help in
cross-cultural interpersonal communication was limited although it did help.

Most respondents who reported a negative opinion regarding Facebook’s
ability to help develop friendships with Americans (7 out of 9) did not give a specific
reason for its ineffectiveness. One respondent commented, “My age (after 25) decide
that I don't make too young American friends; the fiiends at my age or older don't use
facebook that often”. This respondent perceived that only younger Americans used
F acebook very often. Due to the generation gap between frequent Facebook users and
this respondent, and the infrequent use of F acebook by people of similar age to him or

her, this respondent did not think Facebook helped him develop friendship with

40

Americans. The other respondent reported his or her use of F acebook was only “for
fun” and never to “develop my friendships with others”. This response indicated a
passive way of using Facebook characterized by non-connection usage.

Four respondents reported a neutral opinion regrading F acebook’s ability to
develop fiiendship with Americans. One explained the reason was “not many
American friends on it”. Given the popularity of Facebook among American college
students, the respondent should have a limited number of American fiiends, which
showed the importance of the percent of Americans among Facebook fiiends. Other
respondents implied their acknowledgement of Facebook’s ability to develop
friendships with Americans. For example, one respondent regarded F acebook as “a
supplement to my real world friendship”, and this result lent support to the “offline to
online” communication model proposed by Ellison et al. (2007). Another respondent
mentioned his everyday activity on Facebook was “to check the updates of my friends
and sometimes update my own”, thereby supporting Facebook’s ability of uncertainty

reduction.

Discussion
Result discussion
The researcher will first interpret the descriptive findings presented above.
Chinese students generally have a neutral comment on Facebook-’3 ability of
developing friendship with Americans. They consider Facebook neither efiective nor

ineffective in helping them seeking information about target Americans, develop

41

relationships with new American friends, and get closer to Americans. In addition, the
degree of uncertainty reduction (M =3. 09) and the amount of social capital gained
from Americans with the help of F acebook (M =2. 53) were relatively low. This result
also indicates respondents’ neutral attitude towards Facebook’s ability to help develop
intercultural fiiendship with Americans.

Other results also suggest this neutral attitude. For example, most people had
less than 100 friends on Facebook, and most had less than 40% of Americans among
their F acebook friends. In terms of the range, the average number of Facebook fiiends
was between 51 and 100, and the average percent of Americans among Facebook
friends was less than 20%. If compared with the relevant result about Americans
provided by Steinfield et al. (2008), Chinese students have a smaller number of
friends on Facebook than Americans, who had 223.09 fiiends on average in 2006 and
339.26 friends in 2007.

The reason for Chinese users’ neutral attitude towards Facebook’s capability
of developing friendships with Americans should lie in the difficulty of cross-cultural
interaction. Numerous studies have reflected this difficulty to communicate with a
foreigner. First, URT argued that similarity was positively correlated with uncertainty
reduction, so the more similar two people are, the more easily they can interact
effectively with each other. Thus, if two people come from a similar cultural
background, it is easier for them to have effective interactions and develop close
relationship. Second, interpersonal communication research has found the homophily

principle that people liked to interact with those similar to them, as reviewed above.

42

Moreover, cross-cultural interaction research has found that cultural barrier could
prevent people having effective communication and thus could prevent relationship
development. Therefore, people generally have the inertia to interacting frequently
with others from the same cultural background. As Facebook represents offline
relationships, the ineffective interaction between Chinese and Americans that takes
place offline can also be displayed online. In other words, F acebook does not
diminish but simply represents this difficulty in cross-cultural interpersonal
communication. As a result, respondents had a neutral opinion regarding Facebook’s

ability to help develop the fiiendship with Americans.

Factors that aflectfriending decisions.

The researcher studied what factors affected Chinese’ decision to start a
relationship with Americans on F acebook. The most important factor was having
previous offline contact. The predominance of previous contact in making friending
decisions lends support to the findings of Ellison et al. (2007) and Lampe et al. (2006)
that “offline to online” was a common circumstance of relationship development
through Facebook. The “offline to online” model, or the importance of previous
contact, actually includes two situations. On the one hand, people intend to maintain
relationships that have already been made. On the other hand, people want to know
more about those they have met before, or at least have heard of before. Under the
second situation, Facebook plays the role of reducing uncertainty and developing

superficial relationships to friendships. These two situations parallel with the two

43

main functions of F acebook in relationship development by Ellison et al. (2007):
maintaining existing relationships and seeking information about new ties. The
“offline to online” communication model, as well as the importance of previous
offline contact in affecting fiiending decisions, again reflects that F acebook represents
offline relationships online.

Secondly, sharing similarity was another important factor that affected
people’s friending decisions. This finding lends support to research findings about the
tendency to homophily in interpersonal communication (McPherson et al., 2001).
This finding also suggests the difficulty of cross-cultural communication and supports
the previous research about the impact of similarity on intercultural relationship.

The result that similarity could result in the beginning of a relationship online
indicates the possibility of another communication model in CMC. This relationship
starts online and forms driven by similarity, for instance, similar interest. No matter
whether this relationship ultimately stays online or extends to the offline world, this
relationship may finally move forward to mature and intense fiiendship. The reason
for this speculation is that similarity can create many opportunities for interpersonal
communication and relationship development, such as mutual activities, and
interesting and in-depth conversations. As long as people use appropriate
communication strategies and frequently interact with their online “fiiends” in a
longer period of time, superficial computer-mediated relationship could also develop
to friendship ultimately (Walther, 1992).

Finally, that friending anyone was considered as a slightly insignificant factor

that affected fiiending decisions shows that seeking friendship still serves an
important motive of using Facebook by Chinese students. Friending anyone without
any selection or differentiation can lead to a large number of friends, but that implies
an indifferent attitude toward friendship. Compared with the quality of fiiendship on
Facebook, this kind of users cares more about the number of friends. On the contrary,
if people consider whether to have previous offline contact or share similarity while
making friending decisions, they may hold a more serious attitude towards “Facebook
fiiendship” because they use some criteria to filter some people that they do not think
will become friends with them in the fiiture. Thus, this finding indicates that Chinese

F acebook users are motivated to find out friendship through Facebook.

Uncertainty reduction model.

RQ2 explored the relationship between F acebook usage and the degree of
uncertainty reduction. Friend-related F acebook use pattern and percent of Americans
among Facebook fiiends positively predicted the degree of uncertainty reduction.
What differs from previous research on the impact of using F acebook on uncertainty
reduction like Andon (2007) is that Facebook intensity did not turn out a significant
predictor. This result might result from the special feature of the sample used in the
present research. This sample consists of Chinese only and over half of the
respondents are graduate students. However, previous research all used American
undergraduate students as respondents. The different sample may account for the

insignificant result of Facebook intensity.

45

In terms of particular usage of Facebook, it was friend-related activity that was
positively correlated with uncertainty reduction. The more a user seeks information
about their friends, or tracks their fiiends, or interacts with their friends on Facebook,
the more he or she knows about them. Given the relatively high standardized
coefficients (0.74) produced in the model, friend-related Facebook use pattern plays
an important role in uncertainty reduction.

Facebook does provide many channels to information seeking, tracking, and
interpersonal interaction, such as reading News Feed, checking profiles, and sending
messages. By going through these activities, people can accumulate more knowledge
about their friends. Moreover, the third-party source also provides a great deal of
information that can supplement the knowledge of information seekers about the
target person. The potential problem of uncertainty increase by the multisource and
multimodal nature of Facebook may exist but was not studied in the present research.
This can be a topic for future research. Based on the present research data, it is safe to
conclude that people can reduce uncertainty about their contacts by using Facebook.

Percent of Americans among Facebook fiiends was another significant
predictor of uncertainty reduction. American fiiends can be a channel to
understanding American culture. Chinese can apply the knowledge of other
Americans to a target American to understand the particular target person’s language,
behavior, and thought. Hence, the more American friend Chinese have, the better they
may understand other Americans’ thoughts and behaviors. F urthennore, a higher

percent of Americans among Facebook friends may indicate a higher possibility of

46

network convergence, which means the network of a Chinese student overlaps with
that of his or her American fiiends. Research has shown that network convergence
was valuable to helping reduce uncertainty. People could ask others for the
information about the target person, which was a strategy for uncertainty reduction
summarized by Tidwell and Walther (2002). Hence, in this case, the higher percent
Americans comprised among a Chinese ’3 Facebook fiiends, the more likely he or she
can obtain information about an American from others. Therefore, to sum up, the
more American friends a Chinese student has on Facebook, the more probably he or
she will know his or her American fiiends better. One thing that deserves attention is
the relatively small standardized coefficients (0.26). This result reflects that the
percent of Americans among Facebook friends is not a strong predictor for the

outcome of uncertainty reduction.

Social capital model.

RQ3 discussed the relationship between uncertainty reduction, Facebook
usage and the amount of social capital obtained from Americans. Uncertainty
reduction, self-related Facebook use pattern, percent of Americans among Facebook
friends, and friend-related Facebook use pattern turned out to be significant predictors
for the amount of social capital obtained from Americans.

Facebook intensity, which previous studies have shown to be significantly -
related to social capital (Ellison et al., 2007; Steinfield et al., 2008), turned out to be

insignificant in the current study. This might also a result of the all-Chinese sample in

47

which half are graduates.

Given its highest standardized coefficients (0.56), self-related use pattern
constituted the strongest factor for the amount of bonding social capital. Self-related
use pattern is aimed at letting a person’s own situation known to his or her fiiends.
Thus, the more information disclosed, the more probably this person can expect to
obtain support from his or her fiiends.

Uncertainty reduction, as expected before, positively predicted the amount of
bonding social capital. This result implies the importance of building mutual
knowledge between fiiends. The more knowledge a person had about his or her
friends, the more probably this person could expect to obtain bonding social capital
from fiiends because friends can provide appropriate help based on his or her needs.
However, given the standardized coefficients (0.33), uncertainty reduction only
displayed a moderate correlation with the amount of bonding social capital. Hence,
based on this finding, what matters more seems to be how to use Facebook to let
friends know about the user’s own situation and needs.

Percent of Americans among F acebook fiiends remained important for
obtaining bonding social capital. According to Gudykunst (2005), this variable
suggests the size of a network with Americans and so indicates the possibility of
receiving bonding social capital from Americans. Therefore, the higher percent
Americans comprise among a Chinese student’s network, the more likely he or she
can obtain bonding social capital from Americans. A Chinese cannot expect to receive

bonding social capital from Americans unless he or she has a certain number of

48

American friends.

An unexpected finding is the role of friend-related use pattern in this model.
Its standardized coefficients (-0.44) implies a negative, moderate relationship between
friend-related use pattern and the amount of social capital that can be obtained, which
is a stark contrast to our theoretical argumentation. An explanation may be related to
the scale. Most items of the scale focus on one-sided communication behaviors rather
than reciprocal behaviors except “chat with fiiends”. However, fiiendship is a
reciprocal relationship, and bonding social capital should also exist in this two-way
human relation. Therefore, according to the scale used in the present research, by
doing friend-related activities people provide bonding social capital for their friends
based on the knowledge of their friends’ situation, instead of receiving bonding social

capital from them.

Summary.

The current research produced a number of findings concerning relationship
formation and development between Chinese and Americans via Facebook. First,
from the perspective of friending motivation, most Facebook use by Chinese students
is serious as fiiending anyone was not considered an important factor that affects
people’s fiiending decisions. People use Facebook to find and develop friendships,
rather than merely make meaningless “Facebook fiiends”. This motivation might be
stronger among international students as they have come to a totally strange

environment and so are in great need of friendships. In terms of specific factors that

49

affect fiiending decisions, having previous offline contact was found to be the
foremost factor. This result supports the “offline to online” communication mode on
Facebook by Ellison et al. (2007), suggesting that people tend to move their offline
contacts to F acebook and develop their relationships by using a variety of F acebook
services. In addition, sharing similarity, another factor that affects fiiending decisions,
suggests the possibility of “online to offline” communication model. Finding out
those that share similarity in some aspects to them, people begin to interact with them
on the Internet. Due to similarity, it is easier to find out common topics, have an
interesting and lasting conversation with those “friends”, and conduct mutual
activities, thereby more easily giving rise to a stronger relationship.

Second, Facebook users can enjoy the benefit of uncertainty reduction and
obtain bonding social capital from their fiiends by using Facebook. Facebook offers
people more information about those they have recently met, and thus creates more
opportunities for their further interaction. In the process of fiirther interaction,
uncertainty about fiiends is diminished, thereby creating a foundation for exchanging
bonding social capital between each other. People can also use Facebook to make
their needs and problems known to their fiiends so that their fiiends can offer help to
them based on their situations. Additionally, F acebook may be also effective to
provide bridging social capital. Ellison et al. (2007) proposed that F acebook might
make it easier to activate latent ties, which were defined by Haythornthwaite (2005)
as social network ties “that are technically possible but not yet activated socially” (p.

137). Facebook can let its users gather a great deal of information about other people

50

and link them to more people. As a result, F acebook enables people to identify “those
who might be useful in some capacity” (Ellison et al., 2007; p. 1 162), from whom
people can also obtain social capital. In this case, it is bridging social capital that
Facebook generates for its users.

Thirdly, the present study reflects that Chinese students generally have a
neutral opinion regarding Facebook’s ability to form and develop fiiendship with
Americans. Another result that reflects their neutral is that respondents perceived they
could obtain a greater amount of bonding social capital from Chinese than Americans.
Although the specific reason for this result was not studied, based on previous
research about cross-cultural interpersonal communication, cultural similarity could
be a factor. The other reason might be that Facebook was less effective to generate
bonding social capital than bridging social capital, which Ellison et al. (2007) had

already found. Future research can address this issue further.

Limitations and implications

This study used a snowball sample with a size of 69. This method cannot
produce a representative sample of the whole population, so the sample used in the
present research may be biased. The sample is also in a relatively small size, so it
prevents some possible further analyses and may weaken current findings. In addition,
as fiiendship is reciprocal, it should be better to include American students into the I
sample and examine their opinions regarding Facebook’s ability to help them interact

with Chinese.

51

In terms of methodology, survey can have some problems. For instance,
self-reported data can be biased and easily manipulated by respondents. This can also
affect the result of the present study. Future research can use a better methodology,
such as network analysis and content analysis.

Moreover, the researcher did not distinguish graduate students from
undergraduate students and compared the two groups in terms of Facebook use and
the outcome of using Facebook to form and develop friendship with Americans. This
is because of the limited sample size. There are only 30 undergraduate students and 39
graduate students in the sample. Thus, if they were separated, it might be hard to
produce significant results. Although the present study did not find a significant
difference between graduate and undergraduate students in terms of their usage of
F acebook and the outcome of using Facebook, this might be a result of the
all-Chinese sample. It should be better to distinguish graduate students and
undergraduate students in the future.

Furthermore, the researcher made scales that measure friending factors and
Facebook use pattern. These scales may bias some research findings generated in the
present study. For instance, the scale that measured Facebook use pattern focuses
mainly on one-sided communication, which violates the reciprocal nature of
friendship. As a result, there is still a great deal of space for future improvement of
those scales.

The current study also made some contributions to relevant research. The

researcher introduced the concept of Facebook use pattern to explain the impact of

52

F acebook usage on uncertainty reduction and bonding social capital. The result
supports uses and gratifications theory, and suggests a plausible model about the
impact of F acebook on relationship development. The present study also produced a
scale that studied factors that affect fiiending decisions. These contributions can be
used and further developed in the future research.

In terms of practical implications, Facebook can make some improvement in
its design to become a better tool of social interaction for international students. As
this study found the importance of similarity in relationship formation and also
suggested the importance of cultural similarity in cross-cultural fiiendship
development, Facebook can explore more designs to help international users
understand the American culture and merge cultural gaps, which is important because
F acebook has become more and more international. For example, Facebook can
create web pages for some very popular music bands, movies, TV shows, or activities,
and link them to the personal interest listed on Facebook. On those web pages there
can be an introduction about the popular movies, music and TV shows. Other contents
should also be included, such as users’ comments, photos, and some trial services that
uSers can enjoy to know more about popular cultures. By going to these web pages
and using these services, international users can know more about the host culture and
narrow the cultural gap with Americans. As a result, intercultural communication may
become easier. Furthermore, given Facebook’s potential in facilitating bonding and
bridging social capital, Facebook should consider create more opportunities for

interactions because effective interaction can help generate social capital between

53

people. For example, Facebook can improve group’s design to facilitate interactions
between group members. Since Facebook groups are usually formed based on shared
interest or experience, this similarity provides more opportunities for further
interactions between group members, making more likely to form a close relationship
between group members as well as exchange social capital.

This study is exploratory in examining the impact of Facebook in developing
cross-cultural friendships, so many findings did not go into great depth. This
limitation provides a lot of opportunities for future research. For example, future
research can study why it is difficult to develop intercultural friendships. Proposed in
the present research, “online to offline” communication model deserves further study
with predicted outcome value taken into account. Secondly, this study was restricted
to Chinese only, and more than half of the respondents were graduate students, so
results may not be able to be generalized to other ethnic groups. For example, as
cultural similarity may be a crucial barrier to cross-cultural communication and
F acebook cannot diminish this barrier, it is possible that people from a cultural
background closer to American culture, such as Canadians or British, can have better
interactions with Americans with the help of Facebook. Therefore, future research can
replicate the present study with another sample. Finally, this study focused on bonding
social capital only, but Facebook users may be more likely to obtain bridging social
capital from using Facebook (Ellison et al., 2007). Thus, future study can research the
impact of Facebook use on bridging social capital and how cross-cultural friendship

can help accumulate bridging social capital.

54

Table 1

Appendices

Perceived eflectiveness of Facebook in developing cross-cultural friendships

(Cronbach s alpha =0. 85)

 

Individual items and scale

Mean (M)

Standard Deviation (SD)

 

How effective do you think
Facebook is for helping you
develop relationships with
new American friends?

2.91

1.1

 

How effective do you think
Facebook is for helping you learn
personal information about
Americans you meet at classes,
arties or on other occasions?

3.19

1.26

 

 

How effective do you think
Facebook is for getting closer to
Americans you meet at classes,
parties or on other occasions?

 

2.87

 

1.16

 

Table 2

Social capital people can obtain from Americans (Cronbach Is alpha=0. 88) and
Chinese (Cronbach s alpha=0. 8) respectively by using Facebook

 

Individual items and scale

Americans

Chinese

 

Mean
(M)

(SD)

Standard
Deviation

Mean

(M)

Standard
Deviation
(SD)

 

There are several
Americans/ Chinese on
Facebook I trust to help
solve my problems.

2.84

1.13

3.63

1.21

 

When I’m making an
important decision,
Facebook allows me to
ask advice from
Americans/Chinese about
it.

2.58

1.01

3.29

0.99

 

 

I can always find an
American/Chinese through
Facebook when I feel
lonely.

 

2.31

1.06

 

 

3.12

 

 

55

 

 

Table 2 continued

 

If I needed an emergency
loan of $500, I can find
an American/ Chinese on
F acebook I can turn to.

2.01

1.27

3.04 1.28

 

Among my Facebook
friends, I can find an
American/Chinese

that would put his/her
reputation on the line for
me.

2.62

1.05

3.47 0.99

 

Among my F acebook
friends, I can find an
American/Chinese

who would be good job
reference for me.

2.84

1.07

3.21 1.07

 

Among my Facebook
friends, I can find an
American/Chinese
that would share the
last dollar with me.

2.18

1.15

3.22 1.22

 

 

Among my F acebook
fiiends, I can always find
an American that would
help me fight an injustice

 

2.83

 

1.07

 

3.26 1.1

 

 

Table 3

Uncertainty reduction scale (Cronbach Is alpha=0. 87)

 

Individual items and scale

Mean (M)

Standard Deviation(SD)

 

Information on Facebook makes

me confident of how my

American fiiends will react and behave.

2.91

1.08

 

Information on F acebook helps me

accurately predict the values my

American friends have.

2.82

1.13

 

lnforrnation on F acebook helps me

accurately predict my American

friends’ attitudes.

3.16

1.15

 

 

Information on Facebook helps me

accurately predict my American
friends’ feelings and emotions.

 

3.3

 

1.2

 

56

 

 

Table 3 continued

 

Information on F acebook helps me

get to know my American fiiends well.

3.31

1.2

 

 

Information on F acebook helps me
to be certain that my American
friends like me.

3.02

 

 

1.12

 

Table 4

Factors that aflectfriending decisions (Cronbach s alpha=0. 85 )

 

Individual items and scale

Mean (M)

Standard deviation (SD)

 

Having previous oflline contact
(Cronbach Is alpha=0. 84)

3.14

1.05

 

When I decide whether to be friends
with someone on F acebook, whether
I have heard of this person is

to me.

3.28

1.26

 

When I decide whether to be fiiends
with someone on Facebook, whether
I have met this person face-to-face
is to me.

2.96

1.29

 

When I decide whether to be fiiends
with someone on Facebook, whether

I have talked with this person
is to me.

3.12

1.27

 

When I decide whether to be fiiends
with someone on Facebook, whether
this person is my fiiend in reality

is to me.

3.23

1.3

 

Sharing similarity
(Cronbach s alpha=0. 83)

2.83

0.81

 

When I decide whether to be friends
with someone on Facebook, whether
this person has friends in common
with me is to me.

2.9

1.23

 

 

When I decide whether to be fiiends
with someone on F acebook, whether
this person is in the same network as

me is to me.

 

2.84

 

1.21

 

57

 

 

Table 4 continued

 

When I decide whether to be fiiends
with someone on F acebook, whether
this person shares some interests
with me is to me.

2.97

1.08

 

When I decide whether to be friends
with someone on F acebook, whether
this person comes from the same
social status with me is to
me.

2.52

1.07

 

When I decide whether to be friends
with someone on Facebook, whether
this person shares the same basic
values with me is to me.

3.03

 

When I decide whether to be friends
with someone on Facebook, whether
this person also comes from China
is to me.

2.56

1.16

 

When I decide whether to be friends
with someone on Facebook, whether
this person is a friend of someone I’m
interested in is to me.

3.03

1.2

 

F riending anyone
(Cronbach Is alpha=0. 82)

2.56

0.99

 

I’ll be friends with anyone because
I want the number of my Facebook
friends to be as large as possible.

2.43

1.27

 

I’ll be fiiends with anyone because
having lots of fiiends makes me
look popular.

2.43

1.21

 

I’ll be fiiends with anyone because
then I can see more profiles.

2.65

1.22

 

 

I’ll be friends with anyone because
it’s easier to accept their friend
request than refuse.

 

2.71

 

1.19

 

Table 5

F acebook intensity

 

 

Individual items and scale

 

Mean (M)
or N (n%)

 

Standard
Deviation (SD)

 

58

 

 

Table 5 continued

 

In the past week, on average,
approximately how many minutes
have you spent on Facebook each
day?

27.78

49.33

 

About how many TOTAL F acebook
friends do you have at MSU and
elsewhere?

1- less than 50,

2— 51-100,

3- 101-150,

4- 151-200,

5- 201-250,

6- 251—300,

7- 301—350,

8- 351-400,

9- 401-450,

10-451-500,

11-more than 500

2.13

33 (47.83%)
19 (27.54%)
7 (10.14%)
4 (5.8%)

4 (5.8%)

0

o

1(1.45%)

o

0

1(1.45%)

1.24

 

 

Facebook attitudinal scale
(Cronbach 3' alpha =0. 91)
Facebook is part of my everyday
activity.

I’m proud to tell people I’m on
Facebook.

Facebook has become part of
my daily routine.

I feel out of touch when I
haven’t logged onto

Facebook for a while.

I feel I am part of the

F acebook community.

I would be sorry if

Facebook shut down.

 

2.63

2.61

2.79

2.72

2.26

2.6

2.81

 

0.95

1.22

0.97

1.25

1.17

1.14

1.19

 

Table 6

Facebook use pattern (Cronbach 8 alj

nha=0. 93)

 

Individual items and scale

Mean (M)

Standard Deviation

(SD)

 

Self-related F acebook use pattern
(Cronbach 8 alpha =0. 79)

2.38

0.76

 

 

 

update profiles

2.43

 

0.99

 

 

 

Table 6 continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

update status 2.58 0.99
post new photos/videos/notes 2.38 0.93
post new photos/videos/comments 2.12 0.99
in your Facebook groups

F fiend-related F acebook use 2.57 0.75
pattern

(Cronbach s alpha=0. 89)

send messages 2.49 0.93
check your fiiends’ profiles 2.86 1.09
read News Feed 2.76 1.2
write on your friends’ Walls 2.69 0.94
watch videos/view photos/read 2.68 0.93
notes/your friends’ status and

leave a comment

chat with your fiiends 2.49 0.92
read your F acebook groups’ updates 2.38 0.89
repost videos/photos/notes 2.2 1.01
that you like

Table 7

Demographic information, and the use of the Internet and F acebook (N =69)

 

 

 

 

 

 

Mean (M) or N (%) Standard Deviation (SD)
Gender
Male 27 (39.13%)
Female 42 (60.87%)
Year at school
Freshman 10 (14.49%)
Sophomore 18 (26.09%)
Junior 1(1.45%)
Senior 1(1.45%)
Graduate student 39 (56.52%)
Ethnicity
Chinese, born in China 65 (94.29%)
Chinese, with US citizenship 1 (1.45%)
or permanent residency
Chinese, with other countries’ 1 (1 .45%)
citizenship or permanent
residency
I don’t want to disclose 1 (1.45%)
Length in US (years) 1.66 1.59

 

 

 

60

 

 

Table 7 continued

 

 

 

Length of Internet use 267.22

everyday (minutes)

About what percentage of 1.86 1.75
your total F acebook Friends

are AMERICAN?

1- less than 20%, 40(58.82%)

2— 20%-39%,
3- 40%-59%
4-60%-79%

5-80%-100%

8(11.76%)
11(16.18%)
6(8.7%)

 

3(4.35%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 8

Paired sample T—test of factors that aflectfriending decisions on F acebook

Pairs Mean Standard t (If p
Deviation (SD)

Similarity- 0.28 1.09 2.12 68 <0.05

friending anyone

Similarity- -0.31 0.96 -2.7 68 <0.01

offline contact

Offline contact- 0.59 1.52 3.21 68 <0.01

friending anyone

Table 9

Regression predicting the extent of uncertainty reduction

Independent variables standardized coefficients p

Intercept l .76 * *

F acebook intensity -0.1

Self-related use pattern -O.3

Friend-related use pattern 0.74 **

Percent of Americans 0.26 **

among Facebook friends

Time in US -0.06

Gender 0.18

Undergraduate versus -0.06

Graduate

 

 

 

 

N=69

F=3.84, p<.01, adj. R2 =0.23

 

** p<.01

61

 

 

 

Table 10

Regression predicting the amount of American bonding social capital

 

 

 

 

 

 

Independent variables standardized coefficients p
Intercept 1 .45 * *
Uncertainty reduction 0.33 ***
F acebook intensity -0.08
Self-related use pattern 0.56 **
Friend-related use -O.44 *
pattern
Percent of Americans 0.19 *
among Facebook
fiiends
Time in US -0.04
Gender -O.2
Undergraduate versus -0.05
Graduate

=69 F=5.25, p<.001, adj. R2 =0.34

 

* p<.05, ** p<.01, *** p<.001

62

 

References

Adams, R., Blieszner, R., & de Vries, B. (2000). Definitions of friendship in the third
age: Age, gender, and study location effects. Journal of aging studies, 14(1),
1 17-13 3.

Albrecht, T., & Adelrnan, A. (1984). Social support and life stress. Human
Communication Research, 11(1), 3-32.

Andon, S. P. (2007). Evaluating computer-mediated communication on the university
campus: The impact of Facebook. com on the development of romantic
relationship. Unpublished mater’s thesis, Florida State University, Tallahassee.
Retrieved April 9, 2009, from
http://etd. lib. fsu.edu/theses/available/etd-O 1 052007-1 70005/

Altman, 1., & Taylor, DA. (1973). Social Penetration: The Development of
Interpersonal Relationships. New York, NY: Holt, Rinehart, & Winston.

Bargh, J .A., & McKenna, K.Y.A. (2004). The Internet and social life. Annual Review
of Psychology, 55(1), 573-590.

Berger, C.R., & Calabrese, R. J. (1975). Some explorations in initial interaction and
beyond: Toward a developmental theory of interpersonal communication. In
R.T. Craig, & H.L. Muller(Eds.), (2007), Theorizing communication: Reading
across traditions (pp. 325-338). Thousand Oaks, CA: Sage.

Bochner, S., McLeod, B., & Lin, A. (1977). Friendship patterns of overseas students:
A functional model. International Journal of Psychology, 12(4), 277-294.

Bourdieu, P., & Wacquant, L. (1992). An Invitation to Reflexive Sociology. Chicago:
University of Chicago Press.

Boyd, D. (2006). Friends, Friendsters, and MySpace Top 8: Writing community into
being on social network sites. First Monday, 11(12). Retrieved June 15, 2008
from
http:// www. firstmondaycrg/ issues/ issue 1 1_12/boyd/index.html

Boyd, D., & Ellison, N. (2007). Social network sites: Definition, history, and
scholarship. Journal of Computer-Mediated Communication, 13(1), 210-230.

Bukowski, W., Hoza, B.,& Boivin, M. (1994). Measuring friendship quality during
pre- and early adolescence: The development and psychometric properties of
the friendship qualities scale. Journal of Social and Personal Relationships, 1 I ,
471-484.

63

Centola, D., Gonzélez—Avella, J ., Eguiluz-V., & San Miguel, M. (2007). Homophily,
cultural drift and the co-evolution of cultural groups. Journal of Conflict
Resolution, 51(6), 905-929.

Culnan, M. J. & Markus, M. L. (1987). Information technologies. In F. M. Jablin, L.L.
Putman, K.H. Roberts, & L. W. Porter(Eds.), Handbook of organizational

communication: an interdisciplinary perspective (pp. 420-433). Newbury Park,
CA: Sage

Dirnicco, J., Millen, D., Geyer, W., Dugan, C., Brownholtz, B., & Muller, M. (2006).
Motivations for social networking at work. Proceedings of the 2008 Computer
Supported Cooperative Work. New York: ACM Press.

Ellison, N., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “friends”:
Exploring the relationship between college students’ use of online social

networks and social capital. Journal of Computer-Mediated Communication,
12(3), 1143-1168.

F ehr, B. (2004). A prototype model of intimacy interactions in same-sex fiiendships.
In D. J. Mashek, & A. Aron (Eds), Handbook of closeness and intimacy, (pp.
9-26). Mahwah, NJ: Lawrence Erlbaum Publishers.

F 0110, D., & Raynes-Goldie, K. (2006). Hyperfiiends and beyond: Friendship and
social norms on live journal. In M. Consalvo & C. Haythomthwaite (Eds),
Internet Research Annual Volume 4: Selected Papers from the Association of
Internet Researchers Conference. New York: Peter Lang. Retrieved March 31,
2009, from http://k4t3.org/publications/hyperfiiendship.pgf

Fumham, A., & Alibhai, N. (1985). The fiiendship networks of foreign students: A
replication and extension of the functional model. International Journal of
Psychology, 20, 709-722.

Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology,
78(6), 1360-1380.

Granovetter, M. (1983). The strength of weak ties: A network theory revisited.
Sociological Theory, 1, 201-233.

Gudykunst, W. B. (1985). The influence of cultural similarity, type of relationship,

and self—monitoring on uncertainty reduction processes. Communication
Monographs, 52, 203-217.

Gudykunst, W.B. (2005). An anxiety/uncertainty management (AUM) theory of

effective communication: Making the mesh of the net finer. In W.B.
Gudykunst (Ed.), Theorizing About Intercultural Cultural Communication (pp.

64

281-322). Thousand Oaks, CA: SAGE Publications.

Haythomthwaite, C. (2005). Social networks and Internet connectivity effects.
Information, Communication, & Society, 8(2), 125-147.

Joinson, A.N. (2008) “Looking at”, “looking up”, or “keeping up with ” people?
Motives and uses of F acebook Proceedings from 2008 Conference on Human
Factors in Computing Systems (CHI 2008), 1027-1036. New York: ACM
Press.

Katz, E., Blumler, J .G, & Gurevitch, M. (1973). Uses and gratifications research. The
Public Opinion Quarterly, 37(4), 509-523.

Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V., & Crawford, A.
(2002). Internet paradox revisited. Journal of Social Issues, 58, 49-74.

Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., Scherlis, W.
(1998). Internet paradox: A social technology that reduces social involvement
and psychological well-being? The American Psychologist, 53, 1017-1031.

Lampe, C., Ellison, N., & Steinfield, C. (2006). A Face(book) in the crowd: Social
searching vs. social browsing. Proceedings of the 2006 20th Anniversary
Conference on Computer Supported Cooperative Work. New York: ACM
Press.

Lee, H.O. & Boster, El. (1991). Social information for uncertainty reduction during
initial interactions. In S.T. Ting, & F. Korzenny(Eds.), Cross-cultural
interpersonal communication (pp. 189-212). Newbury Park, CA: Sage.

McKenna, K.Y.A., Green, A.S., & Gleason, M.E.J. (2002). Relationship formation on
the Internet: What’s the big attraction? Journal of Social Issues, 58(1), 9-31.

McPherson, M., Smith-Lovin, L., & Cook, J .M. (2001). Birds of a feather:
Homophily in social networks. Annual Review of Sociology, 27(1), 415-444.

Merkle, E.R. & Richardson, RA. (2000). Digital dating and virtual relating:
Conceptualizing computer mediated romantic relationships. Family Relations,
49(2), 187-192.

Nyland, R., Marvez, R., & Beck, J. (2007). MySpace: Social networking or social
isolation? Paper presented at the AEJMC Midwinter Conference, Reno.

Papacharissi, Z., & Mendelson, A. (in press). Friends, networks, and zombies: The

social utility of Facebook. In S. Papathanassopoulos (Ed.), Directions for
Media Studies in the 21‘" Century. Manuscript submitted for publication.

65

Patterson, B.R., Bettini, L., & Nussbaum, J .F. (1993). The meaning of friendship
across the life span: Two studies. Communication Quarterly, 41, 145-160.

Paxton, P. (1999). Is social capital declining in the United States? A multiple indicator
assessment. Amerian Journal of Sociology, 105(1), 88-127.

Putnam, RD. (2000). Bowling alone. New York: Simon & Schuster.

Quan-Haase, A., & Wellman, B. (2004). How does the Internet affect social capital?
In M. Huysman, & V. Wulf (Eds), Social Capital and Information Technology,
(pp. 113-132). Cambridge, MA: MIT press.

Ramirez, A., & Walther, J. (2009). Information seeking and interpersonal outcomes
using the Internet. In T.D. Afifi, & W.A.Afifi (Eds), Uncertainty, Information
Management, and Disclosure Decisions, (pp.67-84). New York: Routledge.

Rogers, E.M., & Steinfatt, TM. (1999). Intercultural Communication. Prospect
Heights, IL: Waveland Press, Inc.

Steinfield, C., Ellison, N., & Lampe, C. (2008). Social capital, self-esteem, and use of
online social network sites: A longitudinal analysis. Journal of Applied
Developmental Psychology, 29, 434-445.

Tidwell, L. C. & Walther, J. B. (2002). Computer-mediated communication effects on
disclosure impressions, and interpersonal evaluations, getting to know one
another a bit at a time. Human Communication Research, 28(3), 317-348.

Valenzuela, S., Park, N., & Kee, K. (2008). Lessons from Facebook: The eflect of
social network sites on college students ’social capital. Paper presented at the
9th International Symposium on Online Journalism, Austin, TX.

Valkenburg, P. M., Peter, J ., & Schouten, A. P. (2006). Friend networking sites and
their relationship to adolescents' well being and social self-esteem.
CyberPsychology and Behavior; 9, 5 84-590.

Walther, J .B. (1992). Interpersonal effects in computer-mediated interaction: A
relational perspective. Communication Research, 19, 52-90.

Walther, J .B. (1996). Computer-mediated communication: Impersonal, interpersonal
and hyperpersonal interaction. Communication Research, 23(1), 3-43.

Walther, J. B., & Boyd, S. (2002). Attraction to computer-mediated social support. In

C. A. Lin, & D. J. Athin (Eds), Communication Technology and Society:
Audience Adoption and Use (pp. 153-188). Cresskill, NJ: Hampton Press, Inc.

66

Walther, J. B.. & Tidwell. L. C. (1995). Nonverbal cues in computer-mediated
communication, and the effect of chronemics on relational communication.
Journal of Organizational Computing, 5, 355-378.

Walther, J. B., Van Der Heide, 8., Kim, S-Y., Westerman, D., & Tong, S. T. (2008).
The role of friends’ appearance and behavior on evaluation of individuals on

F accbook: Are we known by the company we keep? Human Communication
Research, 34(1), 28-49.

Wellman, B., Haase, A. Q., Witte, J., 8: Hampton, K. (2001). Does the Internet
increase, decrease, or supplement social capital? Social networks, participation,
and community commitment. A rncrican Behavioral Scientist, 45(3), 436-455.

Westemran, D. (2008). How do people really seek information about others?
Information seeking across lntcmet and traditional communication channels.
Journal ofComp:(rer—ll'lcdiatcd Communication, 13, 751-767.

\H’illiams. D. (2006). On and Olft Is Not: Scales for Social Capital in an Online Era.
Journal of Comp:tier-Mediated Communictm'on, 11(2). Retrieved September
21, 2008 from: http://jcrne.indiana.etlur’voll1r'is_sue2/willia_ms.lrtml

 

67

063 3642

   

”iiiiliiiliiililliiiiliiiiiiiil“