<%BANNER%>

Problematic Use of Online Social Networking Sites for College Students

Permanent Link: http://ufdc.ufl.edu/UFE0024085/00001

Material Information

Title: Problematic Use of Online Social Networking Sites for College Students Prevalence, Predictors, and Association with Well-Being
Physical Description: 1 online resource (92 p.)
Language: english
Creator: Spraggins, Andrea
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: addiction, facebook, internet, myspace, networking, problematic, social
Psychology -- Dissertations, Academic -- UF
Genre: Counseling Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Social networking sites like MySpace and Facebook have become such a popular communication application on the Internet that these sites are now some of the most-trafficked websites in the world. Though use of social networking sites has become widespread, little research has been conducted about these sites and their impact on users. Our study attempted to partially address this gap in the literature by examining social networking site use among college students, specifically when the utilization of these types of sites becomes problematic or addictive. Researchers have begun finding evidence for the existence of Internet addiction or problematic use of the Internet in general. However, specific Internet applications which may hold addictive potential have been rarely researched, with the exception of online gambling and gaming. Given the popularity of social networking sites, the aims of our study were to determine if problematic use of social networking sites can develop, and to investigate some possible predictors of problematic use, including social anxiety and a lack of belonging to an offline social network. The less-threatening social environment that social networking sites offer compared to face-to-face interactions may make these sites particularly seductive for socially anxious individuals lacking friends. The relationship of problematic social networking site use with well-being was also examined in this study. An online survey was completed by 367 undergraduates, a majority of which identified themselves as social networking site users. Our study found evidence for the existence of problematic social networking site use by utilizing a measure modified from an existing measure of general problematic Internet use. Social anxiety was found to be positively related with problematic use, but our study failed to find support for lack of belonging to an offline social network as another predictor. However, a model placing loneliness as a mediator of the relationship between social anxiety and problematic use was supported. Finally, our study found evidence for a link between problematic use and well-being. Increased symptoms of problematic use were associated with decreased self-esteem, happiness, satisfaction with life, and increased depression and loneliness.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Andrea Spraggins.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Neimeyer, Greg J.
Local: Co-adviser: Morgan, James I.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024085:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024085/00001

Material Information

Title: Problematic Use of Online Social Networking Sites for College Students Prevalence, Predictors, and Association with Well-Being
Physical Description: 1 online resource (92 p.)
Language: english
Creator: Spraggins, Andrea
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: addiction, facebook, internet, myspace, networking, problematic, social
Psychology -- Dissertations, Academic -- UF
Genre: Counseling Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Social networking sites like MySpace and Facebook have become such a popular communication application on the Internet that these sites are now some of the most-trafficked websites in the world. Though use of social networking sites has become widespread, little research has been conducted about these sites and their impact on users. Our study attempted to partially address this gap in the literature by examining social networking site use among college students, specifically when the utilization of these types of sites becomes problematic or addictive. Researchers have begun finding evidence for the existence of Internet addiction or problematic use of the Internet in general. However, specific Internet applications which may hold addictive potential have been rarely researched, with the exception of online gambling and gaming. Given the popularity of social networking sites, the aims of our study were to determine if problematic use of social networking sites can develop, and to investigate some possible predictors of problematic use, including social anxiety and a lack of belonging to an offline social network. The less-threatening social environment that social networking sites offer compared to face-to-face interactions may make these sites particularly seductive for socially anxious individuals lacking friends. The relationship of problematic social networking site use with well-being was also examined in this study. An online survey was completed by 367 undergraduates, a majority of which identified themselves as social networking site users. Our study found evidence for the existence of problematic social networking site use by utilizing a measure modified from an existing measure of general problematic Internet use. Social anxiety was found to be positively related with problematic use, but our study failed to find support for lack of belonging to an offline social network as another predictor. However, a model placing loneliness as a mediator of the relationship between social anxiety and problematic use was supported. Finally, our study found evidence for a link between problematic use and well-being. Increased symptoms of problematic use were associated with decreased self-esteem, happiness, satisfaction with life, and increased depression and loneliness.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Andrea Spraggins.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Neimeyer, Greg J.
Local: Co-adviser: Morgan, James I.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024085:00001


This item has the following downloads:


Full Text





PROBLEMATIC USE OF ONLINE SOCIAL NETWORKING SITES FOR COLLEGE
STUDENTS: PREVALENCE, PREDICTORS, AND ASSOCIATION WITH WELL-BEING




















By

ANDREA SPRAGGINS


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2009




































2009 Andrea Spraggins









ACKNOWLEDGMENTS

I would like to express my appreciation to a number of people for supporting me in my

dissertation. First, I would like to thank my advisor, Dr. Greg Neimeyer, for all of his guidance

and support of this project. Thanks also go to the members of my committee, Dr. Sondra Smith,

Dr. Jim Morgan, and Dr. Jeff Farrar, who gave their valuable time to this project. My parents and

family also have my deepest appreciation for all of their support and love which helped get me to

this place and continues to keep me grounded. Most important, I thank my husband, Barrett

Spraggins, who has been my rock throughout graduate school. Barrett continues to be my biggest

supporter and I appreciate him dearly.










TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ...............................................................................................................3

LIST OF TABLES ........... ................... ..........................6

LIST OF FIGURES ....................................................... ........7

ABSTRACT ............................ .......................................................8

CHAPTER

1 INTRODUCTION ................................................ .......... .. .. 10

2 LITERATURE REVIEW ......................................................... ...............13

History of Social Networking Sites ................... ......................................14
MySpace ............... ................................................................. ....15
F aceb ook ............... .. ... ........... .... ......................................... 15
Problem atic Internet U se or Internet A ddiction....................................................................16
First aim: Existence of Problematic Social Networking Site Use ........................................22
Second Aim: Predictors of Problematic Social Networking Site Use...............................24
Etiology of Problematic Internet Use.................................................................... 25
Predictors of Problematic Social Networking Site Use.................................................27
Third Aim: Problematic Social Networking Site Use and Well-Being................................30
Study Overview ...................................................... ........34

3 M ATERIALS AND M ETHODS ................................................. ............... 36

Participants ................................................36
Measures ......... .................. ............. .........................................36
Belonging to an Offline Social Network.................................................................. 37
Problematic Social Networking Site Use ............................. ...............38
Social Anxiety .......................................................40
Well-Being ................................................40
Depression......................................................... ........................40
Loneliness........................................41
Self-Esteem ........................................ .........41
Happiness ................................................42
Satisfaction with life.................................... ........42
Methods ............................................................................43

4 RESULTS ..................................................................................................................44

Social Networking Site Use...................................... ......................................44
Examination of the Modified GPIUS ............... .... ........ .............44



4










Prediction 1: Prevalence of Problematic Social Networking Site Use............... ...............46
Prediction 2: Predictors of Problematic Social Networking Site Use ..............................48
Prediction 3: Problematic Social Networking Site Use & Well-Being .............. ...............55

5 DISCUSSION ...................................... .................. .............. ........60

R review of Study Findings ...............................................................60
Social N etw working Site U se...............................................60
Measure of Problematic Social Networking Site Use .................................................61
Prediction 1: Existence of Problematic Social Networking Site Use..............................62
Prediction 2: Predictors of Problematic Social Networking Site Use ............................66
Prediction 3: Problematic Social Networking Site Use and Well-Being .....................70
Study Implications ............ .................. .........................72
Limitations ....................................... ..... ..........74
Future Research ...................................................... ........75

APPENDIX

A INTERPERSONAL SUPPORT EVALUATION LIST (ISEL) BELONGING SCALE....77

B MODIFIED GENERALIZED PROBLEMATIC INTERNET USE SCALE (GPIUS).........78

C SOCIAL AVOIDANCE AND DISTRESS SCALE (SAD) ..............................................80

D OXFORD HAPPINESS QUESTIONNAIRE (OHQ) ...........................................................82

L IST O F R E F E R E N C E S ......................................................................................................... 84

BIOGRAPHICAL SKETCH ................................... ................ .......................92









LIST OF TABLES


Table page

4-1 D distribution of G P IU S Scores...................................................................................... 47

4-2 Correlations Among Problematic Social Networking Site Use (GPIUS scores),
Social Anxiety (SAD scores), and Belonging (ISEL scores) .......................................50

4-3 Means and Standard Deviations of Well-Being Variables for Social Networking Site
Users ............................................... ......... 56

4-4 Correlations Between Well-Being Variables and Problematic Social Networking Site
Use (GPIU S Scores) .......................................... ........ ..............58

4-5 Means and Standard Deviations of Well-Being Variables for Problematic Social
Networking Site Users (GPIUS > 87) and Non-Problematic Users (GPIUS < 88)...........59









LIST OF FIGURES

Figure page

4-1 M edition model ............................................ ............. ..... ... 52









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PROBLEMATIC USE OF ONLINE SOCIAL NETWORKING SITES FOR COLLEGE
STUDENTS: PREVALENCE, PREDICTORS, AND ASSOCIATION WITH WELL-BEING

By

Andrea Spraggins

August 2009

Chair: Gregory Neimeyer
Major: Counseling Psychology

Social networking sites like MySpace and Facebook have become such a popular

communication application on the Internet that these sites are now some of the most-trafficked

websites in the world. Though use of social networking sites has become widespread, little

research has been conducted about these sites and their impact on users. Our study attempted to

partially address this gap in the literature by examining social networking site use among college

students, specifically when the utilization of these types of sites becomes problematic or

addictive. Researchers have begun finding evidence for the existence of Internet addiction or

problematic use of the Internet in general. However, specific Internet applications which may

hold addictive potential have been rarely researched, with the exception of online gambling and

gaming. Given the popularity of social networking sites, the aims of our study were to determine

if problematic use of social networking sites can develop, and to investigate some possible

predictors of problematic use, including social anxiety and a lack of belonging to an offline

social network. The less-threatening social environment that social networking sites offer

compared to face-to-face interactions may make these sites particularly seductive for socially

anxious individuals lacking friends. The relationship of problematic social networking site use

with well-being was also examined in this study. An online survey was completed by 367









undergraduates, a majority of which identified themselves as social networking site users. Our

study found evidence for the existence of problematic social networking site use by utilizing a

measure modified from an existing measure of general problematic Internet use. Social anxiety

was found to be positively related with problematic use, but our study failed to find support for

lack of belonging to an offline social network as another predictor. However, a model placing

loneliness as a mediator of the relationship between social anxiety and problematic use was

supported. Finally, our study found evidence for a link between problematic use and well-being.

Increased symptoms of problematic use were associated with decreased self-esteem, happiness,

satisfaction with life, and increased depression and loneliness.









CHAPTER 1
INTRODUCTION

The need for friendships is a natural part of human existence (Baumeister and Leary, 1995)

and the Internet is a pervasive part of our culture, so it is no surprise that the Internet has become

a source for making friends and maintaining relationships. Social networking sites are the most

recent technological creation on the Internet that serve this purpose. MySpace and Facebook, two

of the most popular online social networking sites, have become an international phenomenon

and an important part of how adolescents and young adults communicate with one another.

Though use of social networking sites has become widespread, little research has been conducted

on the usage and impact of these sites. Our study attempted to partially address this gap in the

literature by examining social networking site use among college students, specifically when the

utilization of these types of sites becomes problematic or addictive.

Like social networking sites, problematic Internet use (PIU), or Internet addiction as it is

called by some researchers, is a relatively new phenomenon that has come to the attention of

researchers in recent years. Literature shows that some individuals can become dependent upon

the Internet and that this can interfere with professional, social, and personal functioning, as well

as negatively impact well-being, much like other types of addiction (Caplan, 2002, 2003; Davis,

2001; Young, 1996a). Most of the PIU literature has focused on Internet use in general without

examining specific applications, though exceptions include research on online gambling

(Griffiths, 2003) and online gaming addictions (Wan & Chiou, 2006). No research has examined

the addictive potential that social networking sites may hold. Social networking sites may be

particularly seductive due to the communicative and interactive features they offer in a less-

threatening social environment. Studies of problematic general Internet use have shown that

individuals with Internet dependency utilize the communication or interactive functions of the









Internet significantly more than non-communicative functions (Morahan-Martin and

Schumacher, 2000; Young, 1996a). Therefore, the first aim of our study was to assess the

prevalence of the problematic usage of social networking sites in college students, a uniquely

vulnerable population to PIU (Kandell, 1998).

A second aim of our study was to investigate some possible predictors of problematic

social networking site use. Studies of problematic use of the Internet in general have shown that

socially anxious individuals may be at greater risk for developing PIU than non-socially anxious

individuals due to the unique and less-threatening environment that the Internet provides for

social interaction (Caplan, 2007; Yen, Ko, Yen, Wu, and Yang, 2007). Given that social

networking sites are a key source of social interaction on the Internet, it was proposed in our

study that socially anxious individuals may be particularly prone to problematic or dependent

usage of these types of sites. In addition to social anxiety, another predictor was also considered

that may mediate the relationship between social anxiety and problematic social networking site

use. It is thought that individuals with higher levels of social anxiety may lack a sense of

belonging to an offline social network due to the anxiety associated with forming face-to-face

friendships. Socially anxious individuals may then turn to social networking sites to compensate

for the lack of social support in their lives, and become dependent on the ability of social

networking sites to fulfill this need. Therefore, our study examined a model in which social

anxiety leads to problematic social networking site use partially through a decreased sense of

belonging to an offline social network.

Finally, the third aim of our study was to examine how problematic usage of social

networking sites relates to levels of well-being in college students. Research has shown a

significant relationship between problematic use of the Internet in general and lower levels of









well-being, including increased levels of depression and loneliness, and decreased self-esteem

(Caplan, 2002; Morahan-Martin & Schumacher, 2003; Young & Rogers, 1998). Our study

sought to replicate this finding with problematic usage of social networking sites.









CHAPTER 2
LITERATURE REVIEW

Social networking sites became a recognized phenomenon on the Internet with the

development of Friendster in 2002. Since this time, a number of other social networking sites,

such as MySpace and Facebook, have been created and are used by millions of people all over

the world. Boyd and Ellison (2007) define social networking sites as "web-based services that

allow individuals to (1) construct a public or semi-public profile within a bounded system, (2)

articulate a list of other users with whom they share a connection, and (3) view and traverse their

list of connections and those made by others within the system." The authors highlight that the

purpose of these sites is primarily to communicate with and make visible to others a person's

social network. Though meeting "strangers" outside of one's existing network is possible on

these sites, the primary way in which these sites are utilized by site members is to maintain

current relationships with friends and family. These sites are differentiated from other Internet

sites like message boards and online communities in that the sites are primarily organized around

people (the user and the user's social network), rather than around a common interest or goal

(Boyd & Ellison, 2007). Social networking sites are more egocentrically structured in that the

user is the center of his or her own community, as opposed to public forums or message boards

which are topically structured.

These websites take the communicative functions encouraged by older technology like e-

mail and AOL instant messenger to another level. Profile pages give users a place to express

their individual personalities, and they can share with the world intimate details of their lives.

Even photographs of the user and the user's friends can be uploaded to profile pages. One of the

most important purposes of profile pages, however, is to provide a place for users to display

friend lists and messages from friends. Other users can view friend profiles and expand their own









network through communication with "friends of friends." Users can send others friend requests

in order to add other users to their own friend list. Friend lists can grow to be quite large, with

many users maintaining lists of friends in triple digits, while others preferring to keep their friend

lists limited to only close, existing friendships. In general, however, most friend lists are

composed of users who have some shared connection, such as a shared class year at a school

(Ellison, Steinfeld, & Lampe, 2007). In recent years, privacy concerns relating to social

networking sites have become an issue covered by the media (George, 2006). To protect the

privacy of the user and the user's friends, most sites provide privacy features that allow for users

to specify who can view their profile and their list of friends.

History of Social Networking Sites

Though Friendster is often given credit as the site that began the online social networking

phenomenon, it was not the first social networking site created. The first of these sites, called

SixDegrees.com, was launched in 1997 (Boyd & Ellison, 2007). Using Six Degrees, site

members could develop personal profiles and friend lists, as well as explore others' friend lists.

By 2000, however, the site shut down due to insufficient members and was declared by the site

founder to be "ahead of its time." The site failed because people simply did not have enough

friends online to promote extensive use of the website. Other social networking sites sprung up

during the late 1990s, including AsianAvenue, BlackPlanet, MiGente, and LiveJournal, but none

enjoyed enough popularity to make much of an impact culturally. That changed with the creation

of Friendster in 2002 (Boyd & Ellison, 2007). Originally designed as a competitor to Match.com,

it grew into a massive social networking site where members could network within their current

social circle as well as meet "friends of friends." The site became hugely popular and was

recognized by the media in the beginning of 2003. A large membership surge as a result of the

media coverage overwhelmed the site and ultimately, the site's popularity declined. Despite the









failure of Friendster, it began a new cultural phenomenon that would forever change the way

millions of people use the Internet and how a generation of adolescents communicate with one

another. From the start of the media recognition in 2003, many new social networking sites were

created, the most successful of which are My Space and Facebook.

MySpace

MySpace is a popular social networking site which was developed in August of 2003

(www.myspace.com). It is currently the world's sixth most popular website and has become a

fixture in popular culture. Until recently, MySpace was the most popular of the social

networking sites, receiving 80% of the visits to online social networking websites. As of January

2008, the site counted 110 million people as monthly active users, with new registrations

occurring at a rate of 300,000 people per day (Owyang, 2008). MySpace began in the United

States but has become an international phenomenon with the creation of versions for China and

the United Kingdom. The premise of MySpace is to bring friends together and to provide a place

for networking and making new online friends. Users post profiles about themselves to create

their own page that other users can visit. Pictures, favorite music, and even videos can be

uploaded in order to personalize profile pages. Each user maintains a friend list, and friends

leave messages for each other that everyone can read. If a user wants to become friends with

another user, it is as simple as sending an online friend request.

Facebook

Facebook is another social networking site that has seen a surge in popularity in the past

year. As of April 2008, Facebook surpassed MySpace to become the most-trafficked social

media website in the world and the fourth most popular website in general (www.facebook.com).

As of July 2008, the site has over 80 million active members worldwide and is growing at a rate

of 250,000 new registrants a day. Founded in February 2004, Facebook was developed by a









Harvard University student who intended it as a way to get to know other students on the

Harvard campus. The site is based upon the paper "facebooks" given to incoming Freshman

students at many universities to help students learn more about one another. Due to the extreme

popularity of the site at Harvard, the creator expanded Facebook to include other universities. As

the demand for the site grew, expansion to high schools and large companies also took place.

The site now has over 55,000 regional, work-related, collegiate, and high school networks.

Facebook use on college campuses is quite popular. Ellison, Steinfield, and Lampe (2007) found

that 94% of the 286 undergraduates sampled in their study were Facebook users. Like MySpace,

users of Facebook can post messages on each other's profile pages and private dialogs can occur

via email-like messages to another user's inbox. In addition, Facebook members can join virtual

groups based on common interests. Users can also search Facebook for other users that have

something in common with them, like a hometown, college, or high school.

Though social networking sites like MySpace and Facebook have gained widespread use

in recent years, little research has been conducted regarding the use of these sites. Our study was

interested in the excessive or problematic use of these sites, including the prevalence, predictors,

and impact of problematic usage of social networking sites for college students. Given the non-

existent research on this topic, the literature on problematic general Internet use or Internet

addiction was examined as a basis for our study.

Problematic Internet Use or Internet Addiction

In recent years, literature has begun to accumulate about the existence of Internet addiction

or problematic Internet use (PIU). Countries like South Korea and China are at the forefront of

the research in this area, as Internet addiction has become a serious issue in these countries

(Block, 2008). The South Korean government considers 2.l1% of its population ages 6-19

(approximately 210,000 children) to be Internet addicts and a large portion of this population









requires treatment, including psychotropic medications and/or hospitalization (Choi, 2007).

Internet addiction is now considered one of the most severe public health issues in South Korea.

The situation in China is not much different, with one report stating that 13.7% of Chinese

adolescents (approximately 10 million) are considered Internet addicts (Block, 2008). In the

United States, however, Internet addiction is only slowly coming to the attention of researchers.

Part of the difficulty in identifying the presence of Internet addiction in this country is due to a

lack of an agreed upon definition, the proper terminology to be used, and the etiology of Internet

addiction. Internet-related problems have been termed Internet addiction (Young, 1996a),

pathological Internet use (Davis, 2001), Internet dependency (Scherer, 1997), and problematic

Internet use (Caplan, 2007; Yellowlees & Marks, 2007). The term utilized is dependent upon

how one defines the Internet-related problems. Griffiths (1998) believed in the term Internet

addiction, as he saw Internet use which interferes with an individual's functioning as a type of

technological addiction. He points out that Internet addiction does not necessarily mean elevated

levels of Internet use, but that some elevated use does turn into an addiction. Kandell (1998) also

defines Internet addiction, which he states broadly as "a psychological dependence on the

Internet, regardless of the type of activity once logged on" (p. 12).

Other researchers propose more detailed definitions of Internet addiction, some developing

assessment instruments based on these definitions. Young (1996a, 1998, 1999) was one of the

first to examine Internet addiction and has become a major figure in this area, even developing a

website (www.netaddiction.com) for the Center for Internet Addiction Recovery. To define

Internet addiction, Young modified the DSM-IV criteria for pathological gambling to develop a

set of criteria which comprises an eight-item Internet Addiction Diagnostic Questionnaire (DQ).

Examples of items include, "Do you feel preoccupied with the Internet (think about previous on-









line activity or anticipate next on-line session)?", "Do you feel the need to use the Internet with

increasing amounts of time in order to achieve satisfaction?", "Have you repeatedly made

unsuccessful efforts to control, cut back, or stop Internet use?", "Do you feel restless, moody,

depressed, or irritable when attempting to cut down or stop Internet use?", and "Have you

jeopardized or risked the loss of a significant relationship, job, educational or career opportunity

because of the Internet?" To be considered an "Internet addict" a person must meet at least five

of the diagnostic criteria in the questionnaire. In order to examine the prevalence of Internet

addiction in the general population, Young (1996a) administered the Internet Addiction DQ to

596 participants recruited through advertisements, including postings on electronic support

groups geared towards Internet addiction and as a search result for those who searched for the

keywords "Internet addiction" on web search engines. The study found that 80% of the

participants could be classified as Internet dependents, calling into question the sensitivity of the

measure and the sampling methods utilized.

Beard and Wolf (2001) modified Young's (1996a) criteria due to what they believed was a

lack of clarity and an improper comparison to a DSM-IV disorder. Instead of pathological

gambling, the authors believed that the DSM-IV substance abuse criteria were more appropriate

for developing the diagnostic criteria of Internet addiction. Unlike Young, the authors were

specific about which five criteria must be met in order to give a diagnosis of Internet addiction.

These include: (1) a preoccupation with the Internet, (2) a need to use the Internet with

increasing amounts of time in order to achieve satisfaction, (3) unsuccessful efforts have been

made to control, cut back, or stop Internet use, (4) when attempting to cut down or stop Internet

use, the person is restless, moody, depressed, or irritable and (5) the person has stayed online

longer than originally intended. In addition, the authors believe that at least one of the following









criteria must also be met in order to make a diagnosis: the person (1) has jeopardized or risked

the loss of a significant relationship, job, educational or career opportunity because of the

Internet, (2) has lied to family members, a therapist, or others to conceal the extent of

involvement with the Internet, and (3) uses the Internet as a way of escaping from problems or

for relieving a dysphoric mood (e.g., feelings of helplessness, guilt, anxiety, depression). The

diagnostic criteria set forth by Beard and Wolf were not intended by the authors to be used in a

self-report format, but rather as a diagnostic interview by a professional.

Griffiths (1998) also defined Internet addiction based on the DSM-IV criteria for substance

abuse. His criteria include symptoms of tolerance, withdrawal, craving, and negative life

consequences, as well as additional symptoms of salience of the activity to the individual,

changes in mood when engaging in the activity, and a tendency to relapse after the activity is

discontinued. Brenner (1997) additionally modified DSM-IV criteria for substance abuse as a

basis for his 32-item true-false Internet-RelatedAddictive Behavior Inventory (IRABI). The

measure includes questions related to negative life consequences and side effects of Internet

addiction such as online relationship problems and resultant time management issues. Culture-

specific measures of Internet addiction have also been developed, mainly for use in China and

Taiwan. These include the Chinese Internet Addiction Scale (CIAS; Chen and Chou, 1999), the

Chinese-translated IRABI (Chou & Hsiao, 2000), and the Internet Addiction Scale for Taiwan

High School Students (IAST; Lin & Tsai, 1999).

Though many researchers define problems with Internet use as an addiction, other

researchers do not believe that the term "addiction" should be associated with Internet use, as

this term is reserved for physiological dependence between a person and a stimulus, such as a

substance. Davis (2001) suggests the term pathological Internet use (PIU) be used instead, and









that dependence be discussed in relation to the Internet rather than addiction. He defines PIU as

consisting of two types: specific pathological Internet use and generalized pathological Internet

use. Specific PIU refers to a content-specific dependency, such as a dependency on online

gaming or online gambling. It is assumed that some form of this dependence would still exist

even in the absence of the Internet (e.g., gambling addiction at casinos instead of online

gambling). Generalized PIU, on the other hand, refers to misuse of the Internet independent of

specific Internet content. It usually involves a dependency on the unique social environment that

the Internet can provide, and therefore the dependence in an alternative form would not

necessarily exist in the absence of the Internet, as it would with specific PIU. Generalized PIU is

characterized by a general sense of "wasting time" online without a clear purpose, or using the

Internet for social functions in order to remain in a virtual social life. Symptoms of PIU include:

obsessive thoughts about the Internet, diminished impulse control, inability to decrease Internet

use, anticipating future online use, less time spent on other pleasurable offline activities, social

isolation, sense of guilt about online use, and a feeling that the Internet is an individual's only

friend.

Caplan (2002) agreed with Davis (2001) that the term addiction should not be associated

with the Internet. He suggested the term problematic Internet use (PIU) to describe the

maladaptive cognitions and behaviors associated with Internet use that result in a negative

impact on academic, professional, and social functioning. Following Davis's work, Caplan

characterized PIU as including cognitive and behavioral symptoms such as mood alteration,

perception of social benefits of the Internet, compulsive use, excessive use, withdrawal, and

perceived social control when interacting with others online compared to face-to-face. Based on

his own definition and on Davis's model, Caplan developed a measure, the General Problematic









Internet Use Scale (GPIUS), to assess PIU. This measure will be discussed in more detail in the

materials section of this paper as it was modified and utilized in our study. Accordingly,

Caplan's term problematic internet use (PIU) will be used for the remainder of this paper to

refer to the phenomenon of "Internet addiction," or the maladaptive cognitions and behaviors

related to Internet use which may interfere with normal daily living (Caplan, 2002). The term

problematic social networking site use will be utilized to refer to a specific problematic or

dependent relationship that a person may develop with these types of sites.

In addition to a lack of clarity as to how to characterize PIU, a debate also exists among

researchers as to how to classify PIU within the existing range of mental disorders. Along with

this, many are considering whether "Internet Addiction" should be added to the next revision of

the Diagnostic and Statistical Manual of Mental Disorders (DSM; APA, 1995). Yellowlees and

Marks (2007), the most recent authors to weigh in on the debate, divide the writing on this topic

into two schools of thought. On one side are the researchers who propose that individuals are

addicted to the use of the Internet in general and Internet addiction should be classified as an

emerging psychiatric disorder in the next revision of the DSM. For example, Marks (1990)

advocates the addition of Internet addiction to the list of other non-chemical behavioral

addictions (e.g., pathological gambling) included in the DSM. On the other side of the debate are

those researchers who believe that individuals can develop a problematic relationship with

specific online applications available via the Internet, such as online gambling, shopping,

chatting, or pornography. Researchers in this camp do not believe that PIU warrants a special

diagnosis (e.g., Internet addiction) because addictions to online content can be classified under

pre-existing DSM classifications, such as pathological gambling or impulse control disorder not

otherwise specified (NOS) (Beard & Wolf, 2001; Shapira, Goldsmith, Keck, Khosla, &









McElroy, 2000; Treuer, Fabian, & Furedi, 2001). Mitchell (2000) believes that a separate

diagnosis is not warranted because a problematic relationship with the Internet could be the

result of an underlying, co-morbid psychological disorder. Griffiths (2000) seconds this by his

assertion that "Internet addicts" do not exist, but that the Internet can provide individuals with a

way to engage in other addictions. In other words, addictions to the Internet (i.e., addictions to

the Internet itself) and addictions on the Internet (i.e., the Internet provides another forum for an

individual to engage in his or her addiction, such as online gambling or gaming) have different

meanings and implications. Yellowees and Marks review the research provided by both schools

of thought and conclude that though researchers in both camps do not deny that individuals can

develop a problematic relationship with the Internet, Internet addiction does not warrant a special

diagnosis as an emerging disorder. This certainly is not the end of this debate and our study did

not attempt to settle the argument. Rather, the purpose of our study was to simply determine if a

dependency on a specific Internet application, social networking sites, can exist.

First aim: Existence of Problematic Social Networking Site Use

Most research has examined dependency on the Internet in general without addressing a

possible dependency on specific Internet applications. Our study adds to the literature in this area

by exploring the possible existence of a dependency specifically on social networking sites. As

previously defined, social networking sites are considered websites on the Internet organized

around people (the user and the user's network) which "allow individuals to (1) construct a

public or semi-public profile within a bounded system, (2) articulate a list of other users with

whom they share a connections, and (3) view and traverse their list of connections and those

made by others within the system" (Boyd & Ellison, 2007). Some research provides justification

for believing that a dependency on these types of sites may exist. When researchers of PIU or

Internet addiction ask study participants what Internet applications they typically use, a









significantly higher number of PIU users note that they use the communicative functions of the

Internet over non-communicative functions. Young (1996a) found that the majority of Internet

users classified as "dependent" by her Internet Addiction DQ use online interactive applications.

Chou and colleagues (Chou, Chou, & Tyan, 1999; Chou & Hsiao, 2000) also found that

"addicted" individuals were more likely to use the Internet for communication functions rather

than for other non-communication uses. Morahan-Martin and Schumacher (2000) reported that

"pathological Internet users" in their study utilize the Internet more for meeting new people,

emotional support, and interactive applications than non-pathological users. These findings seem

to support the idea that communicative or interactive applications of the Internet might be

particularly addictive. It has already been shown that a large majority of Internet users are

utilizing communicative applications like social networking sites, given that sites like MySpace

and Facebook are within the top most visited sites in the world. To emphasize, they are some of

the most visited out of all websites, not just social networking sites (Owyang, 2008). This fact

combined with research supporting the addictive nature of Internet communication applications

suggest that a dependency on social networking sites may exist for some people. Therefore, our

study sought to explore the existence and prevalence of problematic social networking site use.

This was examined in a population of college students because this group has been shown to be

frequent users of social networking sites (Ellison et al., 2007) and vulnerable to PIU (Kandell,

1998, Moore, 1995).

College students may be particularly at risk for developing a problematic relationship with

the Internet. Kandell (1998) believes that college students' developmental stage of solidifying

their sense of identity and forming meaningful and intimate relationships puts them at risk for

making the Internet and its social functions an overly central part of their lives. Moore (1995)









also cited accessibility of the Internet and flexibility of schedules as two factors that may

contribute to increased vulnerability for college students to develop PIU. In addition, most

college students are part of the Net Generation, another group which has been identified as a

vulnerable population to PIU. The Net-geners or Millennials, as they are sometimes called, are

individuals born in the early 1980's or later that are the first generation to grow up with personal

computers and the Internet. BusinessWeek (Hempel, 2005) referred to this generation as the

"MySpace Generation" due to the high use of the website among this cohort. Howe & Strauss

(2000) report that Millennials have higher socio-economic status, more education, and are more

ethnically diverse than any other generation before them. Making up over 30% of the United

States population, this generation is one of the largest, and therefore problems to which this

group may be particularly vulnerable should be of interest to researchers.

Some have proposed that characteristics which define this generation could contribute to a

particularly high vulnerability for PIU. Citing Tapscott's (1998) book about the rise of the Net

Generation, Leung (2004) identified the following cohort characteristics that might make this

generation vulnerable to PIU. These include: (1) a global orientation, (2) being emotionally

uninhibited, (3) having a strong belief in the right to information and learning, (4) being

technologically savvy, and (5) having a preoccupation with maturity and adulthood. For Net-

geners with these characteristics, the Internet's social communication applications may be

particularly seductive. Therefore, using college students of this generation to study the possibility

of a dependency on social networking sites is especially appropriate.

Second Aim: Predictors of Problematic Social Networking Site Use

After establishing the existence of problematic social networking site use, the second aim

of our study was to examine possible predictors of problematic usage. Though many possible

predictors exist, our study examined two: social anxiety and the degree to which a person









belongs to an offline social network. The selection of these predictors was based on the proposed

etiology of problematic general Internet use reflected in the literature as well as research

suggesting an association between these predictors and general PIU.

Etiology of Problematic Internet Use

Etiologies of PIU have been suggested by a number of different authors. For example,

Chou and colleagues (1999) applied Stephenson's (1998) Play Theory of Mass Communication

to explain their version of the etiology of PIU. When applied to Internet use, this theory proposes

that the Internet provides a pleasurable and reinforcing communication experience and that this

reinforcement entices the individual to continuously use the Internet to the point that overuse

leads to addiction-like behavior. Another explanation proposed by Suler (1999) states that an

individual becomes dependent on the Internet due to the Internet's ability to satisfy one or more

unfulfilled needs of the individual. Suler identified six needs that the Internet could be used to

fulfill, including (1) sex, (2) an altered state of consciousness, (3) achievement and mastery, (4)

belonging, (5) relationships, and (6) self-actualization/transcendence of self.

The most detailed explanation of PIU development is given by Davis (2001). He

proposed a cognitive-behavioral model of PIU in which maladaptive cognitions, together with

behaviors that intensify problematic responses, result in PIU. In Davis's model, it is the cognitive

symptoms that lead to the behavioral symptoms which are most commonly emphasized in other

models. The conditions under which PIU occurs are considered within a diathesis-stress model,

in which a vulnerability or psychopathology exists which predisposes a person to PIU

development should a stressor occur. In Davis's model, the stressor which may encourage the

PIU to develop is the introduction of the Internet or a new technology on the Internet (e.g., social

networking sites). By itself, the introduction of this new technology does not cause PIU, but it is

a necessary condition for PIU to occur. Reinforcement of the use of Internet technology is a key









factor which allows PIU to develop. For example, Davis believed that socially isolated

individuals may find the social environment offered by the Internet particularly reinforcing

because social applications on the Internet allow for a non-threatening environment for

communication. If an individual is reinforced for using the online activity by receiving some

kind of positive response, the individual will continue Internet use, and may be conditioned to

use the online activity more and more to receive the desired response.

Davis contends that maladaptive cognitions are the most important factor in PIU

development and maintenance. He divides these maladaptive cognitions into two subtypes -

thoughts about self and thoughts about the world. Thoughts about self usually occur in a

ruminative-type cognitive style, in which the individual frequently thinks about his or her

problematic relationship with the Internet (e.g., trying to figure out why overuse is occurring,

reading about Internet overuse, constantly talking to friends about Internet overuse). The

rumination of thoughts interferes with the person's ability to problem solve and engage in new

behaviors. Hence, the PIU is exacerbated by maladaptive cognitions. Cognitions about self that

maintain PIU also include self-doubt, low self-efficacy, and negative self-appraisal (e.g., "I am

worthless offline, but online I am somebody," "I am a failure when I am offline"). The individual

with these types of cognitions may use the Internet to achieve a more positive view of self

through social acceptance in a less-threatening environment. The Internet is considered less

socially threatening than face-to-face social interactions because it allows for a higher degree of

self-disclosure and risk-taking due to reduced auditory and visual cues, allowance of time to

think about responses, and anonymity (Morahan-Martin & Schumacher, 2000; Peter,

Valkenburg, & Schouten, 2005).









Thoughts about the world are the other maladaptive cognitive subtype proposed by Davis

(2001). These involve the generalization of specific events to global patterns or engagement in

all-or-nothing type thinking (e.g., "Nobody loves me offline," "The Internet is the only place I

am respected"). Both maladaptive thoughts about self and thoughts about the world are

automatically triggered by stimuli associated with the Internet, and are automatically enacted

when one engages in online activities. These maladaptive cognitions therefore lead to PIU and

the maintenance of problematic use.

Predictors of Problematic Social Networking Site Use

The predictors of problematic social networking site use included in our study (i.e., social

anxiety and a lack of belonging to a local social network) fit with these given etiologies of

problematic general Internet use. Caplan (2007) speaks directly to the fit of social anxiety within

Davis's (2001) model of PIU development. He suggests that socially anxious individuals have

more maladaptive cognitions about their social competence than others without these issues.

Citing the self-presentational theory of social anxiety (Schlenker & Leary, 1982), Caplan

explains that social anxiety is created due to a desire to create a positive impression in a social

situation, but a lack of confidence in one's social and self-presentation skills creates anxiety in

social encounters. To reduce the anxiety, socially anxious individuals will turn to low-risk

communicative environments in which they have more of a chance of presenting themselves in a

positive way.

To relate this to Davis's model of PIU development, social anxiety would be considered

the underlying vulnerability or existing psychopathology that predisposes the person to PIU

development. When online communication, such as a social networking site, is introduced, a

socially non-threatening environment is created. For socially anxious individuals, their

maladaptive cognitions about their social skills, along with a desire to make social connections in









a less risky way, may lead to a preference for this online social environment instead of face-to-

face communications. PIU development could then occur as the individual continues to receive

reinforcement, or the gaining of social acceptance and support, by using this less-threatening

online environment. The cognitions and behaviors that drove the socially anxious person to

prefer the online environment also act to maintain online use and the problematic or dependent

relationship the person has formed with the online social atmosphere.

Caplan examined his model of social anxiety as a contributor to PIU development by

collecting data from 343 undergraduate students. He found significance for a model that

indicated social anxiety as the most important predictor of preference for online communication

and negative outcomes associated with PIU. Further evidence in support of the relationship

between social anxiety and PIU has also been found. Yen and colleagues (Yen, Ko, Yen, Wu, &

Yang, 2007) showed that individuals classified as Internet addicts by the Chen Internet Addiction

Scale (CIAS; Ko, Yen, Chen, Chen, & Yen, 2005) had higher levels of social phobia, as

indicated by the Social Phobia Inventory (SPIN; Connor et al., 2000) than those not classified as

addicts.

Given that social networking sites provide a type of online social atmosphere, it is thought

that these sites too can provide a less-threatening environment in which social support may be

gained. Socially anxious individuals may be drawn to these types of sites and then may develop a

problematic relationship with them through the model outlined by Davis (2001) and Caplan

(2007). Therefore, our study proposed that social anxiety is a predictor of problematic social

networking site use.

Within this prediction is the assumption that socially anxious individuals turn to these sites

to fulfill a need for social relationships due to a lack of currently belonging to an offline social









network. To feel belongingness is a basic need of most human beings. Maslow (1968) ranked it

in the middle of his motivational hierarchy, and Bowlby (1969) made it an integral part of his

attachment theory. The belongingness hypothesis proposes that humans have an inherent need to

form and maintain lasting, positive, and significant relationships with others (Baumeister &

Leary, 1995). Anant (1966) defined belonging as "a sense of personal involvement in a social

system so that persons feel themselves to be an indispensable and integral part of the system" (p.

21). Hagerty and colleagues differentiate two types of belonging psychological and

sociological (Hagerty, Lynch-Sauer, Patusky, Bouwsema, & Collier, 1992). Psychological

belonging refers to the affective state, or a person's perception of fit and valued involvement in a

system. Sociological belonging refers less to the affective state and more to actual membership

in a group or system. This type of belonging is more observable than psychological belonging as

it can be behaviorally detected via group membership. In our study, the measure used to assess

belonging is a measure of sociological belonging, or actual membership in an offline social

network. Literature has shown that socially anxious individuals often avoid forming face-to-face

relationships (Gambrill, 1996), and have fewer friends (La Greca & Lopez, 1998). For socially

anxious people who may avoid face-to-face relationships and therefore lack belonging to an

offline social network, online social networking sites may provide a less-threatening way to

fulfill this need to belong. However, if the need for belonging is only being met through the use

of these online sites, individuals may become dependent on these sites to fulfill this need.

The idea of socially anxious individuals turning to the Internet for need fulfillment is

supported by McKenna and Bargh (2000), who proposed a social compensation hypothesis. This

theory suggests that socially anxious individuals may compensate for their lack of social support

by using the non-threatening environment of the Internet to meet friends. Valkenburg and Peter









(2007) found support for the social compensation hypothesis by showing that socially anxious

individuals used the Internet for online social interactions more than their non-socially anxious

counterparts, and perceived the Internet as a safer place for intimate self-disclosure. Chak and

Leung (2004) also suggest that the Internet can provide a place for individuals to satisfy social

and emotional needs that are unfulfilled by their limited face-to-face social networks. If the

person finds the use of social networking sites fulfilling and reinforcing enough, problematic

usage with social networking sites may develop through cognitive and behavioral mechanisms

associated with this reinforcement, as outlined by Davis (2001). This idea is also suggested by

Suler (1999), who proposed that PIU develops when an unmet need is fulfilled and reinforced

through Internet use. He specifically noted one such need as the need to belong. In addition,

Griffiths (2000) has shown a relationship between using the Internet for a way to compensate for

a lack of social support and "Internet addiction."

In conclusion, our study examined both the degree to which a person belongs to a social

network and social anxiety as possible contributors to problematic social networking site use.

PIU etiologies and research indicating an association between social anxiety and a lack of

belonging suggest that a lack of belonging may mediate the relationship between social anxiety

and problematic social networking site use. Individuals who are socially anxious may have less

of a local social network, and therefore may turn to social networking sites to compensate for an

unfulfilled need to belong. Problematic usage may develop as a result of the reinforcing

fulfillment of this need provided by social networking sites.

Third Aim: Problematic Social Networking Site Use and Well-Being

Finally, because various studies have shown a relationship between PIU and decreased

well-being, the third aim of our study was to examine if this same relationship exists between

problematic usage of social networking sites and college students' well-being. Research









specifically looking at problematic social networking site use and its association with well-being

does not exist. Therefore, the literature showing a relationship between well-being and PIU in

general was reviewed as support of our study's predictions.

A number of studies have found an association between PIU and a range of variables

representing well-being, including levels of depression, loneliness, and self esteem. Yen and

colleagues (2007) examined the relationship of PIU with other psychological disorders, including

depression. PIU was measured using the Chen Internet Addiction Scale (CIAS; Ko et al., 2005).

Depression was assessed using Center for Epidemiological Studies' Depression Scale (CES-D;

Radloff, 1977). These measures were administered to 2,114 high school students in Taiwan.

Close to 18% of the students were classified as "Internet addicts" according to the CIAS. Results

of the study indicated that "Internet addicts" had significantly higher CES-D scores than non-

addicts, indicating that higher levels of depression were associated with PIU. Young and Rogers

(1998) also found a relationship between PIU and depression. They administered the Internet

Addiction DQ (Young, 1996a) and the Beck Depression Inventory-II (BDI-II; Beck, Steer, &

Brown, 1996) to 312 adults recruited through advertisements on the Internet. They found that

those individuals classified as Internet addicts had higher levels of depression than non-addicts.

Finally, Caplan (2002) additionally found an association between PIU and depression. His

Generalized Problematic Internet Use Scale (GPIUS) was used as a measure of PIU and the BDI-

II as a measure of depression. After administering the measures to 386 undergraduates, he found

GPIUS scores to be positively correlated with depression scores, indicating that as PIU increases

so do symptoms of depression.

Loneliness has also been found to relate to PIU. Kraut and colleagues (Kraut, Patterson,

Lundmark, Kiesler, Mukopadhyay, & Scherlis, 1998) found that levels of loneliness increased









with the amount of time spent online. Caplan (2003) also found loneliness, as measured by the

UCLA Loneliness Scale (Russell, Peplau, & Cutrona, 1980), to be positively associated with

PIU, as measured by the GPIUS. Two additional studies found a similar result, with pathological

Internet users or Internet dependents scoring higher on the UCLA Loneliness scale than non-

pathological users or non-dependents (Morahan-Martin & Schumacher, 2000; Nalwa & Anand,

2003), lending further support for a relationship between loneliness and PIU.

Finally, a negative relationship between PIU and self-esteem has also been found in the

literature. Caplan (2002) examined the relationship between PIU, as measured by the GPIUS,

and self-esteem, as measured by the Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1965). He

found GPIUS scores to have a significant negative relationship with RSE scores, indicating that

individuals with higher levels of PIU have lower levels of self-esteem. Niemz, Griffiths, and

Banyard (2005) also found a significant relationship between PIU and low self-esteem. PIU was

measured using the Pathological Internet Use Scale (Morahan-Martin & Schumacher, 2000) and

self-esteem was assessed with the RSE. Three-hundred seventy one undergraduate students

completed the measures. PIU scores were found to be negatively correlated with RSE scores,

indicating that PIU is associated with lower self-esteem. Armstrong, Phillips, and Saling (2000)

found similar results. Self-esteem scores, as measured by the Coopersmith Self-Esteem

Inventory (SEI; Coopersmith, 1991), predicted PIU, as measured by the Internet-Related

Addictive Behavior Inventory (IRABI; Brenner, 1997). Specifically, lower self-esteem scores

predicted greater PIU.

The limitation in many of these studies is the failure to establish causality in the

relationship that has been found between PIU and well-being. It could be that this relationship

exists because those suffering from decreased well-being such as low self-esteem, loneliness,









and/or depression are more likely to develop PIU. As Davis's (2001) model of PIU development

explains, a pre-existing psychopathology or condition could make a person more vulnerable to

PIU development should a stressor occur. However, this relationship could also exist because

PIU negatively impacts an individual's well-being. Davis's (2001) model provides support for

this idea by suggesting that the maladaptive cognitions which are associated with PIU can have

detrimental effects on self-esteem and life satisfaction. In addition, PIU has been shown to

interfere with normal daily functioning in a number of studies. Scherer (1997) found that 13% of

participants admitted that Internet use had interfered with their academics, work performance, or

social lives and 2% of respondents viewed the Internet as having an overall negative effect on

their daily lives. Internet dependents in Young's (1998) study also reported personal, family, and

work problems in relation to PIU. This interference in life functioning caused by PIU could

negatively impact a person's sense of well-being and satisfaction with life.

Given that theories of PIU development and maintenance find support for both directions

in the causal relationship between PIU and well-being, it is most likely that the causal

relationship goes both ways (Chou, Condron, & Belland, 2005; Morahan-Martin & Schumacher,

2003). As explained by Caplan (2003) and Davis (2001), lower well-being may make one

susceptible to PIU development, but the maladaptive cognitions and behaviors that maintain PIU

may worsen these conditions. As well, the interference that PIU can have with daily functioning

in many areas of life, such as academic, work, and family functioning, may have a negative

impact on a person's well-being.

Though research showing a relationship between general PIU and well-being was used to

support the purpose of our study, our study did not attempt to find evidence for a causal

relationship between these constructs. Rather, our study aimed to establish the existence of a









relationship between problematic social networking site use and well-being given that this area

has not been researched previously.

Study Overview

In summary, the main objective of our study was to explore a novel area in the Internet

addiction literature the problematic use of social networking sites like MySpace and Facebook

among college students. The first aim of our study was to establish the existence of problematic

social networking site use by surveying a college undergraduate population. The second aim of

our study was to explore some possible predictors of problematic usage of social networking

sites, including social anxiety and the degree to which a person belongs to an offline social

network. Finally, our study aimed to explore the association of problematic social networking

site use with well-being for college students. Given these study objectives, the following

predictions were set forth:

(1) Problematic social networking site use does exist and the measure used to assess it will
show a range of scores, including scores indicating a high presence of symptoms of
problematic use. This prediction is supported by research indicating that problematic
general Internet users utilize the communicative functions of the Internet more than non-
communicative functions (Chou et al., 1999; Chou & Hsiao, 2000; Young, 1996a). Given
that social networking sites are the most popular social application on the Internet, it is
predicted that for some, a dependency on social networking sites, rather than the Internet in
general, may exist.

(2) The degree to which a person belongs to an offline social network will mediate the
relationship between social anxiety and problematic social networking site use. In order to
find evidence for this, the following sub-predictions must be met:

a. The degree of social anxiety will be positively related to problematic use of social
networking sites. Higher levels of social anxiety will be predictive of higher levels of
problematic social networking site use. This is based on the literature indicating
social anxiety as a predictor of PIU development (Caplan, 2007; Yen et al., 2007).

b. The degree to which a person belongs to an offline social network will be negatively
related to problematic use of social networking sites. Less belonging will be
predictive of higher problematic social networking site use. This is based on the
social compensation hypothesis (McKenna & Bargh, 2000) that individuals with a









limited social network may turn to the Internet to compensate for this, and that this
need fulfillment may result in PIU (Griffiths, 2000; Suler, 1999).

c. The degree of social anxiety will be negatively related to the degree to which a
person belongs to an offline social network. More socially anxious people will have
a decreased sense of belonging to an offline social network. This is based on the
literature indicating that socially anxious individuals often avoid forming face-to-
face relationships (Gambrill, 1996) and have fewer friends (La Greca & Lopez,
1998).

d. By including the variable of belonging to an offline social network, the relationship
between social anxiety and problematic social networking site use will be
significantly decreased, indicating that the degree to which a person belongs to an
offline social network mediates the relationship between these two variables.

(3) Problematic use of social networking sites is expected to be related to decreased well-
being, based on the literature linking general PIU to decreased well-being (Caplan, 2002;
Morahan-Martin & Schumacher, 2003; Young & Rogers, 1998). Specifically, a significant
negative relationship is expected between problematic use of social networking sites and
the well-being indicators of self-esteem, satisfaction with life, and happiness, and a
significant positive relationship is expected between the problematic social networking site
use and the well-being indicators of depression and loneliness.









CHAPTER 3
MATERIALS AND METHODS

Participants

Undergraduate students attending the University of Florida were recruited for participation

in our study due to this type of population's unique vulnerability to PIU that is noted in the

literature (Kandell, 1998; Leung, 2004; Moore, 1995), as well as the ease of recruiting this

population. Three hundred sixty-seven students completed the online survey. Of these 367

students, approximately 95% (N = 350) were social networking site users. The ethnicity make-up

of this sample consisted of approximately 13% African-American (N = 49), 9% Asian/Pacific

Islander (N = 34), 13% Hispanic/Latino (N = 47), 60% Caucasian (N = 221), 1% Middle

Eastern/Arab (N = 4), and 3% Biracial/Multiracial individuals (N = 12). Eighty-one percent (N=

297) of the participants were female and 19% (N = 70) were male. No significant gender or

ethnic differences were found between social networking site users and non-users. The

participants ranged in age from 17 to 28, with the average age being 20 years (SD = 1.60).

Students classified as Juniors made up the largest portion of the sample (34%, N = 125),

followed by Sophomore students at 26% (N = 95), Senior students at 21% (N = 77), and

Freshman students at 19% (N = 70). The majority of students live off campus (71%, N = 261).

All participants were treated in accordance with the "Ethical Principles of Psychologists and

Code of Conduct" (American Psychological Association, 1992).

Measures

Eight measures were used in our study to assess the following: the degree to which a

person belongs to a local social network, problematic social networking site use, social anxiety,

and the well-being constructs of self-esteem, happiness, satisfaction with life, loneliness, and

depression. A short demographic survey was also employed to provide information on the









characteristics of the study participants, including age, academic classification (e.g., Freshman),

race/ethnicity, and gender. In addition, questions were included that explored the participants'

social networking site use. These questions included an assessment of the type of social

networking site(s) most commonly used by the participant (i.e., MySpace, Facebook, Friendster,

or Other), the length of time the participant has been a member, for what purpose the participant

uses social networking sites, and the average amount of time the participant uses these sites.

Amount of site use was assessed with the questions: (a) "How many days per week do you

usually visit a social networking site such as MySpace or Facebook?", (b) "On a typical day,

how many times do you visit social networking sites, such as MySpace or Facebook?", and (c)

"If you visit one or more of these sites, how many minutes do you usually stay on the site each

time?" These items are based on the items used by Valkenburg and colleagues (2006) to assess

participants' use of the social networking site, CU2. In addition to these questions, two items

were given at the end of the survey which asked: (a) "Has anyone ever told you that your use of

online social networking sites like MySpace or Facebook is problematic or interferes with your

life (daily functioning)?", and (b) "Do you feel that your use of online social networking sites

like MySpace or Facebook is problematic or interferes with your life (daily functioning?)"

Belonging to an Offline Social Network

The Interpersonal Support Evaluation List (ISEL; Cohen & Hoberman, 1983) was used to

assess the degree to which a person belongs to an offline social network. This is multi-

dimensional measure which has two forms a 40-item version for the general population and a

48-item version for college students. Only one of the subscales is relevant for use in the current

study Belonging. This scale is comprised of 12 items which assess the extent to which an

individual feels a part of a social group with common interests or the perceived availability of

friends which one can spend time with. In the case of the college version of this measure, it is the









extent to which an individual feels a part of a social group on campus. Refer to appendix A for a

full list of items. Respondents are asked to rate each item based on their level of agreement with

the statement. Answer choices include: "Mostly False" and "Mostly True." A score of 0 or a

score of 1 is given, depending on whether the statement is worded positively or negatively. Total

scores can range from 0 to 12, with higher scores indicating more belongingness to a social

network. The measure has been shown to have good internal consistency, with estimates ranging

from 0.88 to 0.90 (Cohen & Hoberman, 1983).

Problematic Social Networking Site Use

The Generalized Problematic Internet Use Scale (GPIUS; Caplan, 2002) was modified to

examine levels of problematic social networking site use among study participants. This measure

is a 29-item self-report questionnaire based on Davis's (2001) cognitive-behavioral model of

PIU. It measures the prevalence of cognitive and behavioral symptoms of PIU derived from this

model, as well as the degree to which the problematic use interferes with the individual's

functioning in personal, academic, and professional areas of his or her life. The measure was

modified to address PIU specifically with social networking site use, or the degree to which the

respondent has a problematic or dependent relationship with their use of social networking sites.

This was done by replacing the word "Internet" or "online" with the words "social networking

sites." For example, the item "When not online, I wonder what is happening online" was

changed to "When not on a social networking site, I wonder what is happening on that site." A

full list of items is included in Appendix B. Items encompass seven areas which assess (a) mood

alteration (i.e., the extent to which the individual uses social networking sites to change affective

states), (b) perceived social benefits (i.e., the extent to which an individual perceives social

networking sites as providing greater social benefits than face-to-face communication), (c)

perceived social control (i.e., the extent to which an individual perceives social control when









using social networking sites for communication), (d) withdrawal (i.e., the degree of difficulty in

staying away from using social networking sites), (e) compulsivity (i.e., the inability to control,

reduce, or stop social networking site use, along with feelings of guilt about lack of control), (f)

excessive social networking site use (i.e., the degree to which an individual feels he or she

spends too much time using social networking sites, or more than the planned amount of time),

and (g) negative outcomes (i.e., the severity of personal, social, and professional problems as a

result of social networking site use). Each item asks respondents to rank their agreement with the

statement on a 5-point Likert scale, from 1 (Strongly Disagree) to 5 (Strongly Agree). Scores

may range from 29 to 145. The higher the total score, the greater the degree of problematic social

networking site use. A specific cut-off score which would indicate problematic use was not

designated for this instrument. Caplan did not believe in proposing a cut-off score defining a

problematic Internet user versus a non-problematic user, as the theory on which the measure is

based (i.e., Davis's (2001) cognitive-behavioral model) stated that the adaptive versus

maladaptive nature of Internet use is dependent upon the individual and the effects the use has on

the individual's life.

Caplan (2002) developed the GPIUS by utilizing specific examples of PIU cognitions,

behaviors, and outcomes proposed by Davis's theory, as well as including items from other

measures of PIU in the literature (Armstrong et al., 2000; Morahan-Martin & Schumacher, 2000;

Scherer, 1997; Young, 1998). In order to identify the factor structure, Caplan conducted an

exploratory factor analysis on data he collected from an administration of the GPIUS to 386

undergraduates, resulting in a final list of items which comprise the seven areas outlined above.

The seven areas identified were highly consistent with those cognitions, behaviors, and outcomes

proposed by Davis's model of PIU. Reliability analyses indicated high internal consistencies,









with alpha coefficients ranging from 0.78 to 0.85 for the seven subscales. Validity support is

indicated by the significant relationship between GPIUS scores and measures of depression,

loneliness, and self-esteem (Caplan, 2002, 2003) in a direction consistent with Davis's (2001)

model of PIU.

Social Anxiety

The Social Avoidance and Distress scale (SAD; Watson & Friend, 1969) was used in our

study to assess an individual's degree of social anxiety. This is a 28-item scale in a true-false

format which assesses feelings of distress/discomfort and avoidance of social interactions.

Examples of items include, "I am usually at ease when talking to someone of the opposite sex,"

and "I try to avoid formal social occasions." Refer to Appendix C for a full list of items. This

scale was intended as a measure of comfort in face-to-face social interactions, and this was

emphasized in our study by instructing respondents to answer the items in reference to their face-

to-face social interactions, not online social interactions. Higher scores on this measure indicate

higher levels of social anxiety. Internal consistency estimates have ranged from 0.77 to 0.93

(Watson & Friend, 1969; Caplan, 2007).

Well-Being

Multiple measures were used to assess well-being in our study. These included measures

of depression, loneliness, self-esteem, happiness, and satisfaction with life.

Depression

The Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977) was

utilized to assess depression levels for our study's participants. This measure was designed for

use with a general, non-clinical population and assesses the current frequency of depressive

symptoms for an individual. The 20-item questionnaire emphasizes depressed affect or mood,

psychomotor retardation, loss of appetite, sleep disturbance, and feelings of guilt, worthlessness,









helplessness, and hopelessness. For each item, the respondent is asked to indicate how

frequently, in the past week, he or she has experienced the symptom expressed by the item.

Response choices range from rarely or never (less than 1 day), some or a little of the time (1 or 2

days), occasionally (3 or 4 days), and most or all of the time (5 up to 7 days). A score of 0 is

given for the first response choice up to a score of 3 for the last response choice. Positive items

are reverse scored. Total scores range from 0 to 60, with higher scores indicating the presence of

more symptoms of depression. The measure has a well-established reliability (Hann, Winter, &

Jacobsen, 1999). An internal consistency of 0.85 has been found for the general population

(Radloff, 1997). Validity has been established by finding significant correlations between the

CES-D and other measures of depression, including the BDI-II (Shafer, 2006).

Loneliness

The third version of the UCLA Loneliness Scale (Russell, 1996) was used to measure

levels of loneliness. The scale contains 20 positively and negatively worded items that assess an

individual's experience of loneliness. This revised measure is a simplified adaptation of the older

version due to complaints about the readability of some of the items (Russell, 1996). Possible

item responses range from 1 (never) to 4 (always). Scores may range from 0 to 80, with higher

scores indicating a higher degree of loneliness. Good reliability with college students has been

established (Cronbach's alpha = 0.92) (Russell, 1996). Validity with college students has also

been indicated by the significant positive correlations found between the UCLA Loneliness Scale

and other measures of loneliness, as well as significant negative correlations with measures of

social support in an undergraduate student population (Russell, 1996).

Self-Esteem

The Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1965) is a widely used, 10-item self-

report measure that was utilized in our study to assess self-esteem. The scale items assess an









individual's perception of general self-worth or positive self-esteem (e.g., "I feel that I'm a

person of worth, at least on an equal plane with others"). Each item has a 4-point Likert scale in

which the participant is asked to rate his or her agreement, ranging from 0 (strongly disagree) to

3 (strongly agree). Reverse scored items are also included (e.g., "At times, I think I am no good

at all"). A score is calculated by summing the points for each item, with scores ranging from 0 to

30. The higher the score, the higher the self-esteem. Construct and convergent reliability of the

measure has been demonstrated (Goldsmith, 1986). Internal consistency estimates have ranged

from 0.82 to 0.93 (Goldsmith, 1986). Validity has been established through findings of a

correlation between RSE scores and depression (Rice, Ashby, & Slaney, 1998) as well as other

constructs (e.g., anxiety) in expected directions (Goldsmith, 1986; Rosenberg, 1965).

Happiness

Personal happiness was assessed using the Oxford Happiness Questionnaire (Hills &

Argyle, 2002). This scale contains 29 positively and negatively worded items meant to broadly

measure personal happiness. Participants rate on a 6-point Likert scale their degree of agreement

with each of the items in the scale. Higher scores indicate higher levels of personal happiness.

Sample items include "I feel that life is very rewarding" and "I do not have a particular sense of

meaning and purpose in my life." Refer to Appendix D for a full list of items. Studies have

shown the scale to be correlated in the expected direction with other measures of subjective well-

being. Reliability estimates have averaged around 0.91 (Hills & Argyle, 2002).

Satisfaction with life

The Satisfaction With Life Scale (SWLS; Diener, Emmons, Larsen, & Gri n, 1985) was

used to measure overall satisfaction with life. This scale contains 5-items that are used to assess

participants' overall judgment about their life satisfaction. Participants rate their agreement with

items on a 7-point Likert scale. Scores can range from 5 to 35, with higher scores indicating









greater life satisfaction. A 2-month test-retest correlation coefficient of .82 has been found

(Diener et al., 1985). Reliability estimates have ranged from .78 to .95 (Diener et al., 1985;

Vassar, 2008). Positive correlations have been found between the SWLS and other measures of

subjective well-being, as well as negative correlations with measures of psychopathology

(Diener et al., 1985).

Methods

An online survey development program, Survey Monkey (www.surveymonkey.com), was

used to develop an online survey for our study. The survey contained the demographic questions,

items concerning social networking site use, and the measures of PIU, social anxiety, belonging,

and well-being. Studies have shown that converting paper-based instruments to Internet-based

measures does not distort the validity, reliability, or factor structure of the instruments (Yu & Yu,

2007). In addition to the measures, an Institutional Review Board (IRB) approved consent form

was included in the online survey, and participants could not view the rest of the survey without

giving their consent to voluntarily participate.

To recruit participants, a link to the online survey was given to professors and graduate

student instructors in the Department of Psychology at the University of Florida. The professors

and graduate students were asked to distribute the link to their undergraduate classes through

class listserves or in-class announcements. Four graduate students and two professors agreed to

allow their students to participate. Participation in the study was on a completely voluntary basis.

All instructors used the online survey as an opportunity for extra credit for their students. An

alternative extra credit option was also offered to the students so as to not violate ethical

recruitment practices.









CHAPTER 4
RESULTS

Social Networking Site Use

In our study, 95% (N = 350) of participants identified themselves as social networking site

users. Among these users, a very large majority are members of the social networking site

Facebook (97.8%). Members of MySpace made up 56% of the sample and members of

Friendster made up less than 2% of the sample. Less than 1% of the sample were members of a

social networking site other than the three listed. The site names given were LiveJournal (N = 3),

Xanga (N = 1), Buzznet (N = 1), and HI5 (N = 1). It was also found that the majority of social

networking site users have been members for over a year (92.5%). Only 2.2% of the sample

reported being members for less than 6 months. As to the primary purpose for using these sites,

most students acknowledged they primarily use the site to talk to existing friends already in their

offline social network (76.6%). Approximately 22% reported that they primarily use the site for

both talking to existing friends and to make new friends. Less than 1% of the sample (N=3)

indicated that they use the site primarily to make new friends or meet new people. Time spent on

these social networking sites was also examined. Approximately 69% of social networking site

users indicated that they use a social networking site every day. The majority check the site 1 to

2 times per day (40.7%), though approximately 13% of the sample reported using the site 10 or

more times a day. Most users spend about 10 to 15 minutes on the site (43.2%), while 20%

report using the site for 30 minutes or more each time they log on. Two participants stated that

they "never turn it off' and constantly check the site.

Examination of the Modified GPIUS

Because the GPIUS measure was modified in our study to assess problematic social

networking site use, rather than problematic Internet use in general, reliability and validity were









explored for the measure in our study. A reliability analysis of the modified GPIUS measure

revealed an internal consistency of 0.92, indicating that with this sample, the items in the

measure are highly intercorrelated. A possible conclusion is that the items are related by the

common latent construct of problematic social networking site use.

To provide an indication of validity, GPIUS scores were examined in relation to

participants' responses to the questions: (a) "Has anyone ever told you that your use of online

social networking sites like MySpace or Facebook is problematic or interferes with your life

(daily functioning)?", and (b) "Do you feel that your use of online social networking sites like

My Space or Facebook is problematic or interferes with your life (daily functioning?)" If the

measure indeed assesses problematic social networking site use, it is expected that those who

answered yes to either of these questions would have higher scores on the GPIUS than those who

answered no. Analyses were conducted to compare the GPIUS means for the group who

answered yes for either question to the group who answered no. On average, those who answered

yes to the first question (M= 76.31, SD = 13.12) had higher GPIUS scores than those who

answered no (M= 62.32, SD = 15.05). This difference was significant (t = 5.42, p < 0.001),

suggesting that individuals who have had a someone tell them that their use of social networking

sites may be problematic have on average higher GPIUS scores than those individuals who have

not experienced this. For the second question, those who answered yes (M= 77.59, SD = 13.10)

also had higher GPIUS scores than those who answered no (M= 62.36, SD = 14.98). Again, this

difference was significant (t = 5.62, p < 0.001), suggesting that individuals who feel that their

own use of social networking sites may be problematic have on average higher GPIUS scores

than those individuals who do not feel this way. These results indicate that higher GPIUS scores









may be suggestive of a problematic relationship with social networking sites, and therefore lend

validity to the measure utilized in our study.

Prediction 1: Prevalence of Problematic Social Networking Site Use

The first aim of our study was to explore whether a problematic relationship with the use

of social networking sites like MySpace and Facebook exists in a college undergraduate

population. It was hypothesized that a range of scores would be found for the measure of

problematic social networking site use, including scores indicating a high presence of symptoms

of problematic use. To explore this prediction, the scores of the modified GPIUS were examined.

Scores may range from 29 to 145. For users of social networking sites in this sample, GPIUS

scores ranged from 29 to 128, with a mean of 63.79 (SD = 15.45). Scores were normally

distributed. Refer to Table 4-1 for a summary of the score distribution. A definable cut-off score

which would indicate problematic versus non-problematic use was not developed for this

measure so it is impossible to give a definitive classification of problematic users versus non-

problematic users. However, participants showed a range of scores on this measure, with a

number of participants scoring high. Higher scores indicate more "symptoms" of problematic

social networking site use. Some indication of prevalence of problematic use may be gained by

considering that answering neutrally to every item would give a score of 87. In this sample, 5.7%

of social networking site users had scores higher than an 87. Therefore, this portion of the

participants endorsed at least some of the symptoms of problematic social networking site use. A

series of post-hoc analyses were conducted to determine if demographic variables varied for the

group with GPIUS scores above 87 compared to the group with GPIUS scores below 87. The

analyses found no significant difference between the high group and the lower group with

respect to gender, race/ethnicity, and student status.










Table 4-1. Distribution of GPIUS Scores

Score Range Percentage


29-38

39-48

49-58

59-68

69-78

79-88

89-98

99-108

109-118

119-128


Cumulative Percentage


6.3%

10.0%

19.4%

24.8%

23.6%

10.8%

3.4%

1.1%

0.3%

0.3%


6.3%

16.3%

35.7%

60.5%

84.1%

94.9%

98.3%

99.4%

99.7%

100.0%









Providing further evidence of the existence of problematic use of social networking sites,

10.6% of the sample indicated that they had had someone tell them that their use of social

networking sites like Facebook or MySpace was problematic or interfered with their life.

Approximately 9% admitted that they felt their own site use was problematic or interfered with

their daily functioning. Taken together, these results provide tentative evidence for the existence

of problematic social networking site use, and therefore, support for prediction 1.

Prediction 2: Predictors of Problematic Social Networking Site Use

The second prediction of our study hypothesized that the degree to which a person belongs

to an offline social network mediates the relationship between social anxiety and problematic

social networking site use. To test this hypothesis, a series of regression analyses are suggested

by Baron and Kenny (1986). This is the most common method of mediation testing used in

research (Fritz & MacKinnon, 2007). Baron and Kenny (1986) state that a variable functions as a

mediator if it accounts for the relationship between the predictor and outcome variable. In other

words, a mediator helps to explain why a relationship between two variables exists. To determine

if a variable is indeed a mediator, Baron and Kenny propose that three guidelines must be met.

First, the predictor variable must account for variations in the outcome variable. Second,

the mediator also must account for variations in the outcome variable. Finally, when the mediator

is controlled for, the relationship between the predictor and the outcome variable drops

significantly in the case of partial mediation, or becomes non-significant in the case of full

mediation.

In our study, it was hypothesized that the degree to which a person belongs to an offline

social network mediates the relationship between social anxiety and problematic social

networking site use. Three sub-predictions were made in addition to the mediation prediction in

order to meet Baron and Kenny's (1986) guidelines for mediation testing. These were as follows:









(1) the degree of social anxiety is positively related to problematic use of social networking sites,

(2) the degree to which a person belongs to an offline social network is negatively related to

problematic social networking site use, and (3) the degree of social anxiety is negatively related

to the degree to which a person belongs to an offline social network. To determine if these sub-

predictions would be supported, the correlation coefficients relating these variables were

examined. Social anxiety was indicated by SAD scores (M = 6.55, SD = 5.77). Belonging was

indicated by ISEL scores (M = 9.48, SD = 2.13). GPIUS scores indicated problematic social

networking site use. The score distributions for both the SAD and ISEL measures were highly

skewed and therefore before analyses could be conducted, square root transformations were

performed to correct for non-normality. After performing the transformations, skewness and

kurtosis for both measures were within acceptable ranges (Tabachnick & Fidell, 1996). Pearson

product-moment correlations were then examined for the three variables. Table 4-2 displays the

results.

Two of the sub-predictions were supported. Social anxiety was found to be positively

related to problematic social networking site use (r = 0.218, p < 0.001), as well as negatively

related to the degree to which a person belongs to an offline social network (r = -0.377, p <

0.001). However, one sub-prediction was not supported. The degree to which a person belongs to

an offline social network, as indicated by ISEL scores, was not significantly related to

problematic social networking site use, as indicated by GPIUS scores. Because of this, running a

series of regression analyses to test the mediation hypothesis would be inappropriate. Therefore,

prediction 2 was not supported by the results of our study. The degree of belonging to an offline

social network was not found to mediate the relationship between social anxiety and problematic

social networking site use for this sample.









Table 4-2. Correlations Among Problematic Social Networking Site Use (GPIUS scores), Social
Anxiety (SAD scores), and Belonging (ISEL scores)

GPIUS SAD ISEL


GPIUS --- 0.218* -0.042

SAD 0.218* --- -0.377*

ISEL -0.042 -0.377* ---

*p < 0.001









However, it was interesting to note that the well-being variable of loneliness, another

variable included in our study that is closely related to belonging, did show a significant

relationship to problematic social networking site use (r = 0.281, p < 0.001), as well as to social

anxiety (r = 0.564, p < 0.001). Though the constructs of belonging and loneliness do differ, they

are both associated with interpersonal relatedness (Hagerty, Williams, Coyne, & Early, 1996).

These two concepts have been shown to be highly related in the literature (Hagerty et al., 1992),

and in our study, sense of belonging was significantly related with loneliness (r = -.481, p <

.001). In addition, research in the area of general PIU has supported a significant relationship

between loneliness and PIU (Caplan, 2003; Morahan-Martin & Schumacher, 2000; Nalwa &

Anand, 2003). Given the non-existence of research in the area of problematic social networking

site use, we felt it appropriate to explore this other variable of interpersonal relatedness -

loneliness as a mediator of the relationship between social anxiety and problematic use, though

a prediction about this relationship was not included in the original hypotheses. It may be that the

degree of interpersonal relatedness does indeed mediate the relationship between social anxiety

and problematic social networking site use, but the construct of loneliness is a more appropriate

representation of this relatedness than the construct of belonging for this sample of college

students.

A mediation analysis was conducted to determine if loneliness could be considered a

mediator of the relationship between social anxiety and problematic social networking site use.

In this newly hypothesized mediation model (see Figure 4-1), social anxiety (SAD score)

functions as the predictor variable, loneliness (UCLA scores) as the mediator, and problematic

social networking site use (GPIUS score) as the outcome variable. The mediation guidelines

proposed by Baron and Kenny (1986) were followed to test this model.































Figure 4-1. Mediation model. Values reflect standardized coefficients. *p < 0.001









To establish a relationship between the predictor, social anxiety (SAD scores), and the

outcome variable, problematic social networking site use (GPIUS scores), a simple regression

was conducted (path c in figure 4-1). It was found that social anxiety significantly predicted (3 =

.218, p < .001) problematic social networking site use and explained approximately 5% of the

variance (R2 = .048, F(1, 348) = 17.398, p < .001) in GPIUS scores. Given this result, Baron and

Kenny's (1986) first guideline for mediation was met.

A second regression was run to establish a relationship between the mediator of loneliness

and the outcome variable of problematic social networking site use (path b in figure 4-1). UCLA

scores were entered as the independent variable and GPIUS scores as the dependent variable.

Loneliness was found to be a significant predictor (P = .281, p < .001) of problematic social

networking site use and accounted for approximately 8% of the variance (R2 = .079, F(1, 348) =

29.921, p < .001) in GPIUS scores. Therefore, Baron and Kenny's (1986) second guideline for

mediation was met.

Although not required by Baron and Kenny (1986), a third regression was conducted to

establish a relationship between the predictor, social anxiety, and the mediator, loneliness (path a

in figure 4-1). SAD scores were entered as the independent variable and UCLA scores as the

dependent variable. Social anxiety was found to be a significant predictor (P = .564, p < .001) of

loneliness and accounted for approximately 32% of the variance (R2 = .318, F(1, 348) = 162.340,

p < .001) in UCLA scores.

Finally, to satisfy the third guideline for mediation, a multiple regression was conducted

with SAD and UCLA scores as the independent variables and GPIUS scores as the dependent

variable. This was done in order to control for the effect of the mediator, loneliness, on the

relationship between social anxiety and problematic social networking site use (path c' in figure









4-1). A concern with multiple regression analysis is multicollinearity. Therefore, a check of

collinearity diagnostics (i.e., variance inflation factor (VIF) and tolerance) was included in the

multiple regression. VIFs greater than 10.0 and tolerance values less than .02 raise concerns for

multicollinearity (Bowerman & O'Connell, 1990; Menard, 1995). The VIF (i.e., 1.466) and

tolerance (i.e., .682) values were within the acceptable range for the multiple regression in our

study.

The result of the multiple regression indicated that together the predictors explained almost

9% of the variance (R2 = .084, F(1,348) = 15.986, p < .001) in GPIUS scores. According to

Baron and Kenny (1986), if the previously significant relationship between the predictor and the

outcome is no longer significant or has been reduced when the effect of the mediator is

controlled for, this indicates full (in the case of the former) or partial (in the case of the latter)

mediation. The results seem to support full mediation in this model. Examining the standardized

regression coefficients, the relationship between the mediator of loneliness and the outcome

variable of problematic social networking site use (path b) remains significant (P = .232, p <

.001), whereas the relationship between the predictor of social anxiety and the outcome variable

of problematic social networking site use (path c, now c') is now non-significant (P = .087, p =

.162). This result fulfills Baron and Kenny's final guideline for proving mediation. The findings

lend evidence for a model of loneliness as a mediator of the relationship between social anxiety

and problematic social networking site use.

Although Baron and Kenny (1986) did not require any additional steps for establishing

mediation other than those outlined in this paper, many researchers (Frazier, Mortensen, &

Steward, 2005) advocate the use of Sobel's test (Sobel, 1982) to determine if the reduction in the

relationship between the predictor and the outcome variable is significant when the mediator is









controlled for. Sobel's test divides the mediated effect by a standard error term defined by Sobel

to yield a z score. The z score is then compared to 1.96. If the score is greater, the reduction in

the relationship between the predictor and outcome variable is significant at the .05 level and

mediation is indicated. When Sobel's test was applied to the model in our study, the reduction in

the relationship between social anxiety and problematic social networking site use while

controlling for loneliness was significant (z-value = 5.019, p < 0.001), indicating that loneliness

is a mediator of the relationship between social anxiety and problematic social networking site

use.

Though our study did not find support for belonging as a mediator (prediction 2), it was

able to provide evidence that another indicator of interpersonal relatedness loneliness -

contributes to the relationship between social anxiety and problematic use of social networking

sites.

Prediction 3: Problematic Social Networking Site Use & Well-Being

The third main aim of our study was to examine the relationship between the problematic

use of social networking sites and well-being. Specifically, a significant negative relationship

was predicted between problematic social networking site use and the well-being indicators of

self-esteem, happiness, and satisfaction with life, and a significant positive relationship was

expected between problematic social networking site use and the well-being indicators of

depression and loneliness. Means and standard deviations for the well-being measures are

reported in Table 4-3. All five well-being variables were highly skewed and therefore before

analyses could be conducted, square root transformations were performed to correct for non-

normality. After performing the transformation, skewness and kurtosis for all measures were

within acceptable ranges (Tabachnick & Fidell, 1996). Pearson product-moment correlations

were then calculated utilizing GPIUS scores and the scores from the measures assessing the









Table 4-3. Means and Standard Deviations of Well-Being Variables for Social Networking Site
Users

Well-Being Variable Mean Standard Deviation


Satisfaction With Life (SWL) 25.37 5.50

Loneliness (UCLA) 39.58 8.96

Depression (CES-D) 12.91 9.18

Self-Esteem (RSE) 21.84 4.84

Happiness (OHQ) 128.99 19.20

Social Anxiety (SAD) 10.70 6.31









well-being variables. Correlation coefficients are reported in Table 4-4. One-tailed significance

tests were utilized due to the prediction of relationship directions. The results fully support

prediction 3 of our study. GPIUS scores were found to be significantly correlated with the

variables of well-being in the expected directions. A higher degree of problematic social

networking site use (GPIUS scores) was found to be related to lower well-being, as indicated by

greater levels of loneliness and depression, and lower levels of happiness, self-esteem, and

satisfaction with life. All correlation coefficients indicated a moderate effect size (defined as r =

.30; Cohen, 1977) for the relationship of the well-being variables to problematic social

networking site use, with the exception of the coefficient for satisfaction i/ ilh life, which

indicated a small effect size (defined as r = .10; Cohen, 1977).

A series of post-hoc analyses were conducted to explore how the mean scores on the

well-being variables for the group considered problematic social networking site users (i.e.,

individuals with GPIUS scores greater than 87) compared to the group with GPIUS scores less

than 87. The means and standard deviations of the well-being variables for the group of

problematic social networking site users and non-problematic users can be found in Table 4-5.

Greater well-being is indicated by higher scores on the satisfaction i/ ilh life, self-esteem, and

happiness variables, as well as lower scores on the depression, loneliness, and social anxiety

variables. Problematic users significantly differed from non-problematic users on the well-being

variables of loneliness (t =4.20, p < 0.001), depression (t = 3.70, p < 0.001), self-esteem (t = -

2.41, p < 0.05), happiness (t = -2.82, p < 0.01), and social anxiety (t = 3.27, p = 0.001),

indicating that problematic users have lower levels of well-being than non-problematic users.

The two groups did not significantly differ on the variable of satisfaction ii i/h life.









Table 4-4. Correlations Between Well-Being Variables and Problematic Social Networking Site
Use (GPIUS Scores)

Well-Being Variable Problematic Social Networking Site Use (GPIUS scores)


Satisfaction With Life (SWL) -0.134*

Loneliness (UCLA) 0.281**

Depression (CES-D) 0.281**

Self-Esteem (RSE) -0.299**

Happiness (OHQ) -0.282**

*p<0.01 **p< 0.001









Table 4-5. Means and Standard Deviations of Well-Being Variables for Problematic Social
Networking Site Users (GPIUS > 87) and Non-Problematic Users (GPIUS < 88)

Well-Being Variable Problematic User Non-problematic User


Satisfaction With Life (SWL)

Happiness (OHQ)

Self-Esteem (RSE)

Loneliness (UCLA)

Depression (CES-D)

Social Anxiety (SAD)


23.68 (SD 5.15)

117.04 (SD 19.53)

19.26 (SD 4.50)

47.79 (SD 7.89)

20.36 (SD 10.32)

10.70 (SD 6.37)


25.47 (SD 5.51)

129.68 (SD 18.99)

21.99 (SD 4.82)

39.11 (SD 8.80)

12.48 (SD 8.94)

6.31 (SD 6.65)









CHAPTER 5
DISCUSSION

Our study's overall aim was to explore the topic of college students' usage of social

networking sites and how site use may become problematic or "addictive." The predictions put

forth in our study were primarily based on the limited research that has been conducted in the

area of general problematic Internet use or Internet addiction. It was the goal of our study to add

to this existing literature by examining whether a dependency on a specific Internet application -

social networking sites could exist. In addition, possible predictors of problematic social

networking site use were explored, as well as the relationship of problematic use with well-being

for college students.

Review of Study Findings

Social Networking Site Use

The percentage of participants that identified themselves as social networking site users in

our study came as no surprise given the popularity of these sites that has been reported in the

media (Owyang, 2008). Our study confirmed that social networking sites are a prominent part of

social communication among college students. Additionally, given that a large majority of study

participants indicated that they have been social networking site users for over a year, it is clear

that these sites have embedded themselves in the university culture. They have also become a

part of the college student's daily life, as a majority of respondents use one or more social

networking sites every day, at least one to two times per day, for about 10 to 15 minutes at a

time. Also, the recent surge in popularity of the site Facebook over the site MySpace was

supported by our study results, with almost all of the study participants listing themselves as

Facebook members, while only half acknowledged a membership in MySpace. Taken together,









these results certainly support the idea that this generation of youths is a wired one and that the

term Net Generation or MySpace Generation (Hempel, 2005) is well earned.

Another interesting finding was that the majority of study participants reported using social

networking sites for staying in touch with existing friends already in their offline social network.

This finding adds to the Internet literature which already has suggested the communicative

applications of the Internet are used more for maintaining relationships rather than meeting

strangers. In a survey of United States Internet users, D'Amico (1998) reported that 87% of the

1,001 respondents indicated that they use the Internet frequently to keep in touch with friends

and family. College students in particular spend time talking to friends and family online.

Scherer (1997) found that 98% of college student participants in his study reported using the

Internet for maintaining relationships with family and friends. Additionally, Gross (2004) found

that the majority of adolescent Internet users used the communication functions of the Internet to

talk with already existing, offline friends rather than strangers. Specific to social networking

sites, Ellison and colleagues (2007) reported that the Facebook users in their study indicated

using the social networking site for communicating with people with whom they share an offline

connection significantly more than for meeting new people.

Measure of Problematic Social Networking Site Use

Our study also made a contribution towards the measurement of the problematic use of a

specific Internet application (i.e., social networking sites), rather than problematic use of the

Internet in general. The modified measure utilized in our study to assess problematic social

networking site use was found to have high internal consistency. The validity of the measure was

indicated by showing that those who felt they had a problematic relationship with social

networking sites or who had been told that they may have a problematic relationship showed

significantly higher scores on the modified measure than those who did not feel or who had not









been told they had a problematic relationship. In addition, the measure's correlation with

constructs such as social anxiety and well-being measures in the expected directions (Caplan,

2002, 2007; Morahan-Martin & Schumacher, 2000; Nalwa & Anand, 2003; Yen et al., 2007;

Young & Rogers, 1998) also provided evidence for the validity of this measure. To our

knowledge, no other study has examined problematic use of social networking sites, and

certainly a measure to assess it has not been created previously. Therefore, the modification of

the GPIUS to create an assessment of problematic social networking site use in our study is

novel and could potentially contribute to future assessment in this area.

Prediction 1: Existence of Problematic Social Networking Site Use

The first and most important aim of our study was to explore whether a problematic

relationship with social networking sites may exist in a population of college students. The range

of scores found on the GPIUS measure support the notion that some students are experiencing a

problematic relationship with social networking sites and encourages further exploration of this

possibility. Students' GPIUS scores indicated that some individuals are using social networking

sites to change their affective states and to gain greater social benefits and social control than

they receive in face-to-face relationships. In addition, some students have a difficult time

stopping their social networking site use and keeping themselves from using these sites, while

also utilizing the sites for greater amounts of time than planned. Finally, students' scores also

indicated that for some, these sites interfere with personal, social, and professional (academic)

areas of their lives. Other support for the existence of problematic social networking site use was

also found through two direct questions about problematic use; some study participants

acknowledged that they felt or had been told that their use of social networking sites was

problematic.









Given that the measure used to assess problematic use (i.e., the GPIUS) did not have a

definitive cut-off point which would differentiate problematic users from non-users, as well as

the restricted range of GPIUS scores and a low prevalence rate found in our study, it is difficult

to say with certainty that problematic social networking site use is an identified phenomenon. A

comparison of the range of GPIUS scores for this sample with findings from other studies using

this measure (Caplan, 2002; 2003; 2007) are precluded by the fact that the other studies have

utilized individual subscales of the GPIUS rather than a sum score, as in our study. However,

like our study, Caplan showed in his study a restricted range in PIU scores and a low prevalence

of general PIU in the undergraduates he sampled. To give some indicator of prevalence, he

showed that the percentage of students endorsing symptoms of PIU ranged from 5-15% of the

sample, depending on the individual subscale (Caplan, 2006). In our study, 5-10% of the sample

could be considered problematic social networking site users, taking into account both GPIUS

scores and responses to the two direct questions about problematic use. This prevalence rate is

comparable to other prevalence rates found for general PIU in the literature. Utilizing the

Pathological Internet Use Scale (PIUS; Morahan-Martin & Schumacher, 2000), Morahan-Martin

and Schumacher classified 8.1% of their undergraduate sample as pathological Internet users,

while Niemz and colleagues (2005), utilizing this same measure, found 18.3% of the

undergraduates surveyed to be pathological Internet users. Scherer (1997) found 13% of

undergraduates sampled in his study to be Internet dependents, as classified using criteria

paralleling chemical dependencies. Finally, Greenfield (1999) utilized an online survey of

Internet use and behavior to explore a general population. He found that of the 17,251

individuals (aged 8-85 yrs) surveyed, approximately 6% met the criteria for Internet addiction.









Though some of these prevalence rates are higher than the rate found in our study, the

intended purpose of the assessment instruments utilized in these studies should be considered

when comparing these prevalence findings. Beard (2005) highlights that assessment instruments

of general problematic Internet use or Internet addiction do not take into account the different

types of Internet applications that a person could be dependent upon. Therefore, caution should

be used when comparing findings from our study assessing a specific type of Internet

dependency to prevalence rates indicated for general Internet dependency. For example, the

instruments used in these studies of general PIU could be tapping into a dependency upon online

gaming or online gambling, which may be more prevalent than say a dependency upon social

networking sites. In addition, many of the assessment instruments assume that excessive Internet

use denotes problematic Internet use, and do not take into account that elevated Internet usage

may be due to academic or work requirements (Widyanto & Griffiths, 2005). Other studies have

cited the cut-off point of many of the instruments used to assess Internet dependency as being too

liberal, resulting in inflated prevalence estimates (Niemz et al., 2005).

What can be definitively concluded from our study is that overall, the population of college

students sampled might best be characterized as a non-pathological sample with largely

normative levels of social networking site use. Nevertheless, social networking site use could be

characterized as problematic for perhaps 5-10% of the sample. This prevalence rate is on par

with other rates of mental health disorders found in the DSM-IV (APA, 2000), as pointed out by

Watson (2005). Additionally, Hall and Parsons (2001), noting the discrepancies in the Internet

addiction prevalence rates found in the literature, concluded that even a conservative estimate of

6% is worrisome and warrants further research. With social networking sites like MySpace and

Facebook becoming some of the most-trafficked websites in the world, even the possibility that









the use of these types of sites may become problematic for some individuals is cause for concern.

The findings of our study, then, encourage further investigation of problematic social networking

site use.

Evidence found by our study in support of the existence of problematic social networking

site use is in line with research that has implicated the communication functions of the Internet as

commonly used applications by general "Internet addicts" (Chou et al., 1999; Chou & Hsiao,

2000; Young, 1996a). Our study's findings could also be considered supportive of the literature

that suggests the classification of "Internet addiction" in the next version of the DSM (APA,

1995). Proponents of the addition of Internet addiction suggest that individuals can become

dependent on the environment of the Internet, and not simply to an application on the Internet

which provides another way of engaging in an addiction that already exists offline (Yellowlees &

Marks, 2007). Though our study found support for a dependency on a specific Internet

application (i.e., social networking sites), the service this application provides is not available

offline, like gambling or shopping may be. In other words, social networking sites provide a

social environment unique to the Internet. It is a social atmosphere in which the normal auditory

and visual cues that accompany offline communication are missing (Morahan-Martin &

Schumacher, 2000), creating a less-threatening environment for social interaction. In this case,

the Internet is not providing the users with a way of engaging in an addiction they would already

have offline, such as a gambling addict using an online gaming casino instead of an offline

casino. Rather, social networking sites provide an environment in which a person can socially

interact in a less-threatening atmosphere that cannot be gained offline (Morahan-Martin &

Schumacher, 2000; Peter et al., 2005). In the debate over classification of Internet addiction, our

study seems to fall on the side which supports that dependency on the unique environment of the









Internet itself can occur, as opposed to the Internet providing another forum for engaging in a

pre-existing addiction.

Prediction 2: Predictors of Problematic Social Networking Site Use

The second aim of our study was to examine how the concepts of social anxiety and

belonging to an offline social network may contribute to the development of problematic social

networking site use. Specifically, belonging to an offline social network was explored as a

possible mediator of the relationship between social anxiety and problematic social networking

site use. Within this mediation model, support was required for a number of sub-predictions in

order to test the model. Most importantly, evidence for a significant positive relationship

between the predictor variable, social anxiety, and the outcome variable, problematic social

networking site use, was found. Higher levels of social anxiety were associated with greater

levels of problematic social networking site use in this sample of college students. Individuals

who suffer anxiety in offline social environments may be more likely to develop a problematic

relationship with the use of social networking sites than those who do not suffer from social

anxiety. This finding is in line with the etiology of PIU development suggested by Caplan

(2007), who proposed that socially anxious people prefer situations which minimize their social

risk and the Internet can provide this less-threatening social atmosphere. The maladaptive

cognitions associated with social anxiety result in a preference for online communication, and

the social interaction gained by the individual in a less-threatening way reinforces the

individual's dependence on this online environment for social interaction. PIU can then develop,

or in this case, problematic social networking site use.

Belonging was chosen as a possible mediator of this relationship between social anxiety

and problematic use because inherent in this relationship is the assumption that social

networking sites have become reinforcing for the socially anxious individual due to a need that is









being fulfilled. Socially anxious individuals have been shown to have fewer friendships, and

therefore a need to belong exists (Gambrill, 1996; La Greca & Lopez, 1998). Social networking

sites were thought to be able to fulfill this need by providing a less-threatening social

environment, and that PIU could develop due to the need fulfillment provided by these sites

(Chou et al., 1999; Suler, 1999). However, our study failed to find support for belonging as a

mediator, as it was not found to be significantly related to problematic social networking site use.

Failure to find evidence for this relationship may be explained by the fact that most of the

study participants indicated they use social networking sites more for staying in contact with

existing friends in their social network rather than meeting strangers and making new friends.

The prediction of belonging as a mediator rested upon an assumption that socially anxious

individuals were using these sites to make new friends due to a lack of belonging to a current

offline network. However, it seems that these sites are structured in such a way as to entice those

with pre-existing social networks to use the sites for maintaining friendships or meeting "friends

of friends" (Boyd & Ellison, 2007), rather than as an avenue for meeting strangers and making

new friends. Therefore, it may not be a lack of current friendships that is the driving need to use

social networking sites. However, another variable of interpersonal relatedness may be the

answer.

Despite the failure to find support for belonging as a mediator, another indicator of

interpersonal relatedness loneliness seems to have potential. Though belonging and

loneliness are both indicators of interpersonal relatedness, they differ in their meanings (Hagerty

et al., 1992). In our study, belonging, as measured by the ISEL, was characterized as the

perceived availability of friends which one can spend time with. This type of belonging is

observable, as it can be behaviorally detected via group membership. Loneliness, on the other









hand, is a negative emotional or affective state in which a need for personal connectedness

exists, often due to a disruption that has occurred which resulted in an absence or loss (Hagerty

et al., 1992). A lack of belonging, therefore, is a physical lack of relationships whereas loneliness

is the feeling of non-intimacy or connectedness. Research has supported the idea that loneliness

can occur when one is dissatisfied with existing relationships (Cutrona, Russell, & Peplau, 1979)

rather than simply from a lack of belonging. Findings indicate that loneliness is more related to

the quality of one's social connections as opposed to the quantity of connections (Jackson,

Soderlind, & Weiss, 2008). Loneliness is decreased when a person feels intimacy and closeness

in their relationships (Jones, Carpenter, & Quintana, 1985) and feels supported by others (Pierce,

Sarason, & Sarason, 1991). Hagerty and colleagues illustrate with a case vignette how one can

have personal relationships (belong) but still feel lonely.

Marie moved to a town where she knew no one. The first several weeks she spent much of
her time unpacking and finding her way around. She did spend time with her new
neighbors and coworkers, all of whom seemed to really enjoy her company. Yet Marie felt
very alone, unhappy, and isolated without her friends and family.

This case mirrors the transition that many college students must make from their hometown high

school where they may have had a number of friends and family to a university where they may

know no one. Loneliness is quite prevalent among college students despite the great number of

social opportunities for belonging that exist in a college environment (Jones, Freemon, &

Goswick, 1981).

In our study, loneliness was found to be significantly related to social anxiety and

problematic social networking site use. The link between social anxiety and loneliness has been

supported in the literature (Inderbitzen-Pisaruk, Clark, & Solano, 1992) and research suggests

that socially anxious individuals perceive less intimacy and supportiveness in their existing

friendships (La Greca & Lopez, 1998). In addition, a link between loneliness and general









problematic Internet use has been found (Caplan, 2003; Morahan-Martin & Schumacher, 2000;

Nalwa & Anand, 2003). Though a prediction regarding a mediation model positing loneliness as

the mediator between social anxiety and problematic social networking site use was not made at

the start of our study, the failure of our study to find support for belonging as a mediator

prompted the exploration of this new mediation model. Our study found support for loneliness as

a mediator. It may be that the need driving socially anxious individuals to utilize social

networking sites is not a lack of friendships (belonging) but a lack of intimacy in friendships

(loneliness). Instead of using these sites to meet strangers and make friends, the sites are utilized

for enhancing relationships with existing friends. Socially anxious individuals who may feel

loneliness and desire more closeness in their relationships may turn to social networking sites to

fulfill this desire and relieve their feelings of loneliness by gaining more intimacy in their current

offline relationships. This need fulfillment and reinforcement may then develop into problematic

social networking site use, as suggested by Caplan (2007) and Suler (1999).

Studies have shown that the Internet does indeed provide an opportunity to increase

closeness in offline relationships. Valkenburg and Peter (2007) examined how online

communication with existing friends affects the offline relationship with these same friends.

Their stimulation hypothesis proposed that specific characteristics of the Internet encourage self-

disclosure more readily in online relationships than in face-to-face relationships. Self-disclosure

is important for building intimate relationships (Collins & Miller, 1994), and therefore

communicating with friends online could increase the amount of self-disclosure, and hence

intimacy, that takes place in the relationship. A number of studies have found support for

increased self-disclosure in online communication, not only between strangers, but also between

existing friends (Grinter & Palen, 2002; Leung, 2002). Valkenburg and Peter provided evidence









for the stimulation hypothesis by collecting data from 794 adolescents. They found that for

adolescents who use the Internet to communicate with existing friends, more online

communication resulted in more perceived intimacy in friendships even when offline.

In conclusion, our study supports a model in which loneliness, or the need for closeness in

relationships, mediates the relationship between social anxiety and problematic social

networking site use. Support for a model positing belonging, or the availability of friends, as a

mediator of this relationship was not found for this sample of college students. The findings

indicate that socially anxious individuals may use social networking sites to fulfill a need for

closeness in existing relationships and that they may become dependent upon this online

environment for this need fulfillment.

Prediction 3: Problematic Social Networking Site Use and Well-Being

The final aim of our study was to explore the relationship between problematic social

networking site use and well-being in college students. It was predicted that problematic use

would be related to decreased well-being. Literature showing a relationship between general PIU

and well-being was utilized as support for this prediction (Caplan, 2002; Morahan-Martin &

Schumacher, 2003; Young & Rogers, 1998). To represent well-being, measures of depression,

loneliness, self-esteem, happiness, and satisfaction with life were used. All five of the well-being

indicators showed a significant relationship with problematic social networking site use in the

expected directions, and therefore prediction 3 was fully supported. Results indicated that

individuals showing greater symptoms of problematic use also had more symptoms of

depression, loneliness, and unhappiness, as well as having lower self-esteem and less satisfaction

with their lives.









Given the lack of an experimental design of our study, it is impossible to determine

causality in the relationship between problematic social networking site use and well-being.

Support can be found in the literature for both of these causal paths.

First, it could be that this relationship exists because those suffering from decreased well-

being are more likely to develop PIU. As Davis's (2001) model of PIU development explains, a

pre-existing psychopathology or condition could make a person more vulnerable to PIU

development should a stressor occur. Our study has shown support for a model which suggests

social anxiety and the associated loneliness as contributors to the development of problematic

social networking site use. Loneliness has been shown in the literature to be associated with

decreased well-being (Weiss, 1974). A link has been well-established between loneliness or a

lack of closeness in relationships and the development of depression (Russell, Cutrona, Rose,

and Yurko, 1984). Loneliness has also been found to be related to low self-esteem (Leary, 1990)

and unhappiness (Argyle, 1987; Freedman, 1978). Other evidence of the connection between

closeness in relationships and well-being can be seen in the counseling literature. Baumeister and

Leary (1995) emphasize the contribution of the therapeutic relationship in psychotherapeutic

progress (Rogers, 1959) as support for the association of relationship closeness with well-being.

These findings taken together suggest that the decreased well-being associated with loneliness

could be an additional factor in the development of problematic social networking site use, and

are indicative of a path in which decreased well-being leads to problematic use.

However, this relationship could also exist because problematic use causes a negative

impact on an individual's well-being. Support for this causal path can be found in Davis's (2001)

model of general PIU development. He suggests that maladaptive cognitions lead to PIU

development and help maintain problematic use. These maladaptive cognitions include









cognitions about the self which result in self-doubt, low self-efficacy, and negative self-

appraisals (e.g., "I am worthless offline, but online I am somebody," "I am a failure when I am

offline"). In addition, thoughts about the world involve the generalization of specific events to

global patterns or engagement in all-or-nothing type thinking (e.g., "Nobody loves me offline,"

"The Internet is the only place I am respected"). These maladaptive cognitions associated with

PIU may lead to a decrease in self-esteem, happiness, and satisfaction with life. Additionally,

research has indicated a strong link between maladaptive cognitions and the development of

depression (Beck, 1993). Finally, general PIU has been shown to interfere with normal daily

functioning in many areas of an individual's life (Scherer, 1997; Young, 1998). Therefore, the

interference in daily functioning that problematic social networking site use may cause could

also lead to decreased well-being. Thus, these findings lend support to a causal path in which

problematic social networking site use causes a decrease in well-being.

Given that research findings provide evidence for both causal pathways between

problematic social networking site use and well-being, it is most likely that both of these paths

are correct. Decreased well-being may make one susceptible to developing a problematic

relationship with social networking sites, but the maladaptive cognitions that maintain

problematic use may further decrease well-being (Caplan, 2003; Davis, 2001). Though our study

cannot draw conclusions either way about the causality in the relationship between well-being

and problematic social networking site use, the establishment of a relationship between these two

constructs, and thus support for prediction 3, is an excellent beginning in the research of this

novel area.

Study Implications

First, our study is important in its novel contribution to an area with very limited research

but an area with much importance. Given the extensive use of social networking sites, especially









on college campuses, the usage of these sites and the impact they may have should be of interest

to researchers and to university counselors. Counselors should be aware of the potential

problematic relationship or dependency that college students could develop with social

networking sites, and how this may interfere with personal, social, and academic functioning, as

well as impact students' mental health. In a survey of university counselors, Kiralla (2005) found

that approximately 84% of counselors believed that problematic Internet use is a legitimate

concern for students, but 93% said they did not have sufficient training regarding the diagnosis

or treatment of PIU. In addition to training counselors on general PIU, counselors should also be

made aware of the potentially addictive nature of social networking sites. Information also needs

to be given to students as to how the use of the Internet and social networking sites can become

problematic and interfere with their lives. Scherer (1997) notes that many college students are

not even aware that dependency or problematic usage can develop with the Internet, and that it is

the responsibility of counselors to make students aware of the symptoms and available services.

Many university counseling centers are now online, and given the centrality that the Internet

plays in problematic users' lives, this may be the perfect place to begin the dissemination of

important information regarding the symptoms and effects of general PIU and problematic social

networking site use, as well as the available services for those who may struggle with these

issues. Many of the PIU assessment instruments are already available as simple checklists (e.g.,

Young's Internet Addiction DQ) that could be put online for students to self-assess their PIU

potential.

Counselors can also benefit from our study by learning more about the attraction that

social networking sites may hold for socially anxious individuals and how these sites can both be

beneficial and detrimental for them. Though our study explores the detrimental side of site use,









social networking sites may also provide an excellent tool for socially anxious individuals.

Amichai-Hamburger and Furnham (2007) note that socially inhibited people may benefit from

the non-threatening environment the Internet allows for acquiring, practicing, and improving

social skills if they can transfer these skills to offline social interactions. For some, the transfer of

social skills acquired on the Internet to face-to-face relationships may be natural (McKenna,

Greene, & Gleason, 2002), while for others with more severe social anxiety, this ability may be

more limited. Amichai-Hamburger and Furnham propose that even very socially anxious

individuals can learn to transfer online social skills to offline relationships through a series of

steps that gradually build the individual's exposure to face-to-face interactions via audio and

video functions of the Internet. Therefore, social networking sites could be a valuable tool for

counselors to use with socially anxious individuals if proper steps are taken to help socially

anxious people avoid dependency on these sites for all social interaction and support.

Limitations

Our study has a number of limitations that should be noted. First, all of the measures used

in our study are of a self-report nature. This allows for under- or over-reporting by students that

may affect the reliability and validity of the study findings. Many of the constructs assessed in

our study could be associated with shame, denial, or minimization, and this could also lead to

under-reporting (Block, 2008). Additionally, due to the restricted range of GPIUS scores and a

low prevalence rate, it cannot be concluded definitively that the problematic use of social

networking sites exists, though our study certainly provides a starting point for further research

in this area. Another limitation is that due to the lack of an experimental design, causal

inferences cannot be made. Generalizability of our study findings is also limited to

undergraduate students at a four-year university, and the predominantly female composition of

the sample also limits the generalizability. Finally, our study does not include the numerous other









variables that may contribute to the relationships examined and therefore only a limited picture

of problematic social networking site use is given. These limitations are reasonable given that

little research about social networking sites has been conducted. The purpose of our study was to

simply provide a foundation for the examination of social networking site use and the

problematic relationship that can develop with the use of these sites. Future research exploring

this topic in much more detail will certainly be needed.

Future Research

These limitations suggest a number of directions for future research on problematic social

networking site use. First, a more accurate estimate of the prevalence of problematic use in the

college population is most certainly needed. This suggests that development and validation of

assessment instruments measuring problematic use should continue, and a measure with a

designated cut-off point identifying problematic users versus non-problematic users may be

warranted. Examining the prevalence of problematic social networking site use in populations

other than college students, such as younger adolescents or adults, may also yield different and

interesting findings. Future studies could also benefit from an experimental design in which

conclusions regarding causality in the relationships explored in our study could be made. Finally,

the predictors of problematic use included in our study only accounted for a small portion of the

variance in scores indicating problematic social networking site use, and therefore other

predictors of problematic use should be explored. Establishing variables which are predictive of

problematic use and differentiating these from variables that are the result of problematic use

could help guide practitioners in diagnosing and treating this issue.

These are just a small sampling of future directions for research in the area of problematic

social networking site use. Given the non-existence of research in this area, much research is still

needed. The numerous implications of this type of research coupled with a trend showing that









social networking sites are continuing to grow in popularity (Owyang, 2008) suggest the

importance of continued growth in our knowledge about social networking site use and the

impact these sites may have for users.









APPENDIX A
INTERPERSONAL SUPPORT EVALUATION LIST (ISEL) BELONGING SCALE

This scale is made up of a list of statements each of which may or may not be true about
you. Please check true if the statement applies to you most of the time or check false if the
statement does not usually apply to you.

1. There are people at school or in town who I regularly run with, exercise with, play sports
with, or do other enjoyable activities.

2. I hang out in a friend's room or apartment quite a lot.

3. I can find a person who I enjoy spending time with whenever I want.

4. If I decided at dinner time to take a study break this evening and go to a movie, I could
easily find someone to go with me.

5. People hang out in my room or apartment during the day or in the evening.

6. I belong to a group at school or in town that meets regularly or does things together
regularly.

7. I am not a member of any social groups (such as church groups, clubs, teams, etc.)

8. Lately, I often feel lonely, like I don't have anyone to reach out to.

9. I don't have friends at school or in town who would comfort me by showing some
physical affection.

10. I don't often get invited to do things with other people.

11. I don't usually spend two evenings on the weekend doing something with others.









APPENDIX B
MODIFIED GENERALIZED PROBLEMATIC INTERNET USE SCALE (GPIUS)

Rate the extent to which you agree or disagree with each statement about your use of
online social networking sites. Social networking sites are websites on the Internet where you
can create a profile and connect with friends. Examples of these sites are MySpace, Facebook, or
Friendster.

1. I have used social networking sites to talk with others when I was feeling isolated.

2. I can control how others perceive me when I am using a social networking site.

3. I find it hard to stop thinking about what is waiting for me online on the social
networking site(s) I like to use.

4. I have gone on a social networking site to make myself feel better when I was down or
anxious.

5. I have tried to stop using one or more social networking sites for such long periods of
time.

6. When not on a social networking site, I wonder what is happening on that site.

7. When I am on a social networking site, I socialize with people without worrying about
relationship commitment.

8. I have attempted to spend less time on social networking sites but have not been able to.

9. I am treated better by others while online using social networking sites than I am offline.

10. I use social networking sites to make myself feel better when I'm down.

11. I have missed class or work because of being online on a social networking site.

12. I feel worthless offline, but online on a social networking site I am someone.

13. I want to, or have made, unsuccessful efforts to cut down or control my use of one or
more social networking sites.

14. I have missed a social event or social engagements because of being online on a social
networking site.

15. I am more comfortable with computers than with people.

16. I feel guilty about the amount of time I spend on social networking sites.









17. When I am on a social networking site, I socialize with other people without worrying
about how I look.

18. I lose track of time when I am using a social networking site.

19. I miss being online on a social networking site if I can't get on it.

20. I am treated better in my online relationships through social networking sites than in my
face-to-face relationships.

21. I have used a social networking site for a longer time than I intended.

22. I am more confident socializing on social networking sites than I am offline.

23. I have used a social networking site for longer periods of time than I had expected to.

24. I feel safer relating to people online on social networking sites rather than face-to-face.

25. I feel lost if can't get on a social networking site.

26. I have spent a good deal of time using social networking sites.

27. I have gotten into trouble with my employer or school because of being online on a social
networking site.

28. I am preoccupied with thinking about social networking sites if I can't connect for some
time.

29. I have sought others online on social networking sites when I was feeling isolated.









APPENDIX C
SOCIAL AVOIDANCE AND DISTRESS SCALE (SAD)

For the following items, think about how the statement applies to you in your offline
social life, not how you are online when using the Internet. Please check true if the statement
applies to you most of the time or checkfalse if the statement does not usually apply to you.

1. I feel relaxed even in unfamiliar situations.

2. I try to avoid situations which force me to be very sociable.

3. It is easy for me to relax when I am with strangers.

4. I do not have a particular desire to avoid people.

5. I often find social occasions upsetting.

6. I usually feel calm and comfortable at social occasions.

7. I am usually at ease when talking to someone of the opposite sex.

8. I try to avoid talking to people unless I know them well.

9. If the chance comes to meet new people, I often take it.

10. I often feel nervous or tense in casual get-togethers in which both sexes are present.

11. I am usually nervous with people unless I know them well.

12. I usually feel relaxed.

13. I often want to get away from people.

14. I usually feel uncomfortable when I am in a group of people I don't know.

15. I usually feel relaxed when I meet someone for the first time.

16. Being introduced to people makes me tense and nervous.

17. Even though a room is full of strangers, I may enter it anyway.

18. I would avoid walking up and joining a large group of people.

19. When my superiors want to talk with me, I talk willingly.

20. I often feel on edge when I am with a group of people.









21. I tend to withdraw from people.

22. I don't mind talking to people at parties or social situations.

23. I am seldom at ease in a large group of people.

24. I often think up excuses in order to avoid social arrangements.

25. I sometimes take the responsibility for introducing people to each other.

26. I try to avoid formal social occasions.

27. I usually go to whatever social engagements I have.

28. I find it easy to relax with other people.









APPENDIX D
OXFORD HAPPINESS QUESTIONNAIRE (OHQ)

Please indicate how much you agree or disagree with the following statements. You will
need to read the statements carefully because some are phrased positively and others negatively.
Please give the answer that is true for you in general or most of the time.

1. I don't feel particularly pleased with the way I am.

2. I am intensely interested in other people.

3. I feel that life is very rewarding.

4. I have very warm feelings towards almost everyone.

5. I rarely wake up feeling rested.

6. I am not particularly optimistic about the future.

7. I find most things amusing.

8. I am always committed and involved.

9. Life is good.

10. I do not think that the world is a good place.

11. I laugh a lot.

12. I am well satisfied about everything in my life.

13. I don't think I look attractive.

14. There is a gap between what I would like to do and what I have done.

15. I am very happy.

16. I find beauty in some things.

17. I always have a cheerful effect on others.

18. I can fit in everything I want to.

19. I feel that I am not especially in control of my life.

20. I feel able to take anything on.









21. I feel fully mentally alert.

22. I often experience joy and elation.

23. I do not find it easy to make decisions.

24. I do not have a particular sense of meaning and purpose in my life.

25. I feel I have a great deal of energy.

26. I usually have a good influence on events.

27. I do not have fun with other people.

28. I don't feel particularly healthy.

29. I do not have particularly happy memories of the past.









LIST OF REFERENCES

American Psychological Association. (1992). Ethical principles of psychologists and code of
conduct. American Psychologist, 47, 1597-1611.

American Psychiatric Association (1995). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: American Psychiatric Publishing, Inc.

Amichai-Hamburger, Y., & Furnham, A. (2007). The positive net. Computers In Human
Behavior, 23, 1033-1045.

Anant, S. S. (1966). The need to belong. Canada's Mental Health, 14, 21-27.

Argyle, M. (1987). The psychology of happiness. London: Methuen.

Armstrong, L., Phillips, J.G., & Saling, L.L. (2000). Potential determinants of heavier internet
usage. International Journal ofHuman Computer Studies, 53, 537-550.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173-1182.

Baumeister, R., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments
as a fundamental human motivation. Psychological Bulletin, 117, 497-529.

Beard, K. W. (2005). Internet addiction: A review of current assessment techniques and potential
assessment questions. Cyberpsychology & Behavior, 8(1), 7-14.

Beard, K. W., & Wolf, E. M. (2001). Modification in the proposed diagnostic criteria for Internet
addiction. Cyberpsychology and Behavior, 4(3), 377-383.

Beck, A. T. (1993). Cognitive therapy: Past, present, and future. Journal of Counseling and
Clinical Psychology, 61(2), 194-198.

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manualfor the Beck Depression Inventory
(2nd ed.). San Antonio, TX: The Psychological Corporation.

Block, J. J. (2008). Issues for DSM-V: Internet Addiction. American Journal ofPsychiatry,
165(3), 306-307.

Bowerman, B. L., & O'Connell, R. T. (1990). Linear statistical models: An applied approach
(2nd edition). Belmont, CA: Duxbury.

Bowlby, J. (1980). Attachment and loss. New York: Basic Books.

Boyd, D. M., & Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship.
Journal of Computer-Mediated Communication, 13(1). Retrieved July 20, 2008, from
http://jcmc.indiana.edu/voll3/issuel/boyd.ellison.html









Brenner, V. (1997). Parameters of Internet use, abuse and addiction: The first 90 days of the
Internet usage survey. Psychological Reports, 80, 879-882.

Caplan, S. E. (2002). Problematic Internet use and psychosocial well-being: development of a
theory-based cognitive-behavioral measure. Computers in Human Behavior, 18, 533-575.

Caplan, S. E. (2003). Preference for online social interaction: a theory of problematic Internet
use and psychosocial well-being. Communication Research, 30, 625-648.

Caplan, S. E., & High, A. C. (2006). Beyond excessive use: The interaction between cognitive
and behavioral symptoms of problematic Internet use. Communication Research Reports,
23(4), 265-271.

Caplan, S. E. (2007). Relations among loneliness, social anxiety, and problematic Internet use.
Cyberpsychology & Behavior, 10(2), 234-242.

Chak, K., & Leung, L. (2004). Shyness and locus of control as predictors of Internet addiction
and Internet use. CyberPsychology & Behavior, 7(5), 559-570.

Chen, S. H., & Chou, C. (1999, August). Development of Chinese Internet addiction scale in
Taiwan. Poster presented at the 107th American Psychology Annual convention, Boston,
USA.

Choi, Y. H. (2007). Advancement ofIT and seriousness ofyouth Internet addiction. International
Symposium on the Counseling and Treatment of Youth Internet Addiction. Seoul, Korea.

Chou, C., Chou, J., & Tyan, N. N. (1999). An exploratory study of Internet addiction, usage and
communication pleasure-The Taiwan's case. International Journal of Educational
Telecommunications, 5(1), 47-64.

Chou, C., Condron, L., & Belland, J. C. (2005). A review of the research on Internet addiction.
Educational Psychology Review, 17(4), 363-388.

Chou, C., & Hsiao, M. C. (2000). Internet addiction, usage, gratifications, and pleasure
experience-The Taiwan college students' case. Computers & Education, 35(1), 65-80.

Cohen, J. (1977). Statistical power analysisfor the behavioral sciences (revised edition). New
York: Academic Press.

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

Collins, N. L., & Miller, L. C. (1994). Self-disclosure and liking: A meta-analytic review.
Psychological Bulletin, 116, 457- 475.

Connor, K. M., Davidson, J. R. T., Churchill, L. E., et al. (2000). Psychometric properties of the
Social Phobia Inventory (SPIN): New self-rating scale. British Journal ofPsychiatry, 176,
379-386.









Coopersmith, S. (1991). Self-Esteem Inventories. CA: Consulting Psychologists Press, Inc.

Cutrona, C., Russell, D., & Peplau, A. (1979). Loneliness and the process of social adjustment: A
longitudinal study. Paper presented at the annual meeting of the American Psychological
Association, New York, NY.

D'Amico, M. L. (1998). Internet has become a necessity, U. S. poll shows. Retrieved November
14, 2007, from www.cnn.com/TECH/computing/9812/07/neednet.idg/index.html

Davis, R. A. (2001). A cognitive-behavioral model of pathological Internet use. Computers in
Human Behavior, 17, 187-195.

Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale.
Journal ofPersonality Assessment, 49, 71-75.

Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook "friends:" Social
capital and college students' use of online social network sites. Journal of Computer-
Mediated Communication, 12(4). Retrieved November 14, 2007, from
http://jcmc.indiana.edu/voll2/issue4/ellison.html

Frazier, P. A., Mortensen, H., & Steward, J. (2005). Coping strategies as mediators of the
relations among perceived control and distress in sexual assault survivors. Journal of
Counseling Psychology, 52(3), 267-278.

Freedman, J. (1978). Happy people: What happiness is, who has it, and why. New York:
Harcourt Brace Jovanovich.

Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect.
Psychological Science, 18(3), 233-239.

Gambrill, E. (1996). Loneliness, social isolation and social anxiety. In M. A. Mattaini & B. A.
Thyer (Eds.), Finding solutions to social problems: Behavioral strategies for change (pp.
345-371). Washington, DC: American Psychological Association.

George, A. (2006). Living online: The end of privacy? New Scientist, 2569. Retrieved July 20,
2008, from http://www.newscientist.com/channel/tech/mgl9125691.700-living-online-the-
end-of-privacy. html

Goldsmith, R. E. (1986). Dimensionality of the Rosenberg Self-Esteem Scale. Journal of Social
Behavior & Personality, 1(2), 253-264.

Greenfield, D. N. (1999). Psychological characteristics of compulsive Internet use: A
preliminary analysis. Cyberpsychology & Behavior, 2(5), 403-412.

Griffiths, M.D. (1998). Internet addiction: Does it really exist? In J. Gackenbach (Ed.),
Psychology and the Internet: intrapersonal, interpersonal and transpersonal applications
(pp. 61-75). New York: Academic Press.









Griffiths, M. D. (2000). Does Internet and computer "addiction" exist? Some case study
evidence. Cyberpsychology & Behavior, 3(2), 211-218.

Griffiths, M. D. (2003). Internet abuse in the workplace: Issues and concerns for employers and
employment counselors. Journal of Employment Counseling, 40(2), 87-96.

Grinter, R. E., & Palen, L. (2002, November). Instant messaging in teen life. Paper presented at
the 2002 Association for Computing Machinery Conference on Computer Supported
Cooperative Work. Retrieved November 14, 2007, from
http://doi.acm.org/10.1145/587078.587082

Gross, E. F. (2004). Adolescent Internet use: What we expect, what teens report. Journal of
Applied Developmental Psychology, 25, 633- 649.

Hagerty, B. M., Lynch-Sauer, J., Patusky, K. L., Bouwsema, M., & Collier, P. (1992). Sense of
belonging: A vital mental health concept. Archives of Psychiatric Nursing, 6, 172-177.

Hagerty, B. M., Williams, R. A., Coyne, J. C., & Early, M. R. (1996). Sense of belonging and
indicators of social and psychological functioning. Archives of Psychiatric Nursing, 4,
235-244.

Hann, D., Winter, K., & Jacobsen, P. (1999). Measurement of depressive symptoms in cancer
patients: Evaluation of the Center for Epidemiological Studies Depression Scale (CES-D).
Journal ofPsychosomatic Research, 46(5), 437-443.

Hempel, J. (2005, December 12). The MySpace generation. BusinessWeek. Retrieved November
14, 2007, from http://www.mywire.com/pubs/BusinessWeek/2005/12/12/1102930?page=l

Hills, P., & Argyle, M. (2002). The Oxford Happiness Questionnaire: a compact scale for the
measurement of psychological well-being. Personality and Individual Differences, 33,
1073-1082.

Howe, N., & Strauss, W. (2000). Millennials rising: The next great generation. New York:
Vintage.

Inderbitzen-Pisaruk, H., Clark, M. L., & Solano, C. H. (1992). Correlates of loneliness in
midadolescence. Journal of Youth and Adolescence, 21(2), 151-167.

Jackson, T., Soderlind, A., & Weiss, K. E. (2000). Social Behavior and Personality, 28(5), 463-
470.

Jones, W. H., Carpenter, B. N., & Quintana, D. (1985). Personality and interpersonal predictors
of loneliness in two cultures. Journal ofPersonality and Social Psychology, 48, 1053-
1511.

Jones, W. H., Freemon, J. E., & Goswick, R. A. (1981). The persistence of loneliness: Self and
other determinants. Journal ofPersonality, 49(1), 27-48.









Kandell, J. J. (1998). Internet addiction on campus: the vulnerability of college students.
CyberPsychology & Behavior, 1, 11-17.

Kiralla, L. V. (2005). Internet addiction disorder: A descriptive study of college counselors in
four-year institutions. (Doctoral dissertation, University of La Verne, 2005). Retrieved
November 14, 2007, from PsychInfo database.

Ko, C., Yen, J., Chen, C., Chen, S., & Yen, C. (2005). Proposed diagnostic criteria of Internet
addiction for adolescents. Journal of Nervous and Mental Disorders, 193, 728-733.

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? American Psychologist, 53(9), 1017-1031.

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

La Greca, A. M., & Lopez, N. (1998). Social anxiety among adolescents: Linkages with peer
relations and friendships. Journal ofAbnormal Child Psychology, 26(2), 83-94.

Leary, M. R. (1990). Responses to social exclusion: Social anxiety, jealousy, loneliness,
depression, and low self-esteem. Journal of Social and Clinical Psychology, 9, 221-229.

Leung, L. (2002). Loneliness, self-disclosure and ICQ ("I Seek You") use. CyberPsychology &
Behavior, 5, 241-251.

Leung, L. (2004). Net-generation attributes and seductive properties of the Internet as predictors
of online activities and Internet addiction. Cyberpsychology & Behavior, 7(3), 333-348.

Lin, S. S. J., & Tsai, C. C. (1999). Internet Addiction among High Schoolers in Taiwan. Poster
presented at the 107th American Psychology Association (APA) Annual Convention,
Boston, USA.

Marks, I. (1990). Non-chemical behaviorala) addictions. British Journal ofAddiction, 85, 1389-
1394.

Maslow, A. H. (1968). Toward a psychology of being. New York: Van Nostrand.

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

McKenna, K. Y. A., & Bargh, J. (2000). Plan 9 from cyberspace: The implications of the Internet
for personality and social psychology. Personality and Social Psychology Review, 4, 57-
75.

Menard, S. (1995). Applied logistic regression analysis. Quantitative Applications in the Social
Sciences, 106, 7-196.









Mitchell, P. (2000). Internet addiction: Genuine diagnosis or not? The Lancet, 355, 632-633.

Moore, D. (1995). The emperor's virtual c 1vhe%\ The naked n inhl/ about the Internet culture.
Chapel Hill, NC: Alogonquin.

Morahan-Martin, J., & Schumacher, P. (2000). Incidents and correlates of pathological Internet
use among college students. Computers in Human Behavior, 16, 13-29.

Morahan-Martin, J., & Schumacher, P. (2003). Loneliness and social uses of the Internet.
Computers in Human Behavior, 19, 659-671.

Nalwa, K., & Anand, A. P. (2003). Internet addiction in students: A cause of concern.
Cyberpsychology & Behavior, 6(6), 653-656.

Niemz, K., Griffiths, M., Banyard, P. (2005). Prevalence of pathological Internet use among
university students and correlations with self-esteem, the General Health Questionnaire
(GHQ), and disinhibition. CyberPsychology & Behavior, 8(6), 562-570.

Owyang, J. (2008). Social network stats: MySpace, Facebook, and Reunion. Retrieved July 20,
2008, from http://www.web-strategist.com/blog/2008/01/09/social-network-stats-
facebook-myspace-reunion-j an-2008/

Peter, J., Valkenburg, P. M., & Schouten, A. P. (2005). Developing a model of adolescent
friendship formation on the Internet. Cyberpsychology & Behavior, 8(5), 423-429.

Pierce, G. R., Sarason, I. G., & Sarason, B. R. (1991). General and relationship-based
perceptions of social support: Are two constructs better than one? Journal ofPersonality
and Social Psychology, 61, 1028-1039.

Radloff, L. S. (1977). The Center for Epidemiologic Studies-Depression Scale: A self-report
depression scale for research in the general population. Applied Psychological
Measurement, 1, 385-401.

Rice, K. G., Ashby, J. S., & Slaney, R. B. (1998). Self-esteem as a mediator between
perfectionism and depression: A structural equations analysis. Journal of Counseling
Psychology, 45(3), 304-314.

Rogers, Carl. (1959). A theory of therapy, personality and interpersonal relationships as
developed in the client-centered framework. In S. Koch (Ed.), Psychology: A study of a
science. Vol. 3: Formulations of the person and the social context. New York: McGraw
Hill.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
University Press.

Russell, D. (1996). UCLA Loneliness Scale (Version 3): Reliability, validity, and factor
structure. Journal ofPersonality Assessment, 66, 20-40.









Russell, D., Cutrona, C. E., Rose, J., & Yurko, K. (1984). Sound and emotional loneliness: An
examination of Weiss's typology of loneliness. Journal ofPersonality and Social
Psychology, 46, 1313-1321.

Russell, D., Peplau, L. A., & Cutrona, C. E. (1980). The revised UCLA loneliness scale:
Concurrent and discriminant validity evidence. Journal ofPersonality and Social
Psychology, 39, 472-480.

Scherer, K. (1997). College life on-line: healthy and unhealthy Internet use. Journal of College
Student Development, 38, 655-665.

Schlenker, B. R., & Leary, M. R. (1982). Social anxiety and self-presentation: A
conceptualization and model. Psychological Bulletin, 92, 641-669.

Shafer, A. B. (2006). Meta-analysis of the factor structures of four depression questionnaires:
Beck, CES-D, Hamilton, and Zung. Journal of Clinical Psychology, 62(1), 123-146.

Shapira, N. A., Goldsmith, T. D., Keck, P. E., Jr., Khosla, U. M., & McElroy, S. L. (2000).
Psychiatric features of individuals with problematic internet use. Journal ofA elective
Disorders, 57(1-3), 267-272.

Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation
models. In S. Leinhardt (Ed.), Sociological methodology (pp. 290-312). Washington, D.C.:
American Sociological Association.

Stephenson, W. (1988). The play theory of mass communication. New Brunswick, NJ:
Transaction Books.

Suler, J. R. (1999). To get what you need: Healthy and pathological Internet use.
Cyberpsychology & Behavior, 2(5), 355-393.

Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd edition). New York:
Harper & Row.

Tapscott, D. (1998). Growing up digital: The rise of the Net Generation. New York: McGraw-
Hill.

Treuer, T., Fabian, Z., & Furedi, J. (2001). Internet addiction associated with features of impulse
control disorder: Is it a real psychiatric disorder? Journal of Affective Disorders, 66, 283.

Valkenburg, P. M., & Peter, J. (2007). Preadolescents' and adolescents' online communication
and their closeness to friends. Developmental Psychology, 43(2), 267-277.

Valkenburg, P. M., Peter, J., & Alexander, P. S. (2006). Friend networking sites and their
relationship to adolescents' well-being and social self-esteem. Cyberpsychology &
Behavior, 9(5), 584-590.









Vassar, M. (2008). A note on the score reliability for the Satisfaction With Life Scale: an RG
study. Social Indicators Research, 86(1), 47-57.

Wan, C. S., & Chiou, W. B. (2006). Why are adolescents addicted to online gaming? An
interview study in Taiwan. Cyberpsychology & Behavior, 9(6), 762-766.

Watson, J. C. (2005). Internet addiction diagnosis and assessment: Implications for counselors.
Journal of Professional Counseling: Practice, Theory & Research, 33 (2), 17-30.

Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of
Consulting & Clinical Psychology, 33, 448-457.

Weiss, R. S. (1974). Loneliness: The experience of emotional and social isolation. Cambridge,
MA: MIT Press.

Widyanto, L., & Griffiths, M. (2006). 'Internet addiction': A critical review. International
Journal of Mental Health and Addiction, 4, 31-51.

Yellowlees, P. M., & Marks, S. (2007). Problematic Internet use or Internet addiction?
Computers in Human Behavior, 23, 1447-1453.

Yen, J., Ko, C., Yen, C., Wu, H., & Yang, M. (2007). The comorbid psychiatric symptoms of
Internet addiction: Attention deficit and hyperactivity disorder (ADHD), depression, social
phobia, and hostility. Journal ofAdolescent Health, 41, 93-98.

Young, K. (1996a). Addictive use of the internet: A case that breaks the stereotype.
Psychological Reports, 79, 899-902.

Young, K. S. (1998). Internet addiction: the emergence of a new clinical disorder.
Cyberpsychology & Behavior, 1(3), 237-244.

Young, K. S. (1999). The research and controversy surrounding Internet addiction.
Cyberpsychology & Behavior, 2(5), 381-383.

Young, K. S., & Rogers, R. C. (1998). The relationship between depression and Internet
addiction. Cyberpsychology & Behavior, 1(1), 25-28.

Yu, S., & Yu, M. (2007). Comparison of Internet-based and paper-based questionnaires in
Taiwan using multisample invariance approach. Cyberpsychology & Behavior, 10, 501-
507.









BIOGRAPHICAL SKETCH

Andrea Spraggins attended Vanderbilt University in Nashville, TN, for her undergraduate

degree, where she graduated magna cum laude with a double major in psychology and child

development. She recently completed her coursework for her doctorate degree in counseling

psychology at the University of Florida, and is currently on internship at the University of

Houston Counseling & Psychological Center.





PAGE 1

1 PROBLEMATIC USE OF ONLINE SOCI AL NETWORKING SITES FOR COLLEGE STUDENTS: PREVALENCE, PREDICTORS, AND ASSOCIATION WITH WELL-BEING By ANDREA SPRAGGINS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Andrea Spraggins

PAGE 3

3 ACKNOWLEDGMENTS I would like to express my a ppreciation to a number of pe ople for supporting m e in my dissertation. First, I would like to thank my advisor, Dr. Greg Ne imeyer, for all of his guidance and support of this project. Thanks also go to the members of my co mmittee, Dr. Sondra Smith, Dr. Jim Morgan, and Dr. Jeff Farrar, who gave their valuable time to this project. My parents and family also have my deepest appreciation for all of their support and love which helped get me to this place and continues to keep me grounded. Most important, I thank my husband, Barrett Spraggins, who has been my rock throughout gradua te school. Barrett conti nues to be my biggest supporter and I appreciate him dearly.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 3LIST OF TABLES ...........................................................................................................................6LIST OF FIGURES .........................................................................................................................7ABSTRACT ...................................................................................................................... ...............8 CHAP TER 1 INTRODUCTION .................................................................................................................. 102 LITERATURE REVIEW .......................................................................................................13History of Social Networking Sites ........................................................................................14MySpace ....................................................................................................................... ...15Facebook ...................................................................................................................... ....15Problematic Internet Use or Internet Addiction ...................................................................... 16First aim: Existence of Problema tic Social Networking Site Use .......................................... 22Second Aim: Predictors of Problema tic Social Networking Site Use .................................... 24Etiology of Problematic Internet Use .............................................................................. 25Predictors of Problematic So cial Networking Site Use ...................................................27Third Aim: Problematic Social Ne tworking Site Use and Well-Being ..................................30Study Overview ......................................................................................................................343 MATERIALS AND METHODS ...........................................................................................36Participants .................................................................................................................. ...........36Measures ...................................................................................................................... ...........36Belonging to an Offline Social Network ......................................................................... 37Problematic Social Ne tworking Site Use ........................................................................ 38Social Anxiety .................................................................................................................40Well-Being .................................................................................................................... ..40Depression ................................................................................................................ 40Loneliness ................................................................................................................. 41Self-Esteem .............................................................................................................. 41Happiness ................................................................................................................. 42Satisfaction with life ................................................................................................. 42Methods ..................................................................................................................................434 RESULTS ....................................................................................................................... ........44Social Networking Site Use .................................................................................................... 44Examination of the Modified GPIUS ..................................................................................... 44

PAGE 5

5 Prediction 1: Prevalence of Problem atic Social Networking Site Use ................................... 46Prediction 2: Predictors of Problem atic Social Networking Site Use .................................... 48Prediction 3: Problematic Social Networking Site Use & Well-Being .................................. 555 DISCUSSION .................................................................................................................... .....60Review of Study Findings ......................................................................................................60Social Networking Site Use ............................................................................................. 60Measure of Problematic Soci al Networking Site Use ..................................................... 61Prediction 1: Existence of Problematic Social Networking Site Use .............................. 62Prediction 2: Predictors of Problem atic Social Networking Site Use ............................. 66Prediction 3: Problematic Social Ne tworking Site Use and Well-Being ........................ 70Study Implications ............................................................................................................ ......72Limitations ................................................................................................................... ...........74Future Research ......................................................................................................................75APPENDIX A INTERPERSONAL SUPPORT EVALUATION LIST (ISEL) BEL ONGING SCALE ....77B MODIFIED GENERALIZED PROBLEMATIC INTERNET USE SCALE (GPIUS) ......... 78C SOCIAL AVOIDA NCE AND DISTRESS SCALE (SAD) ................................................... 80D OXFORD HAPPINESS QUESTIONNAIRE (OHQ) ............................................................ 82LIST OF REFERENCES ...............................................................................................................84BIOGRAPHICAL SKETCH .........................................................................................................92

PAGE 6

6 LIST OF TABLES Table page 4-1 Distribution of GPIUS Scores ............................................................................................ 474-2 Correlations Among Problematic Social Networking Site Use (GPIUS scores), Social Anxiety (SAD scores), and Belonging (ISEL scores) ............................................ 504-3 Means and Standard Deviations of WellBeing Variables for So cial Networking Site Users ......................................................................................................................... .........564-4 Correlations Between Well -Being Variables and Problematic Social Networking Site Use (GPIUS Scores) ..........................................................................................................584-5 Means and Standard Deviations of We ll-Being Variables for Problematic Social Networking Site Users (GPI US > 87) and Non-Problema tic Users (GPIUS < 88) ........... 59

PAGE 7

7 LIST OF FIGURES Figure page 4-1 Mediation model ........................................................................................................... .....52

PAGE 8

8 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PROBLEMATIC USE OF ONLINE SOCI AL NETWORKING SITES FOR COLLEGE STUDENTS: PREVALENCE, PREDICTORS, AND ASSOCIATION WITH WELL-BEING By Andrea Spraggins August 2009 Chair: Gregory Neimeyer Major: Counseling Psychology Social networking sites like MySpace a nd Facebook have become such a popular communication application on the In ternet that these sites are now some of the most-trafficked websites in the world. Though use of social ne tworking sites has become widespread, little research has been conducted about these sites an d their impact on users. Our study attempted to partially address this gap in the literature by ex amining social networking site use among college students, specifically when th e utilization of these types of sites becomes problematic or addictive. Researchers have begun finding evidence for the existence of Internet addiction or problematic use of the Internet in general. Howe ver, specific Internet applications which may hold addictive potential have b een rarely researched, with th e exception of online gambling and gaming. Given the popularity of social networking s ites, the aims of our study were to determine if problematic use of social networking sites can develop, and to investigate some possible predictors of problematic use, including social anxiety and a lack of be longing to an offline social network. The less-threate ning social environment that social networking sites offer compared to face-to-face interact ions may make these sites part icularly seductive for socially anxious individuals lacking friends The relationship of problematic social networking site use with well-being was also examined in this study. An online survey was completed by 367

PAGE 9

9 undergraduates, a majority of which identified them selves as social networking site users. Our study found evidence for the existence of problema tic social networking si te use by utilizing a measure modified from an existi ng measure of general problematic Internet use. Social anxiety was found to be positively related with problematic use, but our study failed to find support for lack of belonging to an offline social network as another predictor. However, a model placing loneliness as a mediator of the relationship be tween social anxiety and problematic use was supported. Finally, our study found evidence for a link between problematic use and well-being. Increased symptoms of problematic use were asso ciated with decreased self-esteem, happiness, satisfaction with life, and increased depression and loneliness.

PAGE 10

10 CHAPTER 1 INTRODUCTION The need for friendships is a natural part of hum an existence (Baumeister and Leary, 1995) and the Internet is a pervasive part of our culture, so it is no surpri se that the Internet has become a source for making friends and maintaining relati onships. Social networking sites are the most recent technological creation on the Internet th at serve this purpose. MySpace and Facebook, two of the most popular online social networking si tes, have become an international phenomenon and an important part of how adolescents a nd young adults communicat e with one another. Though use of social networking sites has become widespread, little research has been conducted on the usage and impact of these sites. Our study attempted to partially ad dress this gap in the literature by examining social networking site us e among college students, specifically when the utilization of these types of sites becomes problematic or addictive. Like social networking sites, pr oblematic Internet use (PIU), or Internet addiction as it is called by some researchers, is a relatively new phenomenon that has come to the attention of researchers in recent years. Literature shows that some individuals can become dependent upon the Internet and that this can in terfere with professiona l, social, and personal functioning, as well as negatively impact well-being, much like othe r types of addiction (Caplan, 2002, 2003; Davis, 2001; Young, 1996a). Most of the PIU literature has focused on Internet use in general without examining specific applications, though exceptions include research on online gambling (Griffiths, 2003) and online gaming addictions (Wan & Chiou, 2006). No research has examined the addictive potential that so cial networking sites may hold. So cial networking sites may be particularly seductive due to th e communicative and interactive f eatures they offer in a lessthreatening social environment. Studies of probl ematic general Internet use have shown that individuals with Internet depe ndency utilize the communication or interactive functions of the

PAGE 11

11 Internet significantly more than non-co mmunicative functions (Morahan-Martin and Schumacher, 2000; Young, 1996a). Therefore, the first aim of our study was to assess the prevalence of the problematic usage of social ne tworking sites in colleg e students, a uniquely vulnerable population to PIU (Kandell, 1998). A second aim of our study was to investigate some possible predictors of problematic social networking site use. Studies of problematic use of the Internet in general have shown that socially anxious individuals may be at greater risk for developi ng PIU than non-socially anxious individuals due to the unique a nd less-threatening environment th at the Internet provides for social interaction (Caplan, 2007; Yen, Ko, Yen, Wu, and Yang, 2007). Given that social networking sites are a key source of social interaction on the Internet, it was proposed in our study that socially anxious individuals may be pa rticularly prone to problematic or dependent usage of these types of sites. In addition to soci al anxiety, another predictor was also considered that may mediate the relationship between social anxiety and problematic social networking site use. It is thought that indivi duals with higher levels of soci al anxiety may lack a sense of belonging to an offline social ne twork due to the anxiety associat ed with forming face-to-face friendships. Socially anxious indi viduals may then turn to social networking sites to compensate for the lack of social support in their lives, and become depe ndent on the ability of social networking sites to fulfill this need. Therefore, our study examined a model in which social anxiety leads to problematic so cial networking site use partia lly through a decreased sense of belonging to an offline social network. Finally, the third aim of our study was to examine how problematic usage of social networking sites relates to levels of well-bei ng in college students. Research has shown a significant relationship between prob lematic use of the Internet in general and lower levels of

PAGE 12

12 well-being, including increased leve ls of depression and loneliness, and decreased self-esteem (Caplan, 2002; Morahan-Mar tin & Schumacher, 2003; Y oung & Rogers, 1998). Our study sought to replicate this finding with problem atic usage of social networking sites.

PAGE 13

13 CHAPTER 2 LITERATURE REVIEW Social networking sites becam e a recognized phenomenon on the Internet with the development of Friendster in 2002. Since this time, a number of other social networking sites, such as MySpace and Facebook, have been created and are used by millions of people all over the world. Boyd and Ellison (2007) define social networking sites as web -based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system. The authors highlight that the purpose of these sites is primar ily to communicate with and make visible to others a persons social network. Though meeting strangers outs ide of ones existing network is possible on these sites, the primary way in which these site s are utilized by site me mbers is to maintain current relationships with friends and family. Th ese sites are differentiate d from other Internet sites like message boards and online communities in that the sites are primarily organized around people (the user and the users social network), rather than around a common interest or goal (Boyd & Ellison, 2007). Social networ king sites are more egocentrica lly structured in that the user is the center of his or her own community, as opposed to public forums or message boards which are topically structured. These websites take the communicative func tions encouraged by older technology like email and AOL instant messenger to another level. Profile pages give users a place to express their individual personalities, and they can share with the world intimate details of their lives. Even photographs of the user and the users friends can be uploaded to profile pages. One of the most important purposes of profile pages, howev er, is to provide a place for users to display friend lists and messages from friends. Other users can view friend profiles and expand their own

PAGE 14

14 network through communication with friends of fr iends. Users can send others friend requests in order to add other users to th eir own friend list. Friend lists can grow to be quite large, with many users maintaining lists of friends in triple di gits, while others prefer ring to keep their friend lists limited to only close, existing friendships In general, however, most friend lists are composed of users who have some shared connec tion, such as a shared class year at a school (Ellison, Steinfeld, & Lampe, 2007). In recent ye ars, privacy concerns relating to social networking sites have become an issue covered by the media (George, 2006). To protect the privacy of the user and the users friends, most sites provide privacy features that allow for users to specify who can view their pr ofile and their list of friends. History of Social Networking Sites Though Friendster is often given credit as the si te that began the online social networking phenom enon, it was not the first social networking s ite created. The first of these sites, called SixDegrees.com, was launched in 1997 (Boyd & Ellison, 2007). Using Six Degrees, site members could develop personal profiles and friend lists, as well as explore others friend lists. By 2000, however, the site shut down due to insufficient members and was declared by the site founder to be ahead of its time. The site fa iled because people simply did not have enough friends online to promote extensive use of the website. Other social ne tworking sites sprung up during the late 1990s, including AsianAvenue, BlackPlanet, MiGente, and LiveJournal, but none enjoyed enough popularity to make much of an im pact culturally. That changed with the creation of Friendster in 2002 (Boyd & Ellison, 2007). Origin ally designed as a competitor to Match.com, it grew into a massive social networking site wh ere members could network within their current social circle as well as meet friends of friends. The site became hugely popular and was recognized by the media in the beginning of 2003. A large membership surge as a result of the media coverage overwhelmed the site and ultima tely, the sites popularity declined. Despite the

PAGE 15

15 failure of Friendster, it began a new cultural phenomenon that would forever change the way millions of people use the Internet and how a generation of adolescents communicate with one another. From the start of the media recogniti on in 2003, many new social networking sites were created, the most successful of which are MySpace and Facebook. MySpace MySpace is a popular so cial networking site which was developed in August of 2003 ( www.myspace.com ). It is currently the worlds sixth m ost popular website and has become a fixture in popular culture. Until recently, My Space was the most popular of the social networking sites, receiving 80% of the visits to online social netw orking websites. As of January 2008, the site counted 110 million people as mont hly active users, with new registrations occurring at a rate of 300,000 people per da y (Owyang, 2008). MySpace began in the United States but has become an inte rnational phenomenon with the crea tion of versions for China and the United Kingdom. The premise of MySpace is to bring friends together and to provide a place for networking and making new online friends. User s post profiles about themselves to create their own page that other users can visit. Pict ures, favorite music, and even videos can be uploaded in order to personalize profile pages. Ea ch user maintains a friend list, and friends leave messages for each other that everyone can r ead. If a user wants to become friends with another user, it is as simple as sending an online friend request. Facebook Facebook is another social networ king site that has seen a su rge in popularity in the past year. As of April 2008, Facebook su rpassed MySp ace to become the most-trafficked social media website in the world and the four th most popular website in general ( www.facebook.com ). As of July 2008, the site has over 80 m illion active members worldwide and is growing at a rate of 250,000 new registrants a day. Founded in February 2004, Facebook was developed by a

PAGE 16

16 Harvard University student who intended it as a way to get to know other students on the Harvard campus. The site is based upon the pa per facebooks given to incoming Freshman students at many universities to he lp students learn more about one another. Due to the extreme popularity of the site at Harvard, the creator ex panded Facebook to include other universities. As the demand for the site grew, expansion to high schools and large companies also took place. The site now has over 55,000 regional, work-related, collegiate, and high school networks. Facebook use on college campuses is quite popular. Ellison, Steinf ield, and Lampe (2007) found that 94% of the 286 undergraduates sampled in their study were Facebook users. Like MySpace, users of Facebook can post messages on each others profile pages and private dialogs can occur via email-like messages to another users i nbox. In addition, Facebook members can join virtual groups based on common interests. Users can also search Facebook for other users that have something in common with them, like a hometown, college, or high school. Though social networking sites like MySpace and Facebook have gained widespread use in recent years, little research has been conduc ted regarding the use of these sites. Our study was interested in the excessive or problematic use of these sites, in cluding the prevalence, predictors, and impact of problematic usage of social ne tworking sites for college students. Given the nonexistent research on this topic, the literature on problematic general Internet use or Internet addiction was examined as a basis for our study. Problematic Internet Use or Internet Addiction In recent years, literatu re has begun to accumula te about the existence of Internet addiction or problematic Internet use (PIU). Countries like South Korea a nd China are at the forefront of the research in this area, as Internet addiction has become a serious issue in these countries (Block, 2008). The South Korean government considers 2.1% of its population ages 6-19 (approximately 210,000 children) to be Internet a ddicts and a large port ion of this population

PAGE 17

17 requires treatment, including psychotropic me dications and/or hospi talization (Choi, 2007). Internet addiction is now consider ed one of the most severe public health issues in South Korea. The situation in China is not much different, with one report stating that 13.7% of Chinese adolescents (approximately 10 million) are consid ered Internet addicts (Block, 2008). In the United States, however, Internet ad diction is only slowly coming to the attention of researchers. Part of the difficulty in identifying the presence of Internet addicti on in this country is due to a lack of an agreed upon definition, the proper terminology to be used, and the etiology of Internet addiction. Internet-related problems have been termed Internet addiction (Young, 1996a), pathological Internet use (Dav is, 2001), Internet dependency (S cherer, 1997), and problematic Internet use (Caplan, 2007; Yellowlees & Mark s, 2007). The term utilized is dependent upon how one defines the Internet-related problems. Griffiths (1998) believed in the term Internet addiction, as he saw Internet use which interferes with an individuals functioning as a type of technological addiction. He points out that Internet addiction does not necessarily mean elevated levels of Internet use, but that some elevated use does turn into an a ddiction. Kandell (1998) also defines Internet addiction, which he states broadly as a psychological dependence on the Internet, regardless of the type of activity once logged on (p. 12). Other researchers propose more detailed defini tions of Internet addiction, some developing assessment instruments based on these defini tions. Young (1996a, 1998, 1999) was one of the first to examine Internet addiction and has become a major figure in this area, even developing a website ( www.netaddiction.com) for the Center for In ternet Addiction R ecovery. To define Internet addiction, Young m odified the DSM-IV criteria for pat hological gambling to develop a set of criteria which comprises an eight-item Internet Addiction Diagnostic Questionnaire (DQ). Examples of items include, Do you feel preoccupi ed with the Internet (t hink about previous on-

PAGE 18

18 line activity or anticipate next on-line session)?, Do you feel the need to use the Internet with increasing amounts of time in order to achieve satisfaction?, Have you repeatedly made unsuccessful efforts to control, cut back, or stop Internet use?, Do you feel restless, moody, depressed, or irritable when attempting to cu t down or stop Internet use?, and Have you jeopardized or risked the loss of a significant re lationship, job, educationa l or career opportunity because of the Internet? To be considered an I nternet addict a person mu st meet at least five of the diagnostic criteria in the questionnaire. In order to ex amine the prevalence of Internet addiction in the general populat ion, Young (1996a) administered th e Internet Addiction DQ to 596 participants recr uited through advertis ements, including postings on electronic support groups geared towards Internet addiction and as a search result for those who searched for the keywords "Internet addiction" on web search engines. The study found that 80% of the participants could be classified as Internet dependents, calling in to question the sensitivity of the measure and the sampling methods utilized. Beard and Wolf (2001) modified Youngs (1996a) criteria due to what they believed was a lack of clarity and an improper comparison to a DSM-IV disorder. Inst ead of pathological gambling, the authors believed that the DSM-IV s ubstance abuse criteria were more appropriate for developing the diagnostic criteria of Inte rnet addiction. Unlike Young, the authors were specific about which five criteria must be met in order to give a diagnosis of Internet addiction. These include: (1) a preoccupation with the Inte rnet, (2) a need to use the Internet with increasing amounts of time in order to achieve sa tisfaction, (3) unsuccessful efforts have been made to control, cut back, or stop Internet use, (4) when attempting to cut down or stop Internet use, the person is restless, moody, depressed, or irritable and (5) the pe rson has stayed online longer than originally intended. In addition, the au thors believe that at le ast one of the following

PAGE 19

19 criteria must also be met in order to make a diagnosis: the person (1) ha s jeopardized or risked the loss of a significant relati onship, job, educational or care er opportunity because of the Internet, (2) has lied to family members, a th erapist, or others to conceal the extent of involvement with the Internet, a nd (3) uses the Internet as a wa y of escaping from problems or for relieving a dysphoric mood (e.g., feelings of helplessness, guilt, anxiety, depression). The diagnostic criteria set forth by B eard and Wolf were not intended by the authors to be used in a self-report format, but rather as a diagnostic interview by a professional. Griffiths (1998) also defined Internet addiction based on the DSM-IV criteria for substance abuse. His criteria include symptoms of tolerance, withdrawal, crav ing, and negative life consequences, as well as additi onal symptoms of salience of the activity to the individual, changes in mood when engaging in the activity, and a tendency to relapse after the activity is discontinued. Brenner (1997) additionally modified DSM-IV criteria for substance abuse as a basis for his 32-item true-false Internet-Related Addictive Behavior Inventory (IRABI). The measure includes questions relate d to negative life consequences and side effects of Internet addiction such as online relati onship problems and resultant time management issues. Culturespecific measures of Internet addiction have al so been developed, mainly for use in China and Taiwan. These include the Chinese Internet Addiction Scale (CIAS; Chen and Chou, 1999), the Chinese-translated IRABI (Chou & Hsiao, 2000), and the Internet Addictio n Scale for Taiwan High School Students (IAST; Lin & Tsai, 1999). Though many researchers define problems with Internet use as an addiction, other researchers do not believe that th e term addiction should be asso ciated with Internet use, as this term is reserved for physiological depende nce between a person and a stimulus, such as a substance. Davis (2001) suggests the term pathological Internet use (PIU) be used instead, and

PAGE 20

20 that dependence be discussed in relation to the Internet rather than addiction. He defines PIU as consisting of two types: specific pathological In ternet use and generalized pathological Internet use. Specific PIU refers to a content-specifi c dependency, such as a dependency on online gaming or online gambling. It is assumed that some form of this depe ndence would still exist even in the absence of the In ternet (e.g., gambling addiction at casinos instead of online gambling). Generalized PIU, on the other hand, refers to misuse of the Internet independent of specific Internet content. It us ually involves a dependency on the unique social environment that the Internet can provide, and therefore the de pendence in an altern ative form would not necessarily exist in the absence of the Internet, as it would with specific PIU. Generalized PIU is characterized by a general sense of wasting tim e online without a clear purpose, or using the Internet for social functions in order to remain in a virtual social life. Symptoms of PIU include: obsessive thoughts about the Internet, diminished impulse contro l, inability to decrease Internet use, anticipating future online use, less time spen t on other pleasurable off line activities, social isolation, sense of guilt about online use, and a fee ling that the Internet is an individuals only friend. Caplan (2002) agreed with Davis (2001) that the term addiction should not be associated with the Internet. He suggested the term problematic Internet use (PIU) to describe the maladaptive cognitions and behaviors associated with Internet use that result in a negative impact on academic, professional, and social functioning. Following Daviss work, Caplan characterized PIU as including cognitive and be havioral symptoms such as mood alteration, perception of social benefits of the Internet, compulsive use, excessive use, withdrawal, and perceived social control when interacting with others online compared to face-to-face. Based on his own definition and on Daviss model, Caplan developed a measure, the General Problematic

PAGE 21

21 Internet Use Scale (GPIUS), to assess PIU. This measure will be discussed in more detail in the materials section of this paper as it was modi fied and utilized in our study. Accordingly, Caplans term problematic internet use (PIU) will be used for the remainder of this paper to refer to the phenomenon of Int ernet addiction, or the malada ptive cognitions and behaviors related to Internet use which may interfere with normal daily living (Caplan, 2002). The term problematic social networking site use will be utilized to refer to a specific problematic or dependent relationship that a person may develop with these types of sites. In addition to a lack of clarity as to how to characterize PIU, a deba te also exists among researchers as to how to classify PIU within th e existing range of mental disorders. Along with this, many are considering whether Internet Addiction should be added to the next revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM; APA, 1995). Yellowlees and Marks (2007), the most recent authors to weigh in on the debate, divide the writing on this topic into two schools of thought. On one side are th e researchers who propose that individuals are addicted to the use of the Intern et in general and Internet addic tion should be classified as an emerging psychiatric disorder in the next revision of the DS M. For example, Marks (1990) advocates the addition of Internet addiction to the list of other nonchemical behavioral addictions (e.g., pathological gamb ling) included in the DSM. On the other side of the debate are those researchers who believe that individuals can develop a problematic relationship with specific online applications available via the Internet, su ch as online gambling, shopping, chatting, or pornography. Researchers in this cam p do not believe that PIU warrants a special diagnosis (e.g., Internet addiction) because addictions to on line content can be classified under pre-existing DSM classifications, su ch as pathological gambling or impulse control disorder not otherwise specified (NOS) (B eard & Wolf, 2001; Shapira, Goldsmith, Keck, Khosla, &

PAGE 22

22 McElroy, 2000; Treuer, Fabian, & Furedi, 2001). Mitchell (2000) believes that a separate diagnosis is not warranted because a problematic relationship with the Internet could be the result of an underlying, co-morbid psychological disorder. Griffiths (2000) seconds this by his assertion that Internet addicts do not exist, but that the Internet can provide individuals with a way to engage in other addicti ons. In other words, addictions to the Internet (i.e., addictions to the Internet itself ) and addictions on the Internet (i.e., the Internet provides another forum for an individual to engage in his or her addiction, su ch as online gambling or gaming) have different meanings and implications. Yellowees and Mark s review the research provided by both schools of thought and conclude that t hough researchers in both camps do not deny that individuals can develop a problematic relationship with the Internet Internet addiction does not warrant a special diagnosis as an emerging disorder This certainly is not the end of this debate and our study did not attempt to settle the argument. Rather, the purpose of our study was to simply determine if a dependency on a specific Internet application, so cial networking sites, can exist. First aim: Existence of Problematic Social Networking Site Use Most research has exam ined dependency on th e Internet in general without addressing a possible dependency on specific Intern et applications. Our study adds to the literature in this area by exploring the possible existence of a dependenc y specifically on social networking sites. As previously defined, social netw orking sites are considered we bsites on the Internet organized around people (the user and the users network) which allow i ndividuals to (1) construct a public or semi-public profile within a bounded syst em, (2) articulate a list of other users with whom they share a connections, and (3) view a nd traverse their list of connections and those made by others within the system (Boyd & Elli son, 2007). Some research provides justification for believing that a dependency on these types of sites may exist. When researchers of PIU or Internet addiction ask study participants what Internet applications they typically use, a

PAGE 23

23 significantly higher number of PIU users note that they use the communicative functions of the Internet over non-communicative f unctions. Young (1996a) found that the majority of Internet users classified as dependent by her Internet Addiction DQ use online interactive applications. Chou and colleagues (Chou, Chou, & Tyan, 1999; Chou & Hsiao, 2000) also found that addicted individuals were more likely to use the Internet for communication functions rather than for other non-communication uses. Morahan-Ma rtin and Schumacher (2000) reported that pathological Internet users in their study utilize the Internet more for meeting new people, emotional support, and interactiv e applications than non-pathologi cal users. These findings seem to support the idea that communicative or interactive ap plications of the Internet might be particularly addictive. It has already been shown that a large majority of Internet users are utilizing communicative appl ications like social networking sites, given that sites like MySpace and Facebook are within the top most visited sites in the world. To emphasize, they are some of the most visited out of all websites, not just social networking sites (Owyang, 2008). This fact combined with research supporting the addictive nature of Internet co mmunication applications suggest that a dependency on soci al networking sites may exist for some people. Therefore, our study sought to explore the existence and prevalence of problematic social networking site use. This was examined in a population of college stud ents because this group has been shown to be frequent users of social networ king sites (Ellison et al., 2007) and vulnerable to PIU (Kandell, 1998, Moore, 1995). College students may be particularly at risk for developing a problematic relationship with the Internet. Kandell (1998) believes that college students de velopmental stage of solidifying their sense of identity and forming meaningful and intimate relationships puts them at risk for making the Internet and its social functions an overly central part of their lives. Moore (1995)

PAGE 24

24 also cited accessibility of the Internet and flexibility of schedules as two factors that may contribute to increased vulnerab ility for college students to de velop PIU. In addition, most college students are part of the Net Generati on, another group which ha s been identified as a vulnerable population to PIU. The Net-geners or Millennials, as they are sometimes called, are individuals born in the early 1980s or later that are the first generation to grow up with personal computers and the Internet. Busi nessWeek (Hempel, 2005) referre d to this generation as the MySpace Generation due to the high use of th e website among this cohort. Howe & Strauss (2000) report that Millennials have higher soci o-economic status, more education, and are more ethnically diverse than any other generation before them. Making up over 30% of the United States population, this generation is one of the largest, and ther efore problems to which this group may be particularly vulnerable shoul d be of interest to researchers. Some have proposed that characteristics which de fine this generation could contribute to a particularly high vulnerability fo r PIU. Citing Tapscotts (1998) book about the rise of the Net Generation, Leung (2004) identified the following c ohort characteristics that might make this generation vulnerable to PIU. These include: (1) a global orientation, (2) being emotionally uninhibited, (3) having a strong be lief in the right to information and learning, (4) being technologically savvy, and (5) ha ving a preoccupation with maturity and adulthood. For Netgeners with these characteristics, the Intern ets social communicati on applications may be particularly seductive. Therefore, using college students of this generation to study the possibility of a dependency on social networking sites is especially appropriate. Second Aim: Predictors of Problematic Social Networking Site Use After establishing the existence of problem atic social networking site use, the second aim of our study was to examine possible predicto rs of problematic usage. Though many possible predictors exist, our study examined two: so cial anxiety and the degree to which a person

PAGE 25

25 belongs to an offline social network. The selec tion of these predictors was based on the proposed etiology of problematic general Internet use refl ected in the literature as well as research suggesting an association between th ese predictors and general PIU. Etiology of Problematic Internet Use Etiologies of PIU have been suggested by a number of different authors. For example, Chou and colleagues (1999) applied Stephensons (1998) Play Theory of Mass Communication to explain their version of the etiology of PIU. When applied to In ternet use, this theory proposes that the Internet provides a pleasurable and rein forcing communication experience and that this reinforcement entices the individua l to continuously use the Intern et to the point that overuse leads to addiction-like behavior Another explanation proposed by Suler (1999) states that an individual becomes dependent on the Internet due to the Internets ability to satisfy one or more unfulfilled needs of the individual. Suler identified six needs that the Internet could be used to fulfill, including (1) sex, (2) an altered state of consciousness, (3) achievement and mastery, (4) belonging, (5) relationships, and (6) self-a ctualization/transce ndence of self. The most detailed explanation of PIU development is given by Davis (2001). He proposed a cognitive-behavioral model of PIU in which maladaptive cognitions, together with behaviors that intensify problematic responses, resu lt in PIU. In Daviss model, it is the cognitive symptoms that lead to the behavioral sympto ms which are most commonly emphasized in other models. The conditions under which PIU occurs are considered with in a diathesis-stress model, in which a vulnerability or psychopathology exists which predisposes a person to PIU development should a stressor occur. In Davis s model, the stressor wh ich may encourage the PIU to develop is the introduction of the Internet or a new technology on the Internet (e.g., social networking sites). By itself, the introduction of this new technol ogy does not cause PIU, but it is a necessary condition for PIU to occur. Reinforcem ent of the use of Internet technology is a key

PAGE 26

26 factor which allows PIU to develop. For exampl e, Davis believed that socially isolated individuals may find the social environment offe red by the Internet par ticularly reinforcing because social applications on the Internet allow for a non-threatening environment for communication. If an individual is reinforced for using the online activ ity by receiving some kind of positive response, the individual will c ontinue Internet use, and may be conditioned to use the online activity more and more to receive the desired response. Davis contends that maladaptive cognitions are the most important factor in PIU development and maintenance. He divides thes e maladaptive cognitions into two subtypes thoughts about self and thoughts about the worl d. Thoughts about self usually occur in a ruminative-type cognitive style, in which the individual frequently th inks about his or her problematic relationship with the Internet (e .g., trying to figure out why overuse is occurring, reading about Internet overuse, constantly ta lking to friends about Internet overuse). The rumination of thoughts interferes with the persons ability to problem solve and engage in new behaviors. Hence, the PIU is exacerbated by maladaptive cognitions. Cognitions about self that maintain PIU also include self-doubt, low self -efficacy, and negative self-appraisal (e.g., I am worthless offline, but online I am somebody, I am a failure when I am offline). The individual with these types of cognitions may use the Internet to achieve a more positive view of self through social acceptance in a le ss-threatening environment. The Internet is considered less socially threatening than face-to-face social interactions becau se it allows for a higher degree of self-disclosure and risk-taking due to reduced au ditory and visual cues, allowance of time to think about responses, and anonymity (Morah an-Martin & Schumacher, 2000; Peter, Valkenburg, & Schouten, 2005).

PAGE 27

27 Thoughts about the world are the other maladaptive cognitive subtype proposed by Davis (2001). These involve the generaliz ation of specific events to gl obal patterns or engagement in all-or-nothing type thinking (e.g., Nobody loves me offline, The Intern et is the only place I am respected). Both maladaptive thoughts ab out self and thoughts about the world are automatically triggered by stimuli associated with the Internet, and are automatically enacted when one engages in online activ ities. These maladaptive cognitions therefore lead to PIU and the maintenance of problematic use. Predictors of Problematic So cial Netw orking Site Use The predictors of problematic social networki ng site use included in our study (i.e., social anxiety and a lack of belonging to a local social network) fit w ith these given etiologies of problematic general Internet use. Caplan (2007) speaks directly to th e fit of social anxiety within Daviss (2001) model of PIU development. He s uggests that socially an xious individuals have more maladaptive cognitions about their social competence than others without these issues. Citing the self-presentational th eory of social anxiety (Sch lenker & Leary, 1982), Caplan explains that social anxiety is created due to a desire to create a positive impression in a social situation, but a lack of confidence in ones social and self-pre sentation skills cr eates anxiety in social encounters. To reduce the anxiety, social ly anxious individuals will turn to low-risk communicative environments in which they have mo re of a chance of presenting themselves in a positive way. To relate this to Daviss m odel of PIU development, social anxiety would be considered the underlying vulnerability or existing psychopa thology that predisposes the person to PIU development. When online communication, such as a social networking site, is introduced, a socially non-threatening environment is crea ted. For socially anxious individuals, their maladaptive cognitions about their so cial skills, along with a desire to make social connections in

PAGE 28

28 a less risky way, may lead to a preference for this online social environment instead of face-toface communications. PIU development could then occur as the individual continues to receive reinforcement, or the gaining of social accep tance and support, by usi ng this less-threatening online environment. The cognitions and behavior s that drove the socially anxious person to prefer the online environment also act to mainta in online use and the problematic or dependent relationship the person has formed with the online social atmosphere. Caplan examined his model of social anxiety as a contributor to PIU development by collecting data from 343 undergraduate stude nts. He found significance for a model that indicated social anxiety as th e most important predictor of preference for online communication and negative outcomes associated with PIU. Fu rther evidence in support of the relationship between social anxiety and PIU has also been found. Yen and colleagues (Yen, Ko, Yen, Wu, & Yang, 2007) showed that individuals classified as Internet addicts by the Chen Internet Addiction Scale (CIAS; Ko, Yen, Chen, Chen, & Yen, 2005) had higher levels of social phobia, as indicated by the Social Phobia I nventory (SPIN; Connor et al., 2000) than those not classified as addicts. Given that social networ king sites provide a type of online social atmosphere, it is thought that these sites too can provide a less-threatening environment in whic h social support may be gained. Socially anxious individuals may be drawn to these types of sites and then may develop a problematic relationship with them through th e model outlined by Davis (2001) and Caplan (2007). Therefore, our study proposed that social anxiety is a predictor of problematic social networking site use. Within this prediction is the assumption that socially anxious individual s turn to these sites to fulfill a need for social relationships due to a lack of currently belongi ng to an offline social

PAGE 29

29 network. To feel belongingness is a basic need of most human beings. Maslow (1968) ranked it in the middle of his motivational hierarchy, and Bowlby (1969) made it an integral part of his attachment theory. The belongingness hypothesis propos es that humans have an inherent need to form and maintain lasting, positive, and significant relationships with others (Baumeister & Leary, 1995). Anant (1966) defined belonging as a sense of personal involvement in a social system so that persons feel themselves to be an indispensable and integral part of the system (p. 21). Hagerty and colleagues differentiate two types of belonging psychological and sociological (Hagerty, Lynch-Sauer, Patusky, Bouwsema, & Collier, 1992). Psychological belonging refers to the affective st ate, or a persons perception of fit and valu ed involvement in a system. Sociological belonging refers less to the affective state and more to actual membership in a group or system. This type of belonging is more observable than psychological belonging as it can be behaviorally detected via group member ship. In our study, the measure used to assess belonging is a measure of sociological belonging, or actual membership in an offline social network. Literature has shown that socially anxi ous individuals often avoid forming face-to-face relationships (Gambrill, 1996), and have fewer friends (La Greca & Lopez, 1998). For socially anxious people who may avoid f ace-to-face relationships and ther efore lack belonging to an offline social network, online social networki ng sites may provide a less-threatening way to fulfill this need to belong. However, if the need for belonging is only being met through the use of these online sites, individuals may become dependent on these sites to fulfill this need. The idea of socially anxious individuals turning to the Internet for need fulfillment is supported by McKenna and Bargh (2000), who proposed a social compensation hypothesis. This theory suggests that so cially anxious individual s may compensate for thei r lack of social support by using the non-threatening environment of the In ternet to meet friends. Valkenburg and Peter

PAGE 30

30 (2007) found support for the social compensation hypothesis by showing that socially anxious individuals used the Internet fo r online social interactions more than their non-socially anxious counterparts, and perceived the Internet as a sa fer place for intimate self-disclosure. Chak and Leung (2004) also suggest that the Internet can provide a place for individuals to satisfy social and emotional needs that are unf ulfilled by their limited face-to-face social networks. If the person finds the use of social networking site s fulfilling and reinforcing enough, problematic usage with social networking sites may devel op through cognitive and behavioral mechanisms associated with this reinforcement, as outlined by Davis (2001). This idea is also suggested by Suler (1999), who proposed that PIU develops when an unmet need is fulfilled and reinforced through Internet use. He specifically noted one such need as the need to belong. In addition, Griffiths (2000) has shown a relationship between using the Internet for a way to compensate for a lack of social support and Internet addiction. In conclusion, our study examined both the degree to which a person belongs to a social network and social anxiety as pos sible contributors to problematic social networking site use. PIU etiologies and research indicating an asso ciation between social anxiety and a lack of belonging suggest that a lack of belonging may mediate the relati onship between social anxiety and problematic social networking site use. Individuals who are socially anxious may have less of a local social network, and therefore may turn to social networking sites to compensate for an unfulfilled need to belong. Problematic usage may develop as a result of the reinforcing fulfillment of this need provided by social networking sites. Third Aim: Problematic Social Netw orking Site Use and Well-Being Finally, because various studi es have shown a relationshi p between PIU and decreased well-being, the third aim of our study was to exam ine if this same rela tionship exists between problematic usage of social networking sites and college students well-being. Research

PAGE 31

31 specifically looking at problematic social networking site use and its association with well-being does not exist. Therefore, the literature showing a relationshi p between well-being and PIU in general was reviewed as suppor t of our studys predictions. A number of studies have found an association between PIU and a range of variables representing well-being, including levels of depression, lone liness, and self esteem. Yen and colleagues (2007) examined the relationship of PIU with other psychological disorders, including depression. PIU was measured using the Chen Inte rnet Addiction Scale (CIAS; Ko et al., 2005). Depression was assessed using Center for Epid emiological Studies Depression Scale (CES-D; Radloff, 1977). These measures were admini stered to 2,114 high school students in Taiwan. Close to 18% of the students were classified as Internet addicts according to the CIAS. Results of the study indicated that Int ernet addicts had significantl y higher CES-D scores than nonaddicts, indicating that higher le vels of depression were associ ated with PIU. Young and Rogers (1998) also found a relationship between PIU and depression. They administered the Internet Addiction DQ (Young, 1996a) and the Beck Depr ession Inventory-II (BDI-II; Beck, Steer, & Brown, 1996) to 312 adults recru ited through advertisements on th e Internet. They found that those individuals classified as In ternet addicts had highe r levels of depression than non-addicts. Finally, Caplan (2002) additionally found an association between PIU and depression. His Generalized Problematic Internet Use Scale (GPI US) was used as a measure of PIU and the BDIII as a measure of depression. Af ter administering the measures to 386 undergraduates, he found GPIUS scores to be positively correlated with de pression scores, indicating that as PIU increases so do symptoms of depression. Loneliness has also been found to relate to PIU. Kraut and colleagues (Kraut, Patterson, Lundmark, Kiesler, Mukopadhyay, & Scherlis, 1998) found that leve ls of loneliness increased

PAGE 32

32 with the amount of time spent online. Caplan (20 03) also found loneliness, as measured by the UCLA Loneliness Scale (Russell, Peplau, & Cutrona, 1980), to be positively associated with PIU, as measured by the GPIUS. Two additional st udies found a similar result, with pathological Internet users or Internet de pendents scoring higher on the UCLA Loneliness scale than nonpathological users or non-dependents (Morahan-Martin & Schumacher, 2000; Nalwa & Anand, 2003), lending further support for a relati onship between loneliness and PIU. Finally, a negative relationship between PIU a nd self-esteem has also been found in the literature. Caplan (2002) examined the relations hip between PIU, as measured by the GPIUS, and self-esteem, as measured by the Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1965). He found GPIUS scores to have a si gnificant negative rela tionship with RSE scores, indicating that individuals with higher levels of PIU have lower levels of self -esteem. Niemz, Griffiths, and Banyard (2005) also found a significant relati onship between PIU and low self-esteem. PIU was measured using the Pathological Internet Use Scale (Morahan -Martin & Schumacher, 2000) and self-esteem was assessed with the RSE. Thr ee-hundred seventy one undergraduate students completed the measures. PIU scores were found to be negatively correlated with RSE scores, indicating that PIU is associat ed with lower self-esteem. Arms trong, Phillips, and Saling (2000) found similar results. Self-est eem scores, as measured by the Coopersmith Self-Esteem Inventory (SEI; Coopersmith, 1991), predicted PI U, as measured by th e Internet-Related Addictive Behavior Inventory (I RABI; Brenner, 1997). Specifical ly, lower self-esteem scores predicted greater PIU. The limitation in many of these studies is the failure to establish causality in the relationship that has been found between PIU and well-being. It could be that this relationship exists because those suffering from decreased we ll-being such as low self-esteem, loneliness,

PAGE 33

33 and/or depression are more likely to develop PIU. As Daviss (2001) model of PIU development explains, a pre-existing psychopathology or conditi on could make a person more vulnerable to PIU development should a stressor occur. However, this relationship could also exist because PIU negatively impacts an indi viduals well-being. Daviss ( 2001) model provides support for this idea by suggesting that the maladaptive cogn itions which are associated with PIU can have detrimental effects on self-esteem and life sati sfaction. In addition, PIU has been shown to interfere with normal daily functioning in a numbe r of studies. Scherer (1997) found that 13% of participants admitted that Internet use had interf ered with their academics, work performance, or social lives and 2% of respondents viewed the Internet as having an overall negative effect on their daily lives. Internet depe ndents in Youngs (1998) study also reported personal, family, and work problems in relation to PIU. This inte rference in life functioning caused by PIU could negatively impact a persons sense of we ll-being and satisfaction with life. Given that theories of PIU development and maintenance find support for both directions in the causal relationship between PIU and we ll-being, it is most likely that the causal relationship goes both ways (Chou, Condron, & Belland, 2005; Morahan-Martin & Schumacher, 2003). As explained by Caplan (2003) and Davis (2001), lowe r well-being may make one susceptible to PIU development, but the maladap tive cognitions and behavi ors that maintain PIU may worsen these conditions. As well, the interference that PIU can have with daily functioning in many areas of life, such as academic, work, and family functioning, may have a negative impact on a persons well-being. Though research showing a relationship between general PIU and well-being was used to support the purpose of our study, our study did no t attempt to find evidence for a causal relationship between these constructs. Rather, ou r study aimed to establish the existence of a

PAGE 34

34 relationship between problematic social networki ng site use and well-bein g given that this area has not been researched previously. Study Overview In summ ary, the main objective of our study wa s to explore a novel area in the Internet addiction literature the problematic use of social networking sites like MySpace and Facebook among college students. The first aim of our study was to establish the ex istence of problematic social networking site use by surveying a coll ege undergraduate populatio n. The second aim of our study was to explore some possible predictors of problematic usage of social networking sites, including social anxiety and the degree to which a person belongs to an offline social network. Finally, our study aimed to explore the association of problema tic social networking site use with well-being for college students. Given these study objectives, the following predictions were set forth: (1) Problematic social networking site use does exist and the measure used to assess it will show a range of scores, including scores indicating a high presence of symptoms of problematic use. This prediction is supporte d by research indicating that problematic general Internet users utilize the communicativ e functions of the Internet more than noncommunicative functions (Chou et al., 1999; Chou & Hsiao, 2000; Young, 1996a). Given that social networking sites are the most popular social application on the Internet, it is predicted that for some, a dependency on social ne tworking sites, rather than the Internet in general, may exist. (2) The degree to which a person belongs to an offline social network will mediate the relationship between social anxi ety and problematic social netw orking site use. In order to find evidence for this, the followi ng sub-predictions must be met: a. The degree of social anxiety will be positivel y related to problematic use of social networking sites. Higher levels of social a nxiety will be predictive of higher levels of problematic social networking site use. Th is is based on the literature indicating social anxiety as a predictor of PIU de velopment (Caplan, 2007; Yen et al., 2007). b. The degree to which a person belongs to an offline social network will be negatively related to problematic use of social networking sites. Less belonging will be predictive of higher problematic social netw orking site use. This is based on the social compensation hypothesis (McKenna & Bargh, 2000) that individuals with a

PAGE 35

35 limited social network may turn to the Intern et to compensate for this, and that this need fulfillment may result in PI U (Griffiths, 2000; Suler, 1999). c. The degree of social anxiety will be nega tively related to the degree to which a person belongs to an offline social network. More socially anxious people will have a decreased sense of belonging to an off line social network. This is based on the literature indicating that socially anxious individuals often avoid forming face-toface relationships (Gambrill, 1996) and have fewer friends (La Greca & Lopez, 1998). d. By including the variable of belonging to an offline social network, the relationship between social anxiety and problematic social networking site use will be significantly decreased, indicat ing that the degree to which a person belongs to an offline social network mediates the relationship between these two variables. (3) Problematic use of social networking sites is expected to be related to decreased wellbeing, based on the literature linking general PIU to decreased well-being (Caplan, 2002; Morahan-Martin & Schumacher, 2003; Young & Rogers, 1998). Specifically, a significant negative relationship is expected between probl ematic use of social networking sites and the well-being indicators of self-esteem, satisfaction with life, and happiness, and a significant positive relationship is expected between the problematic social networking site use and the well-being indicators of depression and loneliness.

PAGE 36

36 CHAPTER 3 MATERIALS AND METHODS Participants Undergraduate students attending th e University of Florida were recruited for participation in our study due to this type of populations uni que vulnerability to PIU that is noted in the literature (Kandell, 1998; Leung, 200 4; Moore, 1995), as well as the ease of recruiting this population. Three hundred sixty-se ven students completed the online survey. Of these 367 students, approximately 95% (N = 350) were social networking site users. The ethnicity make-up of this sample consisted of approximately 13 % African-American (N = 49), 9% Asian/Pacific Islander (N = 34), 13% Hispanic /Latino (N = 47), 60% Caucas ian (N = 221), 1% Middle Eastern/Arab (N = 4), and 3% Biracial/Multiracial individuals (N = 12). Eighty-one percent (N = 297) of the participants were female and 19% (N = 70) were male. No significant gender or ethnic differences were found between social networking site users and non-users. The participants ranged in age fr om 17 to 28, with the average age being 20 years (SD = 1.60). Students classified as Juniors made up the largest portion of the sample (34%, N = 125), followed by Sophomore students at 26% (N = 95) Senior students at 21% (N = 77), and Freshman students at 19% (N = 70). The major ity of students live off campus (71%, N = 261). All participants were treated in accordance with the Ethical Principl es of Psychologists and Code of Conduct (American Ps ychological Asso ciation, 1992). Measures Eight m easures were used in our study to assess the following: the degree to which a person belongs to a local social network, problematic social networ king site use, social anxiety, and the well-being constructs of self-esteem, happiness, satisfaction with life, loneliness, and depression. A short demographic survey was also employed to provide information on the

PAGE 37

37 characteristics of the study participants, includi ng age, academic classification (e.g., Freshman), race/ethnicity, and gender. In addition, questions were included that explored the participants social networking site use. These questions in cluded an assessment of the type of social networking site(s) most commonly used by the participant (i.e., MySpace, Facebook, Friendster, or Other), the length of time the participant has been a member, for what purpose the participant uses social networking sites, and the average am ount of time the participant uses these sites. Amount of site use was assessed with the que stions: (a) How many days per week do you usually visit a social networking site such as MySpace or Facebook?, (b) On a typical day, how many times do you visit social networking si tes, such as MySpace or Facebook?, and (c) If you visit one or more of thes e sites, how many minutes do you usually stay on the site each time? These items are based on the items used by Valkenburg and colleagues (2006) to assess participants use of the social networking site CU2. In addition to these questions, two items were given at the end of the survey which asked: (a) Has anyone ever told you that your use of online social networking sites like MySpace or F acebook is problematic or interferes with your life (daily functioning)?, and (b) Do you feel th at your use of online social networking sites like MySpace or Facebook is problematic or inte rferes with your life (daily functioning?) Belonging to an Offline Social Network The Interpersonal Support Evaluation List (I SEL; Cohen & Hoberm an, 1983) was used to assess the degree to which a person belongs to an offline social network. This is multidimensional measure which has two forms a 40item version for the general population and a 48-item version for college students. Only one of the subscales is relevant for use in the current study Belonging. This scale is comprised of 12 items which assess the extent to which an individual feels a part of a social group with common interest s or the perceived availability of friends which one can spend time with. In the case of the college version of th is measure, it is the

PAGE 38

38 extent to which an individual feels a part of a social group on campus. Refer to appendix A for a full list of items. Respondents are asked to rate each item based on their level of agreement with the statement. Answer choices include: Mostly False and Mostly True. A score of 0 or a score of 1 is given, depending on whether the statement is worded positively or negatively. Total scores can range from 0 to 12, with higher scor es indicating more bel ongingness to a social network. The measure has been shown to have good internal consistency, with estimates ranging from 0.88 to 0.90 (Cohen & Hoberman, 1983). Problematic Social Networking Site Use The Generalized Problem atic Internet Use S cale (GPIUS; Caplan, 2002) was modified to examine levels of problematic social networking site use among study participants. This measure is a 29-item self-report questi onnaire based on Daviss (2001) cognitive-behavioral model of PIU. It measures the prevalence of cognitive and behavioral symptoms of PIU derived from this model, as well as the degree to which the prob lematic use interferes with the individuals functioning in personal, academic, and professional areas of his or her life. The measure was modified to address PIU specifica lly with social networking site use, or the degree to which the respondent has a problematic or de pendent relationship with their us e of social networking sites. This was done by replacing the word Internet or online with the words social networking sites. For example, the item When not onlin e, I wonder what is happening online was changed to When not on a social networking site, I wonder what is happening on that site. A full list of items is included in Appendix B. It ems encompass seven areas which assess (a) mood alteration (i.e., the extent to which the individual uses social networking sites to change affective states), (b) perceived social benefits (i.e., the extent to which an indi vidual perceives social networking sites as providing gr eater social benefits than face-to-face communication), (c) perceived social control (i.e., the extent to which an individual perceives social control when

PAGE 39

39 using social networking sites for communication), (d) withdrawal (i.e., the degree of difficulty in staying away from using social networking sites), (e) compulsivity (i.e., the inability to control, reduce, or stop social networking site use, along w ith feelings of guilt about lack of control), (f) excessive social networking site use (i.e., the degree to which an individual feels he or she spends too much time using social networking site s, or more than the planned amount of time), and (g) negative outcomes (i.e., th e severity of personal, social and professional problems as a result of social networking site use). Each item asks respondents to rank their agreement with the statement on a 5-point Likert scale, from 1 (Strongly Disagree ) to 5 ( Strongly Agree ). Scores may range from 29 to 145. The higher the total score, the greater the degree of problematic social networking site use. A specifi c cut-off score which would in dicate problematic use was not designated for this instrument. Caplan did not believe in proposing a cut-off score defining a problematic Internet user versus a non-problematic user, as the theory on which the measure is based (i.e., Daviss (2001) cognitive-behavioral model) stat ed that the adaptive versus maladaptive nature of Internet use is dependent upon the individual and th e effects the use has on the individuals life. Caplan (2002) developed the GPIUS by utilizing specific examples of PIU cognitions, behaviors, and outcomes proposed by Daviss th eory, as well as including items from other measures of PIU in the literature (Armstrong et al., 2000; Mora han-Martin & Schumacher, 2000; Scherer, 1997; Young, 1998). In order to identify the factor structure, Caplan conducted an exploratory factor analysis on data he collected from an administration of the GPIUS to 386 undergraduates, resulting in a fina l list of items which comprise the seven areas outlined above. The seven areas identified were highly consistent with those cognitions, behaviors, and outcomes proposed by Daviss model of PIU. Reliability an alyses indicated high in ternal consistencies,

PAGE 40

40 with alpha coefficients ranging from 0.78 to 0.85 for the seven subscales. Validity support is indicated by the significant relationship between GPIUS scores and measures of depression, loneliness, and self-esteem (Caplan, 2002, 2003) in a direction consistent with Daviss (2001) model of PIU. Social Anxiety The Social Avoidance and Distress scale (SAD; Watson & Friend, 1969) was used in our study to assess an individuals de gree of social anxiety. This is a 28-item scale in a true-false format which assesses feelings of distress/discomfort and avoidance of social interactions. Examples of items include, "I am usually at ease when talking to someone of the opposite sex," and "I try to avoid formal soci al occasions." Refer to Appendix C for a full list of items. This scale was intended as a measure of comfort in face-to-face social intera ctions, and this was emphasized in our study by instructing respondents to answer the items in reference to their faceto-face social interactions, not online social intera ctions. Higher scores on this measure indicate higher levels of social anxiet y. Internal consistency estimate s have ranged from 0.77 to 0.93 (Watson & Friend, 1969; Caplan, 2007). Well-Being Multip le measures were used to assess well-being in our study. These included measures of depression, loneliness, self-esteem, happiness, and sa tisfaction with life. Depression The Center for Epidem iologic Studies Depr ession Scale (CES-D; Radloff, 1977) was utilized to assess depression levels for our study s participants. This measure was designed for use with a general, non-clinical population and assesses the curre nt frequency of depressive symptoms for an individual. The 20-item questio nnaire emphasizes depressed affect or mood, psychomotor retardation, loss of appetite, sleep disturbance, and feelings of guilt, worthlessness,

PAGE 41

41 helplessness, and hopelessness. For each item the respondent is asked to indicate how frequently, in the past week, he or she has experienced the symptom expressed by the item. Response choices range from rarely or never (less than 1 day), some or a little of the time (1 or 2 days), occasionally (3 or 4 days), and most or all of the time (5 up to 7 days). A score of 0 is given for the first response choice up to a score of 3 for the last response choice. Positive items are reverse scored. Total scores ra nge from 0 to 60, with higher sc ores indicating the presence of more symptoms of depression. The measure has a well-established reliab ility (Hann, Winter, & Jacobsen, 1999). An internal consistency of 0.85 has been found for the general population (Radloff, 1997). Validity has been established by finding significant correlations between the CES-D and other measures of depressi on, including the BDI-II (Shafer, 2006). Loneliness The third version of the UCLA Loneliness Sc ale (Russell, 1996) was used to m easure levels of loneliness. The scale contains 20 posit ively and negatively worded items that assess an individuals experience of lonelin ess. This revised measure is a simplified adaptation of the older version due to complaints about the readability of some of the items (Russell, 1996). Possible item responses range from 1 ( never ) to 4 ( always ). Scores may range from 0 to 80, with higher scores indicating a higher degree of loneliness. Good reliability with college students has been established (Cronbachs alpha = 0.92) (Russell, 1996). Validity w ith college students has also been indicated by the significan t positive correlations found between the UC LA Loneliness Scale and other measures of loneliness, as well as signi ficant negative correlations with measures of social support in an undergraduate student population (Russell, 1996). Self-Esteem The Rosenberg Self-Esteem Scale (RSE; Rose nberg, 1965) is a widely used, 10-item selfreport measure that was utilized in our study to assess self-esteem. The scale items assess an

PAGE 42

42 individuals perception of general self-worth or positive self-esteem (e.g., I feel that Im a person of worth, at least on an equal plane with ot hers). Each item has a 4-point Likert scale in which the participant is asked to rate his or her agreement, ranging from 0 ( strongly disagree) to 3 ( strongly agree ). Reverse scored items are also incl uded (e.g., At times, I think I am no good at all). A score is calculated by summing the point s for each item, with scores ranging from 0 to 30. The higher the score, the hi gher the self-esteem. Construct and convergent reliability of the measure has been demonstrated (Goldsmith, 1986) Internal consistency estimates have ranged from 0.82 to 0.93 (Goldsmith, 1986). Validity has been established through findings of a correlation between RSE scores and depression (R ice, Ashby, & Slaney, 1998) as well as other constructs (e.g., anxiety) in expected directions (Gol dsmith, 1986; Rosenberg, 1965). Happiness Personal happiness was assessed using the Ox ford Happiness Questionnaire (Hills & Argyle, 200 2). This scale contains 29 positively and negatively worded items meant to broadly measure personal happiness. Partic ipants rate on a 6-point Likert scale their degree of agreement with each of the items in the scale. Higher scores indicate higher levels of personal happiness. Sample items include I feel that life is very re warding and I do not have a particular sense of meaning and purpose in my life. Refer to Appe ndix D for a full list of items. Studies have shown the scale to be correlated in the expected direction with other measures of subjective wellbeing. Reliability estimates have averaged around 0.91 (Hills & Argyle, 2002). Satisfaction with life The Satisfaction W ith Life Scale (SWLS; Diener, Emmons, Larsen, & Grin, 1985) was used to measure overall satisfaction with life. This scale contains 5-items that are used to assess participants overall judgment a bout their life satisfaction. Particip ants rate their agreement with items on a 7-point Likert scale. Scores can rang e from 5 to 35, with higher scores indicating

PAGE 43

43 greater life satisfaction. A 2month test-retest correlation coefficient of .82 has been found (Diener et al., 1985). Reliability estimates have ranged from .78 to .95 (Diener et al., 1985; Vassar, 2008). Positive correlati ons have been found between the SWLS and other measures of subjective well-being, as well as negative corr elations with measures of psychopathology (Diener et al., 1985). Methods An online survey developm en t program, Survey Monkey ( www.surveymonkey.com ), was used to develop an online survey for our study. T he survey contained the demographic questions, items concerning social networking site use, and the measures of PIU, social anxiety, belonging, and well-being. Studies have show n that converting paper-based in struments to In ternet-based measures does not distort the valid ity, reliability, or factor struct ure of the instruments (Yu & Yu, 2007). In addition to the measures, an Institutio nal Review Board (IRB) approved consent form was included in the online survey, and participants could not view the rest of the survey without giving their consent to vo luntarily participate. To recruit participants, a link to the online su rvey was given to professors and graduate student instructors in the Depart ment of Psychology at the Univers ity of Florida. The professors and graduate students were aske d to distribute the link to their undergraduate classes through class listserves or in-class announc ements. Four graduate students and two professors agreed to allow their students to participate. Participation in the study was on a comple tely voluntary basis. All instructors used the online su rvey as an opportunity for extra credit for their students. An alternative extra credit option was also offered to the students so as to not violate ethical recruitment practices.

PAGE 44

44 CHAPTER 4 RESULTS Social Networking Site Use In our study, 95% (N = 350) of participants identified them selv es as social networking site users. Among these users, a very large major ity are members of the social networking site Facebook (97.8%). Members of MySpace made up 56% of the sample and members of Friendster made up less than 2% of the sample. Le ss than 1% of the sample were members of a social networking site other than the three listed. The site names given were LiveJournal (N = 3), Xanga (N = 1), Buzznet (N = 1), and HI5 (N = 1). It was also found that the majority of social networking site users have been members for ove r a year (92.5%). Only 2.2% of the sample reported being members for less than 6 months. As to the primary purpose for using these sites, most students acknowledged they primarily use the si te to talk to existing friends already in their offline social network (76.6%). Approximately 22% reported that they primarily use the site for both talking to existing friends and to make new friends. Less th an 1% of the sample (N=3) indicated that they use the site primarily to ma ke new friends or meet new people. Time spent on these social networking sites was also examine d. Approximately 69% of social networking site users indicated that they use a so cial networking site every day. Th e majority check the site 1 to 2 times per day (40.7%), though approximately 13% of the sample reported using the site 10 or more times a day. Most users spend about 10 to 15 minutes on the site (43.2%), while 20% report using the site for 30 minutes or more each time they log on. Two participants stated that they never turn it off and constantly check the site. Examination of the Modified GPIUS Because the GPIUS m easure was modified in our study to assess problematic social networking site use, rather than problematic Intern et use in general, reliability and validity were

PAGE 45

45 explored for the measure in our study. A reliabi lity analysis of the modified GPIUS measure revealed an internal consistency of 0.92, indica ting that with this sample, the items in the measure are highly intercorrelate d. A possible conclusion is that the items are related by the common latent construct of problema tic social networking site use. To provide an indication of validity, GPIU S scores were examined in relation to participants responses to the questions: (a) Has anyone ever told you that your use of online social networking sites like MySp ace or Facebook is problematic or interferes with your life (daily functioning)?, and (b) Do you feel that your use of online social ne tworking sites like MySpace or Facebook is problematic or interferes with your life (daily functioning?) If the measure indeed assesses problematic social netw orking site use, it is e xpected that those who answered yes to either of these questions would have higher scores on the GPIUS than those who answered no Analyses were conducted to compar e the GPIUS means for the group who answered yes for either question to the group who answered no On average, those who answered yes to the first question ( M = 76.31, SD = 13.12) had higher GPIUS scores than those who answered no ( M = 62.32, SD = 15.05). This differen ce was significant (t = 5.42, p < 0.001), suggesting that individuals who have had a someone tell them that their use of social networking sites may be problematic have on average higher GPIUS scores than thos e individuals who have not experienced this. For the se cond question, those who answered yes ( M = 77.59, SD = 13.10) also had higher GPIUS scores than those who answered no ( M = 62.36, SD = 14.98). Again, this difference was significant (t = 5.62, p < 0.001), suggesting that indivi duals who feel that their own use of social networking sites may be problematic have on average higher GPIUS scores than those individuals who do not feel this way. These results indicate that higher GPIUS scores

PAGE 46

46 may be suggestive of a problematic relationship w ith social networking si tes, and therefore lend validity to the measure utilized in our study. Prediction 1: Prevalence of Problematic Soc ial Networking Site Use The first aim of our study was to explore whet her a problematic relationship with the use of social networking sites like MySpace and Facebook exists in a college undergraduate population. It was hypothesized that a range of scores would be found for the measure of problematic social networking site use, including scores indicati ng a high presence of symptoms of problematic use. To explore this prediction, the scores of the modified GPIUS were examined. Scores may range from 29 to 145. For users of so cial networking sites in this sample, GPIUS scores ranged from 29 to 128, with a mean of 63.79 (SD = 15.45). Scores were normally distributed. Refer to Table 4-1 fo r a summary of the score distri bution. A definable cut-off score which would indicate problematic versus nonproblematic use was not developed for this measure so it is impossible to give a definitiv e classification of problematic users versus nonproblematic users. However, par ticipants showed a range of sc ores on this measure, with a number of participants scori ng high. Higher scores indicate more symptoms of problematic social networking site use. Some indication of prevalence of problematic use may be gained by considering that answering neutra lly to every item would give a sc ore of 87. In this sample, 5.7% of social networking site users had scores hi gher than an 87. Therefore, this portion of the participants endorsed at least some of the sympto ms of problematic social networking site use. A series of post-hoc analyses were conducted to de termine if demographic variables varied for the group with GPIUS scores above 87 compared to the group with GPIUS scores below 87. The analyses found no significant difference betw een the high group and the lower group with respect to gender, race/ethni city, and student status.

PAGE 47

47 Table 4-1. Distributi on of GPIUS Scores Score Range Percentage Cumulative Percentage 29-38 6.3% 6.3% 39-48 10.0% 16.3% 49-58 19.4% 35.7% 59-68 24.8% 60.5% 69-78 23.6% 84.1% 79-88 10.8% 94.9% 89-98 3.4% 98.3% 99-108 1.1% 99.4% 109-118 0.3% 99.7% 119-128 0.3% 100.0%

PAGE 48

48 Providing further evidence of the existence of problematic use of social networking sites, 10.6% of the sample indicated that they had had someone tell them that their use of social networking sites like Facebook or MySpace was probl ematic or interfered with their life. Approximately 9% admitted that they felt their ow n site use was problematic or interfered with their daily functioning. Taken together, these resu lts provide tentative evidence for the existence of problematic social networking site use, and therefore, support for prediction 1. Prediction 2: Predictors of Proble matic Soc ial Networking Site Use The second prediction of our study hypothesized that the degree to which a person belongs to an offline social network mediates the rela tionship between social anxiety and problematic social networking site use. To te st this hypothesis, a series of regression analyses are suggested by Baron and Kenny (1986). This is the most co mmon method of mediatio n testing used in research (Fritz & MacKinnon, 2007). Baron and Kenny (1986) state that a vari able functions as a mediator if it accounts for the relationship betwee n the predictor and outcome variable. In other words, a mediator helps to explain why a relatio nship between two variable s exists. To determine if a variable is indeed a medi ator, Baron and Kenny propose that th ree guidelines must be met. First, the predictor variable must account for variations in the outco me variable. Second, the mediator also must account for variations in the outcome variable. Finally, when the mediator is controlled for, the relationship between the predictor and the out come variable drops significantly in the case of pa rtial mediation, or becomes non-significant in the case of full mediation. In our study, it was hypothesized that the degree to which a person belongs to an offline social network mediates the relationship between social a nxiety and problematic social networking site use. Three sub-predictions were made in addition to the mediation prediction in order to meet Baron and Kennys (1986) guidelines for mediation testing. These were as follows:

PAGE 49

49 (1) the degree of social anxiety is positively relate d to problematic use of social networking sites, (2) the degree to which a person belongs to an offline social network is negatively related to problematic social networking site use, and (3) the degree of soci al anxiety is negatively related to the degree to which a person belongs to an o ffline social network. To determine if these subpredictions would be supported, the correlation coefficients relating these variables were examined. Social anxiety was indicated by S AD scores (M = 6.55, SD = 5.77). Belonging was indicated by ISEL scores (M = 9.48, SD = 2.13). GPIUS scores i ndicated problematic social networking site use. The score distributions fo r both the SAD and ISEL measures were highly skewed and therefore before analyses could be conducted, square root transformations were performed to correct for non-normality. After performing the transformations, skewness and kurtosis for both measures were within acceptabl e ranges (Tabachnick & Fidell, 1996). Pearson product-moment correlations were then examined fo r the three variables. Table 4-2 displays the results. Two of the sub-predictions were supported. Social anxiety was found to be positively related to problematic soci al networking site use ( r = 0.218, p < 0.001), as well as negatively related to the degree to wh ich a person belongs to an offline social network ( r = -0.377, p < 0.001). However, one sub-prediction was not suppor ted. The degree to which a person belongs to an offline social network, as indicated by ISEL scores, was not significantly related to problematic social networking site use, as indicated by GPIUS sc ores. Because of this, running a series of regression analyses to test the mediation hypothesis woul d be inappropriate. Therefore, prediction 2 was not supported by the results of our study. The degree of belonging to an offline social network was not found to mediate the rela tionship between social anxiety and problematic social networking site use for this sample.

PAGE 50

50 Table 4-2. Correlations Among Problematic Social Networking Site Use (GPIUS scores), Social Anxiety (SAD scores), and Belonging (ISEL scores) GPIUS SAD ISEL GPIUS --0.218* -0.042 SAD 0.218* ---0.377* ISEL -0.042 -0.377* --* p < 0.001

PAGE 51

51 However, it was interesting to note that the well-being variable of loneliness, another variable included in our study that is closely related to bel onging, did show a significant relationship to problematic so cial networking site use ( r = 0.281, p < 0.001), as well as to social anxiety ( r = 0.564, p < 0.001). Though the constructs of belo nging and loneliness do differ, they are both associated with inte rpersonal relatedness (Hagerty, Williams, Coyne, & Early, 1996). These two concepts have been shown to be highly related in the literature (Hagerty et al., 1992), and in our study, sense of belonging was si gnificantly related with loneliness ( r = -.481, p < .001). In addition, research in the area of general PIU has suppor ted a significant relationship between loneliness and PIU (Caplan, 2003; Mo rahan-Martin & Schumacher, 2000; Nalwa & Anand, 2003). Given the non-existence of research in the area of problema tic social networking site use, we felt it appropriate to explore this other variable of interpersonal relatedness loneliness as a mediator of the relationship be tween social anxiety a nd problematic use, though a prediction about this relationship was not included in the original hypotheses. It may be that the degree of interpersonal relatedness does indeed mediate the relationship between social anxiety and problematic social networking site use, but the construct of loneliness is a more appropriate representation of this relatedness than the construct of belonging for this sample of college students. A mediation analysis was conducted to determ ine if loneliness could be considered a mediator of the relationship betw een social anxiety and problematic social networking site use. In this newly hypothesized mediation model (s ee Figure 4-1), social anxiety (SAD score) functions as the predictor vari able, loneliness (UCLA scores) as the mediator, and problematic social networking site use (GPIUS score) as the outcome variable. The mediation guidelines proposed by Baron and Kenny (1986) were followed to test this model.

PAGE 52

52 Figure 4-1. Mediation model. Values refl ect standardized coefficients. *p < 0.001 LONELINESS (UCLA Scores) PROBLEMATIC SNS SITE USE (GPIUS Scores) SOCIAL ANXIETY (SAD Scores) a = .564* b = .281* c = .087, ns c = .218*

PAGE 53

53 To establish a relationship between the predic tor, social anxiety (SAD scores), and the outcome variable, problematic so cial networking site use (GPIUS scores), a simple regression was conducted (path c in figure 4-1). It was found that so cial anxiety significantly predicted ( = .218, p < .001) problematic social networking site use and explained approximately 5% of the variance ( R2 = .048, F (1, 348) = 17.398, p < .001) in GPIUS scores. Given this result, Baron and Kennys (1986) first guideline for mediation was met. A second regression was run to establish a relationship between the mediator of loneliness and the outcome variable of problema tic social networking site use (path b in figure 4-1). UCLA scores were entered as the inde pendent variable and GPIUS scor es as the dependent variable. Loneliness was found to be a significant predictor ( = .281, p < .001) of problematic social networking site use and accounted for approximately 8% of the variance ( R2 = .079, F (1, 348) = 29.921, p < .001) in GPIUS scores. Therefore, Ba ron and Kennys (1986) second guideline for mediation was met. Although not required by Baron and Kenny ( 1986), a third regression was conducted to establish a relationship between the predictor, so cial anxiety, and the me diator, loneliness (path a in figure 4-1). SAD scores were entered as the independent variable and UCLA scores as the dependent variable. Social anxiety was found to be a signif icant predictor ( = .564, p < .001) of loneliness and accounted for approximately 32% of the variance ( R2 = .318, F (1, 348) = 162.340, p < .001) in UCLA scores. Finally, to satisfy the third guideline for mediation, a multiple regression was conducted with SAD and UCLA scores as the independent variables and GPIUS scores as the dependent variable. This was done in order to control for the effect of the medi ator, loneliness, on the relationship between social an xiety and problematic social networking site use (path c in figure

PAGE 54

54 4-1). A concern with multiple regression analysis is multicollinearity. Therefore, a check of collinearity diagnostics (i.e., vari ance inflation factor (VIF) and tolerance) was included in the multiple regression. VIFs greater than 10.0 and tole rance values less than .02 raise concerns for multicollinearity (Bowerman & OConnell, 1990; Menard, 1995). The VIF (i.e., 1.466) and tolerance (i.e., .682) values were within the acce ptable range for the mu ltiple regression in our study. The result of the multiple regression indicated th at together the predictors explained almost 9% of the variance (R2 = .084, F (1,348) = 15.986, p < .001) in GPIUS scores. According to Baron and Kenny (1986), if the previously signifi cant relationship between the predictor and the outcome is no longer significant or has been reduced when the effect of the mediator is controlled for, this indicates full (in the case of the former) or pa rtial (in the case of the latter) mediation. The results seem to support full media tion in this model. Examining the standardized regression coefficients, the relationship between the mediator of lone liness and the outcome variable of problematic soci al networking site use (path b) remains significant ( = .232, p < .001), whereas the relationship between the predictor of social anxiety and the outcome variable of problematic social networking site use (path c now c ) is now non-significant ( = .087, p = .162). This result fulfills Baron and Kennys final guideline for proving mediation. The findings lend evidence for a model of loneliness as a medi ator of the relationship between social anxiety and problematic social ne tworking site use. Although Baron and Kenny (1986) di d not require any additional steps for establishing mediation other than those outlined in this paper, many researchers (Frazier, Mortensen, & Steward, 2005) advocate the use of Sobels test (Sobel, 1982) to dete rmine if the reduction in the relationship between the predictor and the outcome variable is sign ificant when the mediator is

PAGE 55

55 controlled for. Sobels test divi des the mediated effect by a sta ndard error term defined by Sobel to yield a z score. The z score is then compared to 1.96. If th e score is greater, the reduction in the relationship between the pred ictor and outcome variable is significant at the .05 level and mediation is indicated. When Sobels test was a pplied to the model in our study, the reduction in the relationship between social anxiety and problematic social networking site use while controlling for loneliness was significant ( z -value = 5.019, p < 0.001), indi cating that loneliness is a mediator of the relationship between social anxiety and problematic social networking site use. Though our study did not find support for belonging as a mediator (prediction 2), it was able to provide evidence that another indicato r of interpersonal rela tedness loneliness contributes to the relationship between social anxiety and problem atic use of social networking sites. Prediction 3: Problematic Social Netw orking Site Use & Well-Being The third main aim of our study was to exam ine the relationship be tween the problematic use of social networking sites and well-being. Specifically, a significan t negative relationship was predicted between problematic social networ king site use and the well-being indicators of self-esteem, happiness, and satisfaction with life, and a significant positive relationship was expected between problematic so cial networking site use and the well-being indicators of depression and loneliness. Means and standard deviations for the well-being measures are reported in Table 4-3. All five well-being variables were highly skewed and therefore before analyses could be conducted, square root tran sformations were performed to correct for nonnormality. After performing the transformation, sk ewness and kurtosis for all measures were within acceptable ranges (Tabac hnick & Fidell, 1996). Pearson product-moment correlations were then calculated utilizing GPIUS scores and the scores from the measures assessing the

PAGE 56

56 Table 4-3. Means and Standard Deviations of Well-Being Variab les for Social Networking Site Users Well-Being Variable Mean Standard Deviation Satisfaction With Life (SWL) 25.37 5.50 Loneliness (UCLA) 39.58 8.96 Depression (CES-D) 12.91 9.18 Self-Esteem (RSE) 21.84 4.84 Happiness (OHQ) 128.99 19.20 Social Anxiety (SAD) 10.70 6.31

PAGE 57

57 well-being variables. Correlation coefficients are reported in Table 44. One-tailed significance tests were utilized due to the prediction of relationship dir ections. The results fully support prediction 3 of our study. GPIUS scores were found to be significantly correlated with the variables of well-being in the expected directions. A higher degree of problematic social networking site use (GPIUS scor es) was found to be related to lower well-being, as indicated by greater levels of loneliness and depression, and lower levels of happine ss, self-esteem, and satisfaction with life. All correlation coefficients indicated a moderate e ffect size (defined as r = .30; Cohen, 1977) for the relationship of the we ll-being variables to problematic social networking site use, with the exception of the coefficient for satisfaction with life which indicated a small effect size (defined as r = .10; Cohen, 1977). A series of post-hoc analyses were conduc ted to explore how the mean scores on the well-being variables for the group considered pr oblematic social networking site users (i.e., individuals with GPIUS scores greater than 87) compared to the group with GPIUS scores less than 87. The means and standard deviations of the well-being vari ables for the group of problematic social networking si te users and non-problematic us ers can be found in Table 4-5. Greater well-being is indicat ed by higher scores on the satisfaction with life self-esteem and happiness variables, as well as lower scores on the depression loneliness and social anxiety variables. Problematic users si gnificantly differed from non-problematic users on the well-being variables of loneliness (t =4.20, p < 0.001), depression (t = 3.70, p < 0.001), self-esteem (t = 2.41, p < 0.05), happiness (t = -2.82, p < 0.01), and social anxiety (t = 3.27, p = 0.001), indicating that problematic users have lower levels of wellbeing than non-problematic users. The two groups did not significan tly differ on the variable of satisfaction with life

PAGE 58

58 Table 4-4. Correlations Between Well-Being Variables and Problem atic Social Networking Site Use (GPIUS Scores) Well-Being Variable Problematic Social Networking Site Use (GPIUS scores) Satisfaction With Life (SWL) -0.134* Loneliness (UCLA) 0.281** Depression (CES-D) 0.281** Self-Esteem (RSE) -0.299** Happiness (OHQ) -0.282** p < 0.01 ** p < 0.001

PAGE 59

59 Table 4-5. Means and Standard Deviations of Well-Being Vari ables for Problematic Social Networking Site Users (GPIUS > 87) and Non-Problematic Users (GPIUS < 88) Well-Being Variable Problematic User Nonp roblematic User Satisfaction With Life (SWL) 23.68 (SD 5.15) 25.47 (SD 5.51) Happiness (OHQ) 117.04 (SD 19.53) 129.68 (SD 18.99) Self-Esteem (RSE) 19.26 (SD 4.50) 21.99 (SD 4.82) Loneliness (UCLA) 47.79 (SD 7.89) 39.11 (SD 8.80) Depression (CES-D) 20.36 (SD 10.32) 12.48 (SD 8.94) Social Anxiety (SAD) 10.70 (SD 6.37) 6.31 (SD 6.65)

PAGE 60

60 CHAPTER 5 DISCUSSION Our studys overall aim was to explore the t opic of college student s usage of social networking sites and how site use may become pr oblematic or addictive. The predictions put forth in our study were primarily based on the lim ited research that has been conducted in the area of general problematic Internet use or Internet addiction. It was the go al of our study to add to this existing literature by examining whether a dependency on a specific Internet application social networking sites could exist. In addition, possible pred ictors of problematic social networking site use were explore d, as well as the relationship of problematic use with well-being for college students. Review of Study Findings Social Networking Site Use The percentage of participants that identif ied themselves as so cial networking site users in our study came as no surprise given the popularity of these sites that has been reported in the media (Owyang, 2008). Our study confirmed that social networking si tes are a prominent part of social communication among college students. Additi onally, given that a large majority of study participants indicated that they have been social networking site users for over a year, it is clear that these sites have embedded themselves in th e university culture. They have also become a part of the college students da ily life, as a majority of res pondents use one or more social networking sites every day, at least one to two times per day, for about 10 to 15 minutes at a time. Also, the recent surge in popularity of the site Facebook over the site MySpace was supported by our study results, with almost all of the study participants listing themselves as Facebook members, while only half acknowledged a membership in MySpace. Taken together,

PAGE 61

61 these results certainly support the idea that this generation of youths is a wired one and that the term Net Generation or MySpace Generati on (Hempel, 2005) is well earned. Another interesting finding was th at the majority of study partic ipants reported using social networking sites for staying in touch with existing friends already in their offline social network. This finding adds to the Inte rnet literature which already has suggested the communicative applications of the Internet ar e used more for maintaining rela tionships rather than meeting strangers. In a survey of United States Internet users, DAmico (1998) reported that 87% of the 1,001 respondents indicated that they use the Internet frequently to keep in touch with friends and family. College students in particular spend time talking to friends and family online. Scherer (1997) found that 98% of college student part icipants in his study reported using the Internet for maintaining relati onships with family and friends Additionally, Gross (2004) found that the majority of adolescent In ternet users used the communicati on functions of the Internet to talk with already existing, offline friends rather than strangers. Specifi c to social networking sites, Ellison and colleagues (2007) reported th at the Facebook users in their study indicated using the social networking site for communicating with people with whom they share an offline connection significantly more than for meeting new people. Measure of Problematic Social Networking Site Use Our study also m ade a contribution towards th e measurement of the problematic use of a specific Internet application (i.e ., social networking sites), rather than problematic use of the Internet in general. The modifi ed measure utilized in our stud y to assess problematic social networking site use was found to ha ve high internal consistency. Th e validity of the measure was indicated by showing that those who felt they had a problematic relationship with social networking sites or who had been told that th ey may have a problematic relationship showed significantly higher scores on the modified measur e than those who did not feel or who had not

PAGE 62

62 been told they had a problematic relationshi p. In addition, the meas ures correlation with constructs such as social anxiety and well-bei ng measures in the expected directions (Caplan, 2002, 2007; Morahan-Martin & Schumacher, 2000; Nalwa & Anand, 2003; Yen et al., 2007; Young & Rogers, 1998) also provided evidence for the validity of this measure. To our knowledge, no other study has examined problema tic use of social networking sites, and certainly a measure to assess it has not been cr eated previously. Therefore, the modification of the GPIUS to create an assessment of problematic social networking site use in our study is novel and could potentially contribute to future assessment in this area. Prediction 1: Existence of Proble matic Soc ial Networking Site Use The first and most important aim of our study was to explore whether a problematic relationship with social networki ng sites may exist in a population of college students. The range of scores found on the GPIUS measure support the notion that some students are experiencing a problematic relationship with social networking sites and encourages further exploration of this possibility. Students GPIUS scores indicated that some individua ls are using social networking sites to change their affective states and to gain greater social benefits and social control than they receive in face-to -face relationships. In addition, some students have a difficult time stopping their social networking si te use and keeping themselves from using these sites, while also utilizing the sites for greater amounts of tim e than planned. Finally, students scores also indicated that for some, these sites interfere wi th personal, social, and professional (academic) areas of their lives. Other support for the existenc e of problematic social networking site use was also found through two direct questions about problematic use; some study participants acknowledged that they felt or had been told that their use of social networking sites was problematic.

PAGE 63

63 Given that the measure used to assess problem atic use (i.e., the GPIUS) did not have a definitive cut-off point which would differentiate problematic users from non-users, as well as the restricted range of GPIUS scores and a low prevalence rate found in our study, it is difficult to say with certainty that probl ematic social networking site use is an identified phenomenon. A comparison of the range of GPIUS scores for this sample with findings from other studies using this measure (Caplan, 2002; 2003; 2007) are precl uded by the fact that th e other studies have utilized individual subscales of the GPIUS rather than a sum score, as in our study. However, like our study, Caplan showed in his study a restricted range in PIU scores and a low prevalence of general PIU in the undergraduates he sample d. To give some indicator of prevalence, he showed that the percentage of students endorsi ng symptoms of PIU ranged from 5-15% of the sample, depending on the individua l subscale (Caplan, 2006). In our study, 5-10% of the sample could be considered problematic social networki ng site users, taking into account both GPIUS scores and responses to the two direct questions about problematic use. This prevalence rate is comparable to other prevalence rates found for general PIU in the literature. Utilizing the Pathological Internet Use Scale (PIUS; Morahan-Martin & Schu macher, 2000), Morahan-Martin and Schumacher classified 8.1% of their undergradua te sample as pathological Internet users, while Niemz and colleagues (2005), utilizing this same measure, found 18.3% of the undergraduates surveyed to be pathological Internet users. Sche rer (1997) found 13% of undergraduates sampled in his study to be Intern et dependents, as classified using criteria paralleling chemical dependencies. Finally, Gree nfield (1999) utilized an online survey of Internet use and behavior to explore a general population. He found that of the 17,251 individuals (aged 8-85 yrs) survey ed, approximately 6% met the crit eria for Internet addiction.

PAGE 64

64 Though some of these prevalence rates are higher than the rate found in our study, the intended purpose of the assessment instruments ut ilized in these studies should be considered when comparing these prevalence findings. Bear d (2005) highlights that assessment instruments of general problematic Internet use or Internet addiction do not take into account the different types of Internet applications that a person could be depende nt upon. Therefore, caution should be used when comparing findings from our study assessing a specific type of Internet dependency to prevalence rates indicated for ge neral Internet dependency. For example, the instruments used in these studies of general PIU could be tapp ing into a dependency upon online gaming or online gambling, which may be more prevalent than say a dependency upon social networking sites. In addition, many of the assessment instruments assume that excessive Internet use denotes problematic Internet use, and do not take into account that elevated Internet usage may be due to academic or work requirements (Widyanto & Griffiths, 2005). Other studies have cited the cut-off point of many of the instruments used to assess Internet dependency as being too liberal, resulting in inflated preval ence estimates (Niemz et al., 2005). What can be definitively concl uded from our study is that overall, the population of college students sampled might best be characterized as a non-pathological sample with largely normative levels of social networking site use. Nevertheless, social netw orking site use could be characterized as problematic for perhaps 5-10% of the sample. This prevalence rate is on par with other rates of mental heal th disorders found in the DSM-IV (APA, 2000), as pointed out by Watson (2005). Additionally, Hall and Parsons (2001) noting the discrepancies in the Internet addiction prevalence rates found in the literature, concluded that even a c onservative estimate of 6% is worrisome and warrants fu rther research. With social ne tworking sites like MySpace and Facebook becoming some of the most-trafficked webs ites in the world, even the possibility that

PAGE 65

65 the use of these types of sites may become problem atic for some individuals is cause for concern. The findings of our study, then, encourage further investigation of problem atic social networking site use. Evidence found by our study in support of the existence of problematic social networking site use is in line with research that has impli cated the communication functi ons of the Internet as commonly used applications by general Interne t addicts (Chou et al., 1999; Chou & Hsiao, 2000; Young, 1996a). Our studys findings could also be considered supportiv e of the literature that suggests the classification of Internet a ddiction in the next ve rsion of the DSM (APA, 1995). Proponents of the addition of Internet addiction suggest that individuals can become dependent on the environment of the Internet, and not simply to an application on the Internet which provides another way of engaging in an addi ction that already exists offline (Yellowlees & Marks, 2007). Though our study found support fo r a dependency on a specific Internet application (i.e., social networki ng sites), the service this appli cation provides is not available offline, like gambling or shopping may be. In ot her words, social networking sites provide a social environment unique to the Internet. It is a social atmosphere in which the normal auditory and visual cues that accompany offline co mmunication are missing (Morahan-Martin & Schumacher, 2000), creating a less-t hreatening environment for social interaction. In this case, the Internet is not providing the users with a way of engaging in an addiction they would already have offline, such as a gambling addict usi ng an online gaming casino instead of an offline casino. Rather, social networking sites provide an environment in which a person can socially interact in a less-threatening atmosphere that cannot be gained off line (Morahan-Martin & Schumacher, 2000; Peter et al., 2005). In the debate over classification of Internet addiction, our study seems to fall on the side which supports that dependency on the unique environment of the

PAGE 66

66 Internet itself can occur, as opposed to the In ternet providing another forum for engaging in a pre-existing addiction. Prediction 2: Predictors of Proble matic Soc ial Networking Site Use The second aim of our study was to examine how the concepts of social anxiety and belonging to an offline social ne twork may contribute to the deve lopment of problematic social networking site use. Specifical ly, belonging to an offline soci al network was explored as a possible mediator of the relationship between so cial anxiety and problem atic social networking site use. Within this mediation model, support was required for a number of sub-predictions in order to test the model. Most importantly, evidence for a significant positive relationship between the predictor variable, social anxiety, and the outcome variable, problematic social networking site use, was found. High er levels of social anxiety were associated with greater levels of problematic social networking site use in this sample of colle ge students. Individuals who suffer anxiety in offline soci al environments may be more likely to develop a problematic relationship with the us e of social networking sites than those who do not suffer from social anxiety. This finding is in line with the etio logy of PIU development suggested by Caplan (2007), who proposed that socially anxious people prefer situations whic h minimize their social risk and the Internet can provide this less-th reatening social atmosphere. The maladaptive cognitions associated with soci al anxiety result in a prefer ence for online communication, and the social interaction gained by the individua l in a less-threatening way reinforces the individuals dependence on this online environment for social in teraction. PIU can then develop, or in this case, problematic social networking site use. Belonging was chosen as a possi ble mediator of this relatio nship between social anxiety and problematic use because inherent in this relationship is the a ssumption that social networking sites have become reinfo rcing for the socially anxious indi vidual due to a need that is

PAGE 67

67 being fulfilled. Socially anxious individuals have been shown to have fewer friendships, and therefore a need to belong ex ists (Gambrill, 1996; La Greca & L opez, 1998). Social networking sites were thought to be able to fulfill this need by provi ding a less-threatening social environment, and that PIU could develop due to the need fulfillment provided by these sites (Chou et al., 1999; Suler, 1999). However, our study failed to find support for belonging as a mediator, as it was not found to be significantly rela ted to problematic social networking site use. Failure to find evidence for this relationship ma y be explained by the fact that most of the study participants indicated they use social netw orking sites more for staying in contact with existing friends in their social network rather than meeting strangers and making new friends. The prediction of belonging as a mediator rested upon an assumption that socially anxious individuals were using these site s to make new friends due to a lack of belonging to a current offline network. However, it seems that these sites are structured in such a way as to entice those with pre-existing social networks to use the sites for maintaining friendships or meeting friends of friends (Boyd & Ellison, 2007), rather than as an avenue for meeting strangers and making new friends. Therefore, it may not be a lack of cu rrent friendships that is the driving need to use social networking sites. However, another vari able of interpersonal relatedness may be the answer. Despite the failure to find support for belonging as a mediator, another indicator of interpersonal relatedness loneliness seems to have potential. Though belonging and loneliness are both indicators of interpersonal relatedness, they differ in their meanings (Hagerty et al., 1992). In our study, be longing, as measured by the ISEL was characterized as the perceived availability of friends which one can spend time with. This type of belonging is observable, as it can be behavi orally detected via group membership. Loneliness, on the other

PAGE 68

68 hand, is a negative emotional or affective stat e in which a need for personal connectedness exists, often due to a disruption that has occurred which resulted in an absence or loss (Hagerty et al., 1992). A lack of belonging, therefore, is a physical lack of relationships whereas loneliness is the feeling of non-intimacy or connectedness. Research has supported the idea that loneliness can occur when one is dissatisfied with existi ng relationships (Cutrona Russell, & Peplau, 1979) rather than simply from a lack of belonging. Find ings indicate that loneli ness is more related to the quality of ones social connections as opposed to the quantity of connections (Jackson, Soderlind, & Weiss, 2008). Loneliness is decreased when a person feels intimacy and closeness in their relationships (Jones, Carpenter, & Quin tana, 1985) and feels supported by others (Pierce, Sarason, & Sarason, 1991). Hagerty and colleagues illustrate with a cas e vignette how one can have personal relationships (b elong) but still feel lonely. Marie moved to a town where she knew no one. The first several weeks she spent much of her time unpacking and finding her way around. She did spend time with her new neighbors and coworkers, all of whom seemed to really enjoy her company. Yet Marie felt very alone, unhappy, and isolated wi thout her friends and family. This case mirrors the transition that many college students must make from their hometown high school where they may have had a number of friends and family to a university where they may know no one. Loneliness is quite prevalent among co llege students despite the great number of social opportunities for belongi ng that exist in a college environment (Jones, Freemon, & Goswick, 1981). In our study, loneliness was found to be significantly related to social anxiety and problematic social networking site use. The link between social anxiety and loneliness has been supported in the litera ture (Inderbitzen-Pisaruk, Clark, & Solano, 1992) and research suggests that socially anxious individua ls perceive less intimacy and supportiveness in their existing friendships (La Greca & Lopez, 1998). In addi tion, a link between loneliness and general

PAGE 69

69 problematic Internet use has been found (Caplan, 2003; Moraha n-Martin & Schumacher, 2000; Nalwa & Anand, 2003). Though a prediction regardi ng a mediation model positing loneliness as the mediator between social anxiety and problematic social networ king site use was not made at the start of our study, the failure of our study to find support for belonging as a mediator prompted the exploration of th is new mediation model. Our st udy found support for loneliness as a mediator. It may be that the need driving so cially anxious individua ls to utilize social networking sites is not a lack of friendships (belonging) but a l ack of intimacy in friendships (loneliness). Instead of using these sites to meet strangers and make friends, the sites are utilized for enhancing relationships with existing friends Socially anxious indi viduals who may feel loneliness and desire more closeness in their rela tionships may turn to so cial networking sites to fulfill this desire and relieve their feelings of loneliness by gaining more intimacy in their current offline relationships. This need fulfillment and reinforcement may then develop into problematic social networking site use, as suggested by Caplan (2007) and Suler (1999). Studies have shown that the Internet does indeed provide an opportunity to increase closeness in offline relationships. Valke nburg and Peter (2007) examined how online communication with existing friends affects the offline relationship with these same friends. Their stimulation hypothesis proposed that specific characteristics of the In ternet encourage selfdisclosure more readily in online relationships than in face-to-face relationships. Self-disclosure is important for building intimate relations hips (Collins & Miller, 1994), and therefore communicating with friends online could increas e the amount of self-disclosure, and hence intimacy, that takes place in the relationshi p. A number of studies have found support for increased self-disclosure in on line communication, not only between strangers, but also between existing friends (Grinter & Palen, 2002; Leung, 2002). Valkenburg and Pete r provided evidence

PAGE 70

70 for the stimulation hypothesis by collecting data from 794 adolescents. They found that for adolescents who use the Inte rnet to communicate with existing friends, more online communication resulted in more perceived intim acy in friendships even when offline. In conclusion, our study supports a model in whic h loneliness, or the need for closeness in relationships, mediates the re lationship between social anxi ety and problematic social networking site use. Support for a model positing belonging, or the availability of friends, as a mediator of this relationship was not found for this sample of college students. The findings indicate that socially anxious individuals may use social networking sites to fulfill a need for closeness in existing relationshi ps and that they may become dependent upon this online environment for this need fulfillment. Prediction 3: Problematic Social Netw orking Site Use and Well-Being The final aim of our study was to explore th e relationship between problematic social networking site use and well-being in college st udents. It was predicted that problematic use would be related to decreased well-being. Litera ture showing a relationship between general PIU and well-being was utilized as support for th is prediction (Caplan, 2002; Morahan-Martin & Schumacher, 2003; Young & Rogers, 1998). To represent well-being, measures of depression, loneliness, self-esteem, happiness, and satisfaction with life were used. All five of the well-being indicators showed a significant relationship with problematic social networking site use in the expected directions, and therefore prediction 3 was fully supported. Results indicated that individuals showing greater symptoms of pr oblematic use also had more symptoms of depression, loneliness, and unhappine ss, as well as having lower se lf-esteem and less satisfaction with their lives.

PAGE 71

71 Given the lack of an experimental design of our study, it is impossible to determine causality in the relationship between problema tic social networking si te use and well-being. Support can be found in the literature for both of these causal paths. First, it could be that this relationship exis ts because those sufferi ng from decreased wellbeing are more likely to devel op PIU. As Daviss (2001) model of PIU development explains, a pre-existing psychopathology or condition coul d make a person more vulnerable to PIU development should a stressor occur. Our study has shown support for a model which suggests social anxiety and the associated loneliness as contributors to the development of problematic social networking site use. Loneliness has been shown in the literature to be associated with decreased well-being (Weiss, 1974). A link has b een well-established between loneliness or a lack of closeness in relationships and the de velopment of depression (Russell, Cutrona, Rose, and Yurko, 1984). Loneliness has also been found to be related to low self-esteem (Leary, 1990) and unhappiness (Argyle, 1987; Freedman, 1978). Other evidence of the connection between closeness in relationships and we ll-being can be seen in the couns eling literature. Baumeister and Leary (1995) emphasize the contri bution of the therap eutic relationship in psychotherapeutic progress (Rogers, 1959) as support for the associa tion of relationship clos eness with well-being. These findings taken together suggest that the decreased well-being associated with loneliness could be an additional factor in the development of problematic social networking site use, and are indicative of a path in which decreased well-being leads to problematic use. However, this relationship could also exist because problematic use causes a negative impact on an individuals well-b eing. Support for this causal path can be found in Daviss (2001) model of general PIU development. He suggest s that maladaptive cognitions lead to PIU development and help maintain problematic use. These maladaptiv e cognitions include

PAGE 72

72 cognitions about the self which result in se lf-doubt, low self-efficacy, and negative selfappraisals (e.g., I am worthless offline, but onl ine I am somebody, I am a failure when I am offline). In addition, thoughts about the world involve the generali zation of specific events to global patterns or engagement in all-or-nothing type thinking (e.g., Nobody loves me offline, The Internet is the only place I am respected) These maladaptive cogni tions associated with PIU may lead to a decrease in self-esteem, ha ppiness, and satisfaction with life. Additionally, research has indicated a strong link between ma ladaptive cognitions and the development of depression (Beck, 1993). Finally, general PIU has been shown to interfere with normal daily functioning in many areas of an individuals li fe (Scherer, 1997; Young, 1998). Therefore, the interference in daily functioning that problematic social networking site use may cause could also lead to decreased well-being. Thus, these findings lend support to a causal path in which problematic social networking site use causes a decrease in well-being. Given that research findi ngs provide evidence for both causal pathways between problematic social networking site use and well-being, it is most likely that both of these paths are correct. Decreased well-being may make one susceptible to developing a problematic relationship with social networking sites, but the maladaptive cognitions that maintain problematic use may further decrease well-bei ng (Caplan, 2003; Davis, 2001). Though our study cannot draw conclusions either way about the ca usality in the relationship between well-being and problematic social networking site use, the establishment of a relationship between these two constructs, and thus support for prediction 3, is an ex cellent beginning in th e research of this novel area. Study Implications First, our study is im portant in its novel contribution to an ar ea with very limited research but an area with much importance. Given the extensive us e of social networking sites, especially

PAGE 73

73 on college campuses, the usage of these sites and th e impact they may have should be of interest to researchers and to universi ty counselors. Counselors s hould be aware of the potential problematic relationship or de pendency that college students could develop with social networking sites, and how this ma y interfere with personal, soci al, and academic functioning, as well as impact students mental health. In a survey of university counselors, Kiralla (2005) found that approximately 84% of counselors believed th at problematic Internet use is a legitimate concern for students, but 93% said they did not have sufficient training regarding the diagnosis or treatment of PIU. In addition to training coun selors on general PIU, counselors should also be made aware of the potentially addictive nature of social networking sites. Information also needs to be given to students as to how the use of the Internet and so cial networking sites can become problematic and interfere with their lives. Sche rer (1997) notes that many college students are not even aware that dependency or problematic usag e can develop with the Internet, and that it is the responsibility of counselors to make student s aware of the symptoms and available services. Many university counseling centers are now online, and given the centrality that the Internet plays in problematic users lives, this may be the perfect place to be gin the dissemination of important information regarding the symptoms and effects of general PIU and problematic social networking site use, as well as the available services for those who may struggle with these issues. Many of the PIU assessment instruments are already available as simple checklists (e.g., Youngs Internet Addiction DQ ) that could be put online for students to self-assess their PIU potential. Counselors can also benefit from our study by learning more about the attraction that social networking sites may hold fo r socially anxious individuals and how these sites can both be beneficial and detrimental for them. Though our study explores the detrimental side of site use,

PAGE 74

74 social networking sites may also provide an excellent tool for socia lly anxious individuals. Amichai-Hamburger and Furnham (2 007) note that socially inhibited people may benefit from the non-threatening environment the Internet allows for acquiring, practicing, and improving social skills if they can transfer these skills to offline social interactions. For some, the transfer of social skills acquired on the Internet to faceto-face relationships may be natural (McKenna, Greene, & Gleason, 2002), while for others with more severe social anxiety, this ability may be more limited. Amichai-Hamburger and Furnham propose that even very socially anxious individuals can learn to transfer online social skills to offline relationships through a series of steps that gradually build the individuals expo sure to face-to-face interactions via audio and video functions of the Internet. Therefore, social networking sites could be a valuable tool for counselors to use with socially anxious individu als if proper steps are ta ken to help socially anxious people avoid dependency on these site s for all social inter action and support. Limitations Our study has a num ber of limitations that should be noted. First, all of the measures used in our study are of a self-report nature. This al lows for underor over-reporting by students that may affect the reliability and validity of the study findings. Many of the constructs assessed in our study could be associated with shame, denial or minimization, and this could also lead to under-reporting (Block, 2008). Additi onally, due to the restricted range of GPIUS scores and a low prevalence rate, it cannot be concluded defi nitively that the problematic use of social networking sites exists, though our study certainly provides a starting point for further research in this area. Another limitati on is that due to the lack of an experimental design, causal inferences cannot be made. Generalizability of our study findings is also limited to undergraduate students at a four-year university, and the predominantly female composition of the sample also limits the generalizability. Fina lly, our study does not incl ude the numerous other

PAGE 75

75 variables that may contribute to the relationships examined and therefore only a limited picture of problematic social networking site use is gi ven. These limitations are reasonable given that little research about social netw orking sites has been conducted. The purpose of our study was to simply provide a foundation for the examinati on of social networking site use and the problematic relationship that can develop with the use of these si tes. Future research exploring this topic in much more detail will certainly be needed. Future Research These lim itations suggest a number of directions for future research on problematic social networking site use. First, a more accurate estima te of the prevalence of problematic use in the college population is most certa inly needed. This suggests that development and validation of assessment instruments measuring problematic use should continue, and a measure with a designated cut-off point identifying problematic users versus non-problematic users may be warranted. Examining the prevalence of problema tic social networking site use in populations other than college students, such as younger adolescents or adults may also yield different and interesting findings. Future stud ies could also benefit from an experimental design in which conclusions regarding causality in the relationships explored in our study could be made. Finally, the predictors of problematic use included in our study only accounted for a small portion of the variance in scores indicating problematic social networking site use, and therefore other predictors of problematic use should be explored. Establishing variables which are predictive of problematic use and differentiating these from variables that are the result of problematic use could help guide practitioners in di agnosing and treating this issue. These are just a small sampling of future direc tions for research in th e area of problematic social networking site use. Given the non-existence of res earch in this area, mu ch research is still needed. The numerous implications of this type of research coupl ed with a trend showing that

PAGE 76

76 social networking sites are c ontinuing to grow in popularity (Owyang, 2008) suggest the importance of continued growth in our knowledg e about social networking site use and the impact these sites may have for users.

PAGE 77

77 APPENDIX A INTERPERSONAL SUPPORT EVALUATION LIST (ISEL) BEL ONGING SCALE This scale is made up of a list of statements each of which may or may not be true about you. Please check true if the statement applies to you most of the time or check false if the statement does not usually apply to you. 1. There are people at school or in town who I regularly run wi th, exercise with, play sports with, or do other enjoyable activities. 2. I hang out in a friend's room or apartment quite a lot. 3. I can find a person who I enjoy sp ending time with whenever I want. 4. If I decided at dinner time to take a st udy break this evening and go to a movie, I could easily find someone to go with me. 5. People hang out in my room or apartment during the day or in the evening. 6. I belong to a group at school or in town that meets regularly or does things together regularly. 7. I am not a member of any social groups (such as church groups, clubs, teams, etc.) 8. Lately, I often feel lonely, like I don't have anyone to reach out to. 9. I don't have friends at school or in town who would comfort me by showing some physical affection. 10. I don't often get invited to do things with other people. 11. I don't usually spend two evenings on the weekend doing something with others.

PAGE 78

78 APPENDIX B MODIFIED GENERALIZED PROBLEMAT IC INTERNET USE SCALE (GPIUS) Rate the extent to which you agree or disa gree w ith each statement about your use of online social networking sites. Social networking sites are websites on the Internet where you can create a profile and connect with friends. Examples of these sites are MySpace, Facebook, or Friendster. 1. I have used social networking sites to talk with others when I was feeling isolated. 2. I can control how others perceive me when I am using a social networking site. 3. I find it hard to stop thin king about what is waiting for me online on the social networking site(s) I like to use. 4. I have gone on a social networking site to make myself feel better when I was down or anxious. 5. I have tried to stop using one or more soci al networking sites for such long periods of time. 6. When not on a social networking site, I wonder what is happening on that site. 7. When I am on a social networking site, I socialize with people without worrying about relationship commitment. 8. I have attempted to spend less time on social networking sites but have not been able to. 9. I am treated better by others while online using social networking sites than I am offline. 10. I use social networking sites to make myself feel better when Im down. 11. I have missed class or work because of being online on a social networking site. 12. I feel worthless offline, but online on a social networking site I am someone. 13. I want to, or have made, unsuccessful efforts to cut down or control my use of one or more social networking sites. 14. I have missed a social event or social engagements because of being online on a social networking site. 15. I am more comfortable with computers than with people. 16. I feel guilty about the amount of time I spend on social networking sites.

PAGE 79

79 17. When I am on a social networking site, I socialize with other people without worrying about how I look. 18. I lose track of time when I am using a social networking site. 19. I miss being online on a social netw orking site if I cant get on it. 20. I am treated better in my online relationships th rough social networking sites than in my face-to-face relationships. 21. I have used a social networking si te for a longer time than I intended. 22. I am more confident socializing on soci al networking sites than I am offline. 23. I have used a social networking site for l onger periods of time than I had expected to. 24. I feel safer relating to people online on soci al networking sites rather than face-to-face. 25. I feel lost if cant get on a social networking site. 26. I have spent a good deal of time using social networking sites. 27. I have gotten into trouble with my employer or school because of being online on a social networking site. 28. I am preoccupied with thinking about social networking sites if I cant connect for some time. 29. I have sought others online on social netw orking sites when I was feeling isolated.

PAGE 80

80 APPENDIX C SOCIAL AVOIDANCE AND DIS TRESS SCALE (SAD) For the following items, think about how the statement applies to you in your offline social life, not how you are onl ine when using the Internet. Please check true if the statement applies to you most of the time or check false if the statement does not usually apply to you. 1. I feel relaxed even in unfamiliar situations. 2. I try to avoid situations whic h force me to be very sociable. 3. It is easy for me to relax when I am with strangers. 4. I do not have a particular desire to avoid people. 5. I often find social occasions upsetting. 6. I usually feel calm and comf ortable at social occasions. 7. I am usually at ease when talk ing to someone of the opposite sex. 8. I try to avoid talking to people unless I know them well. 9. If the chance comes to meet new people, I often take it. 10. I often feel nervous or te nse in casual get-togethers in which both sexes are present. 11. I am usually nervous with people unless I know them well. 12. I usually feel relaxed. 13. I often want to get away from people. 14. I usually feel uncomfortable when I am in a group of people I dont know. 15. I usually feel relaxed when I meet someone for the first time. 16. Being introduced to people makes me tense and nervous. 17. Even though a room is full of strangers, I may enter it anyway. 18. I would avoid walking up and joining a large gr oup of people. 19. When my superiors want to talk with me, I talk willingly. 20. I often feel on edge when I am with a group of people.

PAGE 81

81 21. I tend to withdraw from people. 22. I dont mind talking to people at parties or social situations. 23. I am seldom at ease in a large group of people. 24. I often think up excuses in orde r to avoid social arrangements. 25. I sometimes take the responsibility for introducing people to each other. 26. I try to avoid formal social occasions. 27. I usually go to whatever social engagements I have. 28. I find it easy to relax with other people.

PAGE 82

82 APPENDIX D OXFORD HAPPINESS QUESTIONNAIRE (OHQ) Please indicate how m uch you agree or disagree with the following statements. You will need to read the statements carefully because some are phrased positively and others negatively. Please give the answer that is true fo r you in general or most of the time. 1. I dont feel particularly pleased with the way I am. 2. I am intensely interested in other people. 3. I feel that life is very rewarding. 4. I have very warm feelings towards almost everyone. 5. I rarely wake up feeling rested. 6. I am not particularly optimistic about the future. 7. I find most things amusing. 8. I am always committed and involved. 9. Life is good. 10. I do not think that the world is a good place. 11. I laugh a lot. 12. I am well satisfied about everything in my life. 13. I dont think I look attractive. 14. There is a gap between what I w ould like to do and what I have done. 15. I am very happy. 16. I find beauty in some things. 17. I always have a cheerful effect on others. 18. I can fit in everything I want to. 19. I feel that I am not especially in control of my life. 20. I feel able to take anything on.

PAGE 83

83 21. I feel fully mentally alert. 22. I often experien ce joy and elation. 23. I do not find it easy to make decisions. 24. I do not have a particular sense of meaning and purpose in my life. 25. I feel I have a great deal of energy. 26. I usually have a good influence on events. 27. I do not have fun with other people. 28. I dont feel part icularly healthy. 29. I do not have particularly happy memories of the past.

PAGE 84

84 LIST OF REFERENCES Am erican Psychological Association. (1992). Ethi cal principles of psychologists and code of conduct. American Psychologist, 47, 1597-1611. American Psychiatri c Association (1995). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Publishing, Inc. Amichai-Hamburger, Y., & Furnham, A. (2007). The positive net. Computers In Human Behavior, 23 1033-1045. Anant, S. S. (1966). The need to belong. Canadas Mental Health, 14 21-27. Argyle, M. (1987). The psychology of happiness London: Methuen. Armstrong, L., Phillips, J.G., & Saling, L.L. (2000). Potential determinants of heavier internet usage. International Journal of HumanComputer Studies, 53 537. Baron, R. M., & Kenny, D. A. (1986). The moderato r-mediator variable di stinction in social psychological research: Con ceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51 1173-1182. Baumeister, R., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117 497-529. Beard, K. W. (2005). Internet addiction: A review of current asse ssment techniques and potential assessment questions. Cyberpsychology & Behavior, 8 (1), 7-14. Beard, K. W., & Wolf, E. M. (2001). Modi cation in the proposed diagno stic criteria for Internet addiction. Cyberpsychology and Behavior, 4 (3), 377. Beck, A. T. (1993). Cognitive therapy: Past, present, and future. Journal of Counseling and Clinical Psychology, 61 (2), 194-198. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory (2nd ed.). San Antonio, TX: The Psychological Corporation. Block, J. J. (2008). Issues fo r DSM-V: Internet Addiction. American Journal of Psychiatry, 165(3), 306-307. Bowerman, B. L., & OConnell, R. T. (1990). Linear statistical mode ls: An applied approach (2nd edition). Belmont, CA: Duxbury. Bowlby, J. (1980). Attachment and loss New York: Basic Books. Boyd, D. M., & Ellison, N. B. ( 2007). Social network sites: Defi nition, history, and scholarship. Journal of Computer-Mediated Communication 13(1). Retrieved July 20, 2008, from http://jcmc.indiana.edu/ vol13/issue1/boyd.ellison.htm l

PAGE 85

85 Brenner, V. (1997). Parameters of Inte rnet use, abuse and addiction: The rst 90 days of the Internet usage survey. Psychological Reports, 80 879. Caplan, S. E. (2002). Problematic Internet use and psychosocial well-being: development of a theory-based cognitive-behavioral measure. Computers in Human Behavior, 18, 533. Caplan, S. E. (2003). Preference for online social interaction: a theory of problematic Internet use and psychosocial well-being. Communication Research 30, 625. Caplan, S. E., & High, A. C. (2006). Beyond excessive use: The interac tion between cognitive and behavioral symptoms of problematic Internet use. Communication Research Reports, 23(4), 265-271. Caplan, S. E. (2007). Relations am ong loneliness, social anxiety, and problematic Internet use. Cyberpsychology & Behavior, 10 (2), 234-242. Chak, K., & Leung, L. (2004). Shyness and locus of c ontrol as predictors of Internet addiction and Internet use. CyberPsychology & Behavior, 7 (5), 559-570. Chen, S. H., & Chou, C. (1999, August). Development of Chinese Internet addiction scale in Taiwan. Poster presented at the 107th Amer ican Psychology Annual convention, Boston, USA. Choi, Y. H. (2007). Advancement of IT and seriousne ss of youth Internet addiction International Symposium on the Counseling and Treatment of Youth Internet Addiction. Seoul, Korea. Chou, C., Chou, J., & Tyan, N. N. (1999). An expl oratory study of Internet addiction, usage and communication pleasureThe Taiwans case. International Journal of Educational Telecommunications, 5 (1), 47. Chou, C., Condron, L., & Belland, J. C. (2005). A revi ew of the research on Internet addiction. Educational Psychology Review, 17 (4), 363-388. Chou, C., & Hsiao, M. C. (2000). In ternet addiction, usage, grati cations, and pleasure experienceThe Taiwan college students case. Computers & Education, 35 (1), 65. Cohen, J. (1977). Statistical power analysis for the behavioral sciences (revised edition). New York: Academic Press. Cohen, S., & Hoberman, H.M. (1983). Positive ev ents and social supports as buffers of life change stress. Journal of Applied Social Psychology, 13 99-125. Collins, N. L., & Miller, L. C. (1994). Self-d isclosure and liking: A meta-analytic review. Psychological Bulletin, 116, 457 475. Connor, K. M., Davidson, J. R. T., Churchill, L. E., et al. (2000). Psycho metric properties of the Social Phobia Inventory (SPI N): New self-rating scale. British Journal of Psychiatry, 176 379-386.

PAGE 86

86 Coopersmith, S. (1991). Self-Esteem Inventories CA: Consulting Psycho logists Press, Inc. Cutrona, C., Russell, D., & Peplau, A. (1979). Loneliness and the process of social adjustment: A longitudinal study Paper presented at the annual meeting of the American Psychological Association, New York, NY. DAmico, M. L. (1998). Internet has become a necessity, U. S. poll shows Retrieved November 14, 2007, from www.cnn.com/TECH/computing/ 9812/07/neednet.idg/index.htm l Davis, R. A. (2001). A cognitive-behavioral model of pathological Internet use. Computers in Human Behavior, 17, 187. Diener, E., Emmons, R. A., Larsen, R. J., & Gri ffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71-75. Ellison, N. B., Steinfield, C., & Lampe, C. (2007) The benefits of Facebook "friends:" Social capital and college students' use of online social network sites. Journal of ComputerMediated Communication 12(4). Retrieved November 14, 2007, from http://jcmc.indiana.edu/vol12/issue4/ellison.htm l Frazier, P. A., Mortensen, H., & Steward, J. (2005). Coping strategies as mediators of the relations among perceived control and di stress in sexual assault survivors. Journal of Counseling Psychology, 52(3), 267-278. Freedman, J. (1978) Happy people: What happiness is, who has it, and why. New York: Harcourt Brace Jovanovich. Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect. Psychological Science, 18(3), 233-239. Gambrill, E. (1996). Loneliness, social isolation and social anxiety. In M. A. Mattaini & B. A. Thyer (Eds.), Finding solutions to social problems: Behavioral strategies for change (pp. 345-371). Washington, DC: American Psychological Association. George, A. (2006). Living online: The end of privacy? New Scientist 2569 Retrieved July 20, 2008, from http://www.newscientist.com/channel/tech/m g19125691.700-living-online-theend-of-privacy.html Goldsmith, R. E. (1986). Dimensionality of the Rosenberg Self-Esteem Scale. Journal of Social Behavior & Personality, 1 (2), 253-264. Greenfield, D. N. (1999). Psychological characteristics of compulsive Internet use: A preliminary analysis. Cyberpsyc hology & Behavior, 2(5), 403-412. Griffiths, M.D. (1998). Internet addiction: Does it really exist? In J. Gackenbach (Ed.), Psychology and the Internet: intrape rsonal, interpersonal and transpersonal applications (pp. 61) New York: Academic Press.

PAGE 87

87 Griffiths, M. D. (2000). Does Internet and computer addiction exist? Some case study evidence. Cyberpsychology & Behavior, 3 (2), 211. Griffiths, M. D. (2003). Internet abuse in the workplace: Issues and concerns for employers and employment counselors. Journal of Employment Counseling, 40 (2), 87-96. Grinter, R. E., & Palen, L. (2002, November). Instant messaging in teen life. Paper presented at the 2002 Association for Computing Machinery Conference on Computer Supported Cooperative Work. Retrieved November 14, 2007, from http://doi.acm.org/10.1145/587078.587082 Gross, E. F. (2004). Adolescent Internet us e: W hat we expect, what teens report. Journal of Applied Developmental Psychology, 25, 633 649. Hagerty, B. M., Lynch-Sauer, J., Patusky, K. L., Bouwsema, M., & Collier, P. (1992). Sense of belonging: A vital ment al health concept Archives of Psychiatric Nursing, 6 172-177. Hagerty, B. M., Williams, R. A., Coyne, J. C., & Early, M. R. (1996). Sense of belonging and indicators of social and psychological functioning. Archives of Psychiatric Nursing, 4 235-244. Hann, D., Winter, K., & Jacobsen, P. (1999). Measurement of depressive symptoms in cancer patients: Evaluation of the Center for Epidemio logical Studies Depression Scale (CES-D). Journal of Psychosomatic Research, 46 (5), 437-443. Hempel, J. (2005, December 12). The MySpace generation. BusinessWeek Retrieved November 14, 2007, from http://www.mywire.com/pubs/Bus inessW eek/2005/12/12/1102930?page=1 Hills, P., & Argyle, M. (2002). The Oxford Happ iness Questionnaire: a compact scale for the measurement of psychological well-being. Personality and Indivi dual Differences, 33 1073. Howe, N., & Strauss, W. (2000). Millennials rising: The next great generation New York: Vintage. Inderbitzen-Pisaruk, H., Clark, M. L., & Solano, C. H. (1992). Correlates of loneliness in midadolescence. Journal of Youth and Adolescence, 21 (2), 151-167. Jackson, T., Soderlind, A., & Weiss, K. E. (2000). Social Behavior and Personality, 28 (5), 463470. Jones, W. H., Carpenter, B. N., & Quintana, D. (1985). Personality and in terpersonal predictors of loneliness in two cultures. Journal of Personality and Social Psychology, 48 10531511. Jones, W. H., Freemon, J. E., & Goswick, R. A. (1981). The persistence of loneliness: Self and other determinants. Journal of Personality, 49(1), 27-48.

PAGE 88

88 Kandell, J. J. (1998). Internet addiction on cam pus: the vulnerability of college students. CyberPsychology & Behavior, 1 11. Kiralla, L. V. (2005). Internet addiction disorder: A descrip tive study of college counselors in four-year institutions (Doctoral dissertation, University of La Verne, 2005). Retrieved November 14, 2007, from PsychInfo database. Ko, C., Yen, J., Chen, C., Chen, S., & Yen, C. (2005). Proposed diagnostic criteria of Internet addiction for adolescents. Journal of Nervous and Mental Disorders, 193 728-733. Kraut, R., Patterson, M., Lundmark, V., Kiesle r, S., Mukopadhyay, T., & Scherlis, W. (1998). Internet paradox. A social technology that reduces social involvement and psychological well-being? American Psychologist, 53 (9), 1017. Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V. & Crawford, A. (2002). Internet paradox revisited. Journal of Social Issues, 58, 49. La Greca, A. M., & Lopez, N. (1998). Social anxiety among adolescents: Linkages with peer relations and friendships. Journal of Abnormal Child Psychology, 26 (2), 83-94. Leary, M. R. (1990). Responses to social excl usion: Social anxiety, jealousy, loneliness, depression, and low self-esteem. Journal of Social and Clinical Psychology, 9 221-229. Leung, L. (2002). Loneliness, self-disclosure and ICQ (I Seek You) use. CyberPsychology & Behavior, 5, 241. Leung, L. (2004). Net-generation attri butes and seductive properties of the Internet as predictors of online activities and Internet addiction. Cyberpsychology & Behavior, 7(3), 333-348. Lin, S. S. J., & Tsai, C. C. (1999). Internet Addiction among High Schoolers in Taiwan Poster presented at the 107th American Psychol ogy Association (APA) Annual Convention, Boston, USA. Marks, I. (1990). Non-chemical (behavioural) addictions. British Journal of Addiction, 85 1389 1394. Maslow, A. H. (1968). Toward a psychology of being. New York: Van Nostrand. McKenna, K. Y. A., Greene, A. S., & Gleason, M. E. J. (2002). Relationship formation on the Internet: Whats the big attraction? Journal of Social Issues, 58 9. McKenna, K. Y. A., & Bargh, J. (2000). Plan 9 from cyberspace: The implications of the Internet for personality and social psychology. Personality and Social Psychology Review 4, 5775. Menard, S. (1995). Applied logistic regression analysis. Quantitative Applications in the Social Sciences, 106 7-196.

PAGE 89

89 Mitchell, P. (2000). Internet addi ction: Genuine di agnosis or not? The Lancet, 355 632. Moore, D. (1995). The emperors virtual clothes: The naked truth about the Internet culture Chapel Hill, NC: Alogonquin. Morahan-Martin, J., & Schumacher, P. (2000). Incide nts and correlates of pathological Internet use among college students. Computers in Human Behavior, 16 13. Morahan-Martin, J., & Schumacher, P. (2003). L oneliness and social uses of the Internet. Computers in Human Behavior, 19, 659. Nalwa, K., & Anand, A. P. (2003). Internet a ddiction in students: A cause of concern. Cyberpsychology & Behavior, 6 (6), 653-656. Niemz, K., Griffiths, M., Banyard, P. (2005). Prevalence of pathologi cal Internet use among university students and correlations with se lf-esteem, the General Health Questionnaire (GHQ), and disinhibition. CyberPsychology & Behavior, 8 (6), 562-570. Owyang, J. (2008). Social network stats: MySpace, Facebook, and Reunion Retrieved July 20, 2008, from http://www.web-strategist.com/bl og/2008/01/09/social-network-statsfacebook-myspace-reunion-jan-2008 / Peter, J., Valkenburg, P. M., & Schouten, A. P. (2005). Developing a model of adolescent friendship formation on the Internet. Cyberpsychology & Behavior, 8 (5), 423-429. Pierce, G. R., Sarason, I. G., & Sarason, B. R. (1991). General and relationship-based perceptions of social support: Are two constructs better than one? Journal of Personality and Social Psychology, 61 1028-1039. Radloff, L. S. (1977). The Center for Epidemio logic StudiesDepression Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1, 385-401. Rice, K. G., Ashby, J. S., & Slaney, R. B. (1998). Self-esteem as a mediator between perfectionism and depression: A structural equations analysis. Journal of Counseling Psychology, 45 (3), 304-314. Rogers, Carl. (1959). A theory of therapy, personality and interperso nal relationships as developed in the client-centered framework. In S. Koch (Ed.), Psychology: A study of a science. Vol. 3: Formulations of the person and the social context. New York: McGraw Hill. Rosenberg, M. (1965). Society and the adolescent self-image Princeton, NJ: Princeton University Press. Russell, D. (1996). UCLA Loneliness Scale (V ersion 3): Reliability, validity, and factor structure. Journal of Personality Assessment, 66 20.

PAGE 90

90 Russell, D., Cutrona, C. E., Rose, J., & Yurko, K. (1984). Sound and emotional loneliness: An examination of Weisss typology of loneliness. Journal of Personality and Social Psychology, 46, 1313-1321. Russell, D., Peplau, L. A., & Cutrona, C. E. (1980). The revised UCLA loneliness scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39, 472. Scherer, K. (1997). College life on-line : healthy and unhealth y Internet use. Journal of College Student Development, 38, 655. Schlenker, B. R., & Leary, M. R. (1982). Social anxiety and se lf-presentation: A conceptualization and model. Psychological Bulletin, 92 641-669. Shafer, A. B. (2006). Meta-analysis of the factor structures of four de pression questionnaires: Beck, CES D Hamilton, and Zung. Journal of Clinical Psychology, 62 (1), 123-146. Shapira, N. A., Goldsmith, T. D., Keck, P. E., Jr., Khosla, U. M., & McElroy, S. L. (2000). Psychiatric features of individuals with problematic internet use. Journal of Aective Disorders, 57 (1), 267. Sobel, M. E. (1982). Asymptotic confidence interv als for indirect effects in structural equation models. In S. Leinhardt (Ed.), Sociological methodology (pp. 290). Washington, D.C.: American Sociological Association. Stephenson, W. (1988). The play theory of mass communication New Brunswick, NJ: Transaction Books. Suler, J. R. (1999). To get what you need : Healthy and pathologi cal Internet use. Cyberpsychology & Behavior, 2 (5), 355. Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd edition). New York: Harper & Row. Tapscott, D. (1998). Growing up digital: The rise of the Net Generation New York: McGrawHill. Treuer, T., Fabian, Z., & Furedi, J. (2001). Internet addic tion associated with features of impulse control disorder: Is it a real psychiatric disorder? Journal of Affective Disorders, 66 283. Valkenburg, P. M., & Peter, J. (2007). Preadol escents and adolescen ts online communication and their closeness to friends. Developmental Psychology, 43 (2), 267-277. Valkenburg, P. M., Peter, J., & Alexander, P. S. (2006). Friend networking sites and their relationship to adolescents well -being and social self-esteem. Cyberpsychology & Behavior, 9 (5), 584-590.

PAGE 91

91 Vassar, M. (2008). A note on the score reliability for the Sa tisfaction With Life Scale: an RG study. Social Indicators Research, 86(1), 47-57. Wan, C. S., & Chiou, W. B. (2006). Why are a dolescents addicted to online gaming? An interview study in Taiwan. Cyberpsychology & Behavior, 9 (6), 762-766. Watson, J. C. (2005). Internet a ddiction diagnosis and assessment: Implications for counselors. Journal of Professional Counseli ng: Practice, Theory & Research 33 (2), 17-30. Watson, D., & Friend, R. (1969). Measurem ent of social-evaluative anxiety. Journal of Consulting & Clinical Psychology, 33, 448. Weiss, R. S. (1974). Loneliness: The experience of emotional and social isolation Cambridge, MA: MIT Press. Widyanto, L., & Griffiths, M. (2006). Internet addiction: A critical review. International Journal of Mental He alth and Addiction, 4 31-51. Yellowlees, P. M., & Marks, S. (2007). Problem atic Internet use or Internet addiction? Computers in Human Behavior, 23, 1447. Yen, J., Ko, C., Yen, C., Wu, H., & Yang, M. ( 2007). The comorbid psychiatric symptoms of Internet addiction: Attention deficit and hype ractivity disorder (ADHD), depression, social phobia, and hostility. Journal of Adolescent Health, 41 93-98. Young, K. (1996a). Addictive use of the inte rnet: A case that breaks the stereotype. Psychological Reports, 79 899. Young, K. S. (1998). Internet addiction: th e emergence of a new clinical disorder. Cyberpsychology & Behavior, 1 (3), 237. Young, K. S. (1999). The research and cont roversy surrounding In ternet addiction. Cyberpsychology & Behavior, 2 (5), 381. Young, K. S., & Rogers, R. C. (1998). The rela tionship between depression and Internet addiction. Cyberpsychology & Behavior, 1 (1), 25. Yu, S., & Yu, M. (2007). Comparison of Intern et-based and paper-based questionnaires in Taiwan using multisample invariance approach. Cyberpsychology & Behavior, 10 501507.

PAGE 92

92 BIOGRAPHICAL SKETCH Andrea Spraggins attended Vanderbilt University in Nashville, TN, for her undergraduate degree, where she gradu ated magna cum laude with a double major in psychology and child development. She recently completed her cour sework for her doctorate degree in counseling psychology at the University of Florida, and is currently on internship at the University of Houston Counseling & Ps ychological Center.