8/2/2019 Social Media 9

1/7

The Theory of Planned Behavior Appliedto Young Peoples Use of Social Networking Web Sites

Emma L. Pelling, B.Behav.Sc. and Katherine M. White, Ph.D.

Abstract

Despite the increasing popularity of social networking Web sites (SNWs), very little is known about the psy-chosocial variables that predict peoples use of these Web sites. The present study used an extended model ofthe theory of planned behavior (TPB), including the additional variables of self-identity and belongingness, topredict high-level SNW use intentions and behavior in a sample of young people ages 17 to 24 years. Additional

analayses examined the impact of self-identity and belongingness on young peoples addictive tendencies to-ward SNWs. University students (N 233) completed measures of the standard TPB constructs (attitude, sub- jective norm, and perceived behavioral control), the additional predictor variables (self-identity and belongingness), demographic variables (age, gender, and past behavior), and addictive tendencies. One weeklater, they reported their engagement in high-level SNW use during the previous week. Regression analysespartially supported the TPB: attitude and subjective norm signficantly predicted intentions to engage in high-level SNW use with intention signficantly predicting behavior. Self-identity, but not belongingness, signficantlycontributed to the prediction of intention and, unexpectedly, behavior. Past behavior also signficantly predictedintention and behavior. Self-identity and belongingness signficantly predicted addictive tendencies towardSNWs. Overall, the present study revealed that high-level SNW use is influenced by attitudinal, normative, andself-identity factors, findings that can be used to inform strategies that aim to modify young peoples high levelsof use or addictive tendencies for SNWs.

Introduction

The popularity and use of social networking Web sites(SNWs) is rapidly increasing with these sites fast be-

coming one of the communication mediums of preference foryoung people.1 Research suggests that SNW use may increaseadolescents self-esteem and well-being if the tone of thefeedback provided by viewers of their profiles is positive butmay decrease self-esteem when the feedback is negative.2 Aswith any type of excessive Internet use, overuse of SNWs hasthe potential to negatively affect an individuals study, work,health, and personal relationships.3 Young adults are morelikely than any other age group to have an SNW4 and engagein higher levels of use (almost 3 hr=day).5 Despite the in-creasing popularity of SNWs, however, there is still littleknown about the psychosocial variables that predict peopleslevel of use.

One well-validated decision-making model that may pro-vide an appropriate framework to understanding and pre-dicting peoples use of SNWs is the theory of plannedbehavior (TPB).6 According to the TPB, behavior is deter-

mined by ones intentions to perform the behavior. Intentionis influenced by attitude, subjective norm, and perceivedbehavioral control (PBC, also said to directly impact behav-ior). Attitude reflects an individuals positive or negativeevaluation of performing the behavior. Subjective norm is theperceived social pressure to perform or not perform the be-havior. PBC is the perceived ease or difficulty that the indi-vidual attaches to performing the behavior. A meta-analysisof TPB studies indicated that the average amount of variancein intention and behavior accounted for by the TPB was 39%and 27% respectively.7 The TPB has been used to predict theperformance of a wide range of behaviors, including thoseinvolving technology.8,9 Despite the support for the TPBmodel, researchers have questioned whether the TPB vari-ables encompass all of the predictors of peoples intentionsand behavior.10 Because SNWs are a medium for identityexpression and self-presentation,11 examination of the role ofself-identity in SNW use is warranted. Additionally, becauseone of the most fundamental motives behind behavior is theneed for relationships with others,12 the effect of the need forbelongingness on young peoples SNW use was explored.

School of Psychology and Counselling, Queensland University of Technology, Queensland, Australia.

CYBERPSYCHOLOGY & BEHAVIORVolume 12, Number 6, 2009 Mary Ann Liebert, Inc.DOI: 10.1089=cpb.2009.0109

755

8/2/2019 Social Media 9

2/7

Self-identity

Based on Strykers13,14 identity theory, self-identity reflectsthe extent to which engaging in a behavior is important toan individuals self-concept.15 According to identity theory,the concept of the self is a set of socially constructed rolesreflecting the extent to which individuals see themselves asfulfilling the criteria for particular societal roles.16 Numerousstudies have demonstrated the ability of self-identity to im-prove the predictive efficiency of the TPB.8,15 Self-identity canbe expressed through interactions with mental and physicalobjects.7 SNW profiles allow for self-identity expressionthrough personalization as users can define themselves(who they are and what they like).11 It is expected, then, thatindividuals who indicate that using SNWs is important totheir self-concept will be more likely to engage in high-leveluse of them.

Belongingness

Human beings possess a core motive to interact with oth-ers, with many social behaviors being driven by the need for belongingness with others.12 The need for belongingness is

defined as the need to be involved with, accepted by, andvalued by others.16 SNWs have the potential to satisfy peo-ples need to belong by facilitating communication withothers, especially those with shared interests.17 People with agreater need for belongingness with others, then, may bemore likely to engage in high-level SNW use in an attempt tosatisfy their need to belong.

Addictive tendencies

When examining peoples high-level use of SNSs, as withother Internet activities, there is the potential issue raised ofsome peoples use encompassing addictive tendencies. It hasbeen argued that for a behavior to be defined as addictive, theactivity must meet six criteria: salience, mood modification,tolerance, withdrawal symptoms, conflict, and relapse.18 Be-havioral addictions may not have observable symptoms, andengaging in the behavior may also be associated with directbenefits.19 In the last few decades, the concept of Internetaddiction has arisen, with some individuals engaging incompulsive and problematic levels of Internet use.20 Researchsuggests that excessive Internet use can have a wide range ofnegative effects, including loneliness; depression; and aca-demic, professional, and personal problems.21 At present, theextent to which high-level users of SNWs experience symp-toms of addiction and negative consequences from their levelof use is currently unclear. The present study explores theimpact of both self-identity and belongingness on youngpeoples addictive tendencies for SNW use.

The present study

The present study aimed to test an extended TPB, includ-ing self-identity and belongingness, to predict young adultsintentions to engage in and subsequent engagement in high-level SNW use. It was expected that attitude, subjective norm,and PBC would predict intention to engage in high-levelSNW use, after taking into account the demographic factorsof age, gender, and past behavior (Hypothesis 1), with theaddition of self-identity and belongingness expected to im-prove the prediction of intention (Hypothesis 2). It was also

expected that intentions and PBC would predict high-levelSNW use in the 1 week follow-up period (Hypothesis 3). Inlight of some young peoples heavy usage of SNWs, the im-pact of self-identity and belongingness on SNW addictivetendencies was explored (Hypothesis 4).

Method

Participants

Participants completing the main questionnaire were 233undergraduate university students (149 female, 84 male) aged17 to 24 years (M 19.22, SD 2.03). One week after com-pleting the main questionnaire, 55% of participants (n129)completed the follow-up questionnaire.

Design

A prospective design was used to examine an extension ofthe TPB to predict high-level SNW use intentions and sub-sequent behavior. The main questionnaire assessed the stan-dard TPB items, self-identity, belongingness, addictivetendencies, and demographic variables (age, gender, and past

behavior). One week following the main questionnaire, con-senting participants completed a follow-up questionnaireassessing their engagement in high-level SNW use during theprevious week.

Measures

Target behavior

The target behavior of high-level SNW use was defined asmaking 4 or more unique visits per day to SNWs (e.g., Fa-cebook, MySpace, and Bebo) in the next week. Unique visitsreferred to the number of times an individual uses an SNW(rather than the number of times they log on to the Web site),as many users log on only once per day but leave the Web

page open and return to view the Web site a number of times.Four or more unique visits per day was deemed to constitutea high level of use for the population of interest based onprevious SNW use data.22 In the present study, 53% of par-ticipants indicated that they had engaged in high-level use atleast once in the previous week.

Main questionnaire

Most of the questionnaire items were worded positively,with some items negatively worded (and subsequently re-coded) to reduce response bias. The standard TPB items werescored on 7-point Likert scales, with the exception of attitude,which was measured using semantic differential scales.

TPB variables

Intention. Three items assessed intention to engage inhigh-level SNW use in the next week. For example, I intendto make 4 or more unique visits to social networking Websites (e.g., Facebook, MySpace, Bebo) in the next week: 1,strongly disagree, to 7, strongly agree. The measure of intentionwas reliable (a 0.92).

Attitude. Attitude was measured using five 7-pointsemantic-differential scales. For example, For me to make 4or more unique visits per day to social networking Web sites

756 PELLING AND WHITE

8/2/2019 Social Media 9

3/7

(e.g., Facebook, MySpace, Bebo) in the next week would be_____: 1, valuable, to 7, worthless (7). The measure of attitudewas reliable (a 0.83).

Subjective norm. Subjective norm was measured usingthree items. For example, Those people who are important tome would approve of me making 4 or more unique visits perday to social networking Web sites (e.g., Facebook, MySpace,

Bebo) in the next week: 1, strongly disagree, to 7 strongly agree.The measure was reliable (a0.82).

PBC. PBC was measured using two items. For example,I have complete control over whether I will make 4 or moreunique visits per day to social networking Web sites (e.g.,Facebook, MySpace, Bebo) in the next week: 1, strongly dis-agree, to 7, strongly agree. The two items correlated signifi-cantly (r(233)0.50, p< 0.001).

Additional TPB variables: self-identity

and belongingness

Self-identity. Three items adapted from Terry et al.15

measured self-identity. For example, I am the type of personwho uses social networking Web sites (e.g. Facebook, My-Space, Bebo): 1, completely false, to 7, completely true. The scalewas reliable (a 0.76).

Belongingness. Ten items drawn from Baumeister andLeary23 measured belongingness. For example, If otherpeople dont seem to accept me, I dont let it bother me: 1, notat all, to 5, extremely. The belongingness scale was reliable(a 0.74).

Past behavior. One item measured past behavior: In theprevious week, on how many days did you make 4 or moreunique visits per day to social networking Web sites (e.g.,

Facebook, MySpace, Bebo)?: from 0 days to 7 days.

Addictive tendencies. Based on previous research,24 eightitems assessed addictive tendencies toward SNWs. For ex-ample, I often think about social networking Web sites (e.g.,Facebook, MySpace, Bebo) when I am not using them: 1,strongly disagree, to 7, strongly agree. The addiction scale wasreliable (a 0.85).

Follow-up questionnaire

One week after the completion of the main questionnaire,participants indicated on an 8-point scale ranging from 0 daysto 7 days how many days in the previous week they had made

4 or more unique visits per day to SNWs.

Procedure

Ethics approval for the study was obtained from the uni-versitys ethics Committee (approval number 0800000160).Data were collected during sessions held on campus. Oneweek after completion of the main questionnaire, consentingparticipants (85%) were e-mailed the follow-up question-naire. Participant responses on the main and follow-upquestionnaires were matched using a unique code identifierto protect anonymity.

Results

Predicting intentions to engage in high-level SNW use

A hierarchical multiple regression analysis was conductedto predict intention to engage in high-level SNW use (seeTable 1). The step 1 variables of age, gender, and past behavior significantly accounted for 46% of the variance,F(3, 222) 63.64,p< 0.001. The addition of the step 2 variables

of attitude, PBC, and subjective norm increased the propor-tion of variance to 66%, F(6, 219) 72.15, p< 0.001. For step 3,the addition of self-identity and belongingness accounted fora further 2% of variance, with the full model accounting for68% of the variance, F(8, 217) 58.61, p< 0.001. When allvariables were entered into the equation, of the significantpredictors, past behavior had the largest beta weight, fol-lowed by subjective norm, attitude, and self-identity.

Predicting high-level SNW use behavior

On average, participants reported at follow-up that in theprevious week, they made 4 or more unique visits per day toSNWs between 1 and 2 days (M 1.40 days, SD2.19 days).

A hierarchical multiple regression was conducted to predicthigh-level SNW use (see Table 1). The step 1 variables of age,gender, and past behavior significantly accounted for 52% ofthe variance, F(3, 120) 43.59, p< 0.001. The addition of the

Table 1. Hierarchical Multiple Regression Analysisincluding Age, Gender, and Past Behavior

in Predicting Intentions to Engage

in and Subsequent Engagement in High-level

Social Networking Web Site Use

Variable B b R2 DR2

Prediction of intentionsStep 1

Age 0.04 0.05 0.46*** 0.46***Gender 0.06 0.02Past behavior 0.32 0.40***

Step 2Attitude 0.30 0.19*** 0.66*** 0.20***Subjective norm 0.45 0.31***PBC 0.09 0.05

Step 3Self-identity 0.24 0.18*** 0.68*** 0.02**Belongingness 0.08 0.03

Prediction of behaviorStep 1

Age 0.01 0.01 0.52*** 0.52***Gender 0.09 0.02

Past behavior 0.48 0.46***Step 2Intention 0.41 0.33** 0.57*** 0.05**PBC 0.04 0.02

Step 3Attitude 0.18 0.10 0.60*** 0.03Subjective norm 0.13 0.07Self-identity 0.31 0.19*Belongingness 0.03 0.01

Note: Weights included are those found in the final step of theanalysis.

PBC, perceived behavioral control.*p

8/2/2019 Social Media 9

4/7

step 2 variables of intention and PBC increased the propor-tion of variance to 57%, F(5, 118) 31.50, p< 0.001. For step 3,the addition of attitude, subjective norm, self-identity, andbelongingness accounted for a further 3% of variance, withthe full model accounting for 60% of the variance, F(9,114) 18.91, p< 0.001. When all variables were in the equa-tion, of the significant predictors, past behavior had thelargest beta weight, followed by intentions and self-identity.

Predicting SNW addictive tendencies

In general, participants in the present study somewhatdisagreed with statements suggesting addictive tendenciestoward SNWs (M 3.06, SD 1.11), although there was asmall subset of participants who did report the presence ofaddictive tendencies for SNWs (with scores above the mid-point of the scale). A hierarchical regression revealed that step1 variables of age and gender did not significantly predictaddictive tendencies, F(2, 228) 2.79, p 0.06. For step 2,with the addition of self-identity and belongingness, the fullmodel significantly accounted for 58% (56% adjusted) of thevariance, F(4, 226) 78.36, p< 0.001. When all variables wereentered into the equation, of the significant predictors, self-identity had the largest beta weight, b 0.71, p< 0.001, fol-lowed by belongingness, b 0.10, p< 0.05 (age and genderwere not significant predictors).

Discussion

The present study tested an extension of the TPB to predictyoung adults intentions to engage in and subsequent en-gagement in high-level SNW use. In partial support of Hy-pothesis 1, attitude and subjective norm emerged assignificant predictors of intention to engage in high-levelSNW use, suggesting that young adults with a more favor-able attitude toward high-level SNW use, and who felt morepressure from others to use these Web sites at high levels,

were more likely to intend to engage in high-level SNW use.Providing partial support for Hypothesis 3, intentions sig-nificantly predicted engagement in high-level SNW use in thefollow-up period. Contrary to prediction, however, PBC didnot emerge as a significant predictor of intentions or behav-ior. Because the effect of PBC decreases as the level of volitionincreases,8 this finding may reflect peoples high volitionalcontrol over using SNWs given their accessibility.

Hypothesis 2 was partially supported because self-identity,but not belongingness, significantly predicted intentions toengage in high-level SNW use. The finding that self-identitysignificantly predicted intentions suggests that the more aperson self-identifies as being an SNW user (i.e., when SNWuse is an important part of the persons self-concept), the

greater the persons intention to engage in high-level SNWuse, and it supports previous research for the role of self-identity in the TPB.11,18 Unexpectedly, self-identity had a di-rect effect on high-level SNW use behavior, suggesting thatthe more SNW use is a salient part of a young adults identity,the greater the individuals use of these Web sites. The find-ings of the present study suggest that the need for belong-ingness with othersdoes not play a role in predicting intentionto engage in high-level SNW use. It is possible that SNW usemay not meet users need for belongingness compared toother forms of communication, such as face-to-face interac-tions. For Hypothesis 4, self-identity and belongingness were

significant predictors of addictive tendencies, suggesting thatindividuals who self-identify strongly as SNW users and havea strong need to belong with others were more likely to reportadditive tendencies toward SNWs. Thus, while need for be-longingness may not influence the amount of SNW use, itmight influence the psychological intensity of use and feelingsthat the user has toward the use of the Web sites.

Strategies that aim to reduce problematic high-level use or

overuse could emphasize that important people in an indi-viduals life (e.g., friends, coworkers) would disapprove oftheir engaging in high-level SNW use (subjective norm), re-inforce the negative consequences (e.g., missing out on otherenjoyable activities) associated with high level use (attitudes),and discourage people from embracing too strongly theidentity of an SNW user, endeavoring to make SNW use lessimportant to their self-concept (self-identity). Strategies thataim to reduce SNW addictive tendencies could discouragepeople from embracing the identity of being an SNW userand emphasize that belongingness needs may be better metvia face-to-face communications.

The present studys strengths include its sound theorybase, prospective design, and a sample most likely to exhibit

high levels of SNW use. Some limitations include that theparticipants were all university students and predominantlyfemale and the low response rate to the follow-up question-naire. Overall, the present study provided some support forthe application of the TPB model in the context of high-levelSNW use with attitude and subjective norm significantlypredicting intention, which, in turn, significantly predictedbehavior. The present study provides support for the inclu-sion of self-identity to the TPB, and both self-identity andbelongingness predicted young adults SNW addictive ten-dencies. By identifying the factors that predict young adultshigh-level SNW use intentions and behavior, the results of thepresent study can be used in efforts to decrease problematicuse and addictive tendencies toward these Web sites.

Acknowledgments

The authors would like to thank Shari Walsh for her as-sistance in the design of the study.

Disclosure Statement

No competing financial interests exist.

References

1. Peluchette J, Karl K. Social networking profiles: an exami-nation of student attitudes regarding use and appropriate-ness of content. CyberPsychology & Behavior 2008; 11:

957.2. Valkenburg P, Peter J, Schouten A. Friend networking sitesand their relationship to adolescents well-being and socialself-esteem. CyberPsychology & Behavior 2006; 9:58490.

3. Caplan S. Problematic Internet use and psychosocial well- being: development of a theory-based cognitive-behavioralmeasurement instrument. Computers in Human Behavior2002; 18:55375.

4. Murnan C. Expanding communication mechanisms: theyrenot just e-mailing anymore. Special Interest Group on Uni-versity and College Computer Services 2002; 5:26772.

5. Raacke J, Bonds-Raacke J. MySpace and Facebook: applyingthe uses and gratifications theory to exploring friend-

758 PELLING AND WHITE

8/2/2019 Social Media 9

5/7

networking sites. CyberPsychology & Behavior 2008; 11:16974.

6. Ajzen I. The theory of planned behavior. OrganizationalBehavior & Human Decision Processes 1991; 50:17911.

7. Connor M, Armitage C. Extending the theory of plannedbehavior: a review and avenues for further research. Journalof Applied Social Psychology 1998; 28:142964.

8. Walsh S, White K. Me, my mobile, and I: the role of self- andprototypical identity influences in the prediction of mobilephone behavior. Journal of Applied Social Psychology 2007;37:240534.

9. Lin H. Understanding behavioral intention to participate invirtual communities. CyberPsychology & Behavior 2006; 9:5407.

10. Sparks P, Guthrie C. Self-identity and the theory of plannedbehavior: a useful addition or an unhelpful artifice? Journalof Applied Social Psychology 1998; 28:13931410.

11. Boyd D. (2006) Friends, Friendsters, and MySpace top 8:writing community into being on social networking web-sites. www.firstmonday.org=issues=issue11_12=boyd=index.html (accessed Aug. 15, 2008).

12. Fiske S. (2004) Social beings: core motives in social psychology.Hoboken, NJ: Wiley.

13. Stryker S. Identity salience and role performance: the rele-vance of symbolic interaction theory for family research.

Journal of Marriage & the Family 1968; 30:55864.14. Stryker S. (1987) Identity theory: developments and exten-

sions. In Yardley K, Honess T, eds. Self and identity: psycho-social perspectives. New York: Wiley, pp. 89103.

15. Terry D, Hogg M, White K. The theory of planned behav-iour: self-identity, social identity and group norms. British

Journal of Social Psychology 1999; 38:22544.16. Leary M, Cox C, Shah J, et al. (2008) Belongingness moti-

vation: a mainspring of social action. In Shah J, Gardner W,eds. Handbook of motivation science. New York: GuilfordPress, pp. 2740.

17. McKenna K, Bargh J. Causes and consequences of socialinteraction on the Internet: a conceptual framework. MediaPsychology 1999; 1:24969.

18. Griffiths M. Nicotine, tobacco and addiction. Nature 1996;384:189.

19. Young K. Internet addiction: a new clinical phenomenonand its consequences. American Behavioral Scientist 2004;48:40215.

20. Beard K, Wolf E. Modification in the proposed diagnosticcriteria for Internet addiction. CyberPsychology & Behavior2001; 4:37783.

21. Caplan S. Preference for online social interaction: a theory ofproblematic Internet use and psychosocial well-being.Communication Research 2003; 30:62548.

22. Microsoft (2008) Reach the net generation by the millions onFacebook. http:==advertising.microsoft.com=WWDocs=User=en-us=Advertise=Partner%20Properties=Facebook=MDAS_Facebook_Dec07.pdf (accessed May 20, 2008).

23. Baumeister R, Leary M. The need to belong: desire for in-terpersonal attachments as a fundamental human motiva-tion. Psychological Bulletin 2005; 117:497529.

24. Ehrenberg A, Juckes S, White K, et al. Personality and self-esteem as predictors of young peoples technology use.

CyberPsychology & Behavior 2008; 11:73941.

Address correspondence to:Dr. Katherine M. White

School of Psychology and CounsellingQueensland University of Technology

Victoria Park RoadKelvin Grove, Queensland

Australia, 4059

E-mail: km.white@qut.edu.au

YOUNG PEOPLES SOCIAL NETWORKING WEB SITE USE 759

8/2/2019 Social Media 9

6/7

8/2/2019 Social Media 9

7/7

Copyright of CyberPsychology & Behavior is the property of Mary Ann Liebert, Inc. and its content may not be

copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.