(In)Visible Threats? The Third-Person Effect in Perceptions of the Influence of Facebook

Abstract

The popularity of Facebook has generated numerous discussions on the individual-level effects of social networking. However, we know very little about people's perceptions of the effects of the most popular social networking site, Facebook. The current investigation reports the findings from a survey designed to help us better understand young people's estimates of the perceived negative effects of Facebook use on themselves and others in regard to three outcome categories: (1) personal relationships, (2) future employment opportunities, and (3) privacy. Congruent with Davidson's third-person effect theory, respondents, when asked about the three outcome categories, believed that the use of Facebook had a larger negative impact on others (e.g., “your closest friends,” “younger people,” “people in your Facebook network of friends,” and “Facebook users in general”) than on themselves. Overall, results were inconclusive when it came to the link between the third-person perceptual gap and support for enhanced regulation of Facebook. Implications and limitations of this research are discussed.

Introduction

Facebook boasts of more than 500 million users who collectively spend 700 billion minutes and post over 30 billion pieces of content on the site each month.¹ These numbers force us to consider the individual and societal-level implications of social networking sites (SNS's). Certainly, Facebook's ability to facilitate interpersonal communication has contributed to its massive expansion. However, despite its communication-related benefits, critics have blasted Facebook over privacy concerns,² invasive advertising,³ and cyberbullying.⁴ Indeed, some of this criticism is warranted, with nearly 20% of undergraduates reporting that they have experienced negative effects of Facebook use.⁵

Although Facebook research remains relatively new, much of the existing literature focuses on college students, given their avid use of social media. As Facebook's popularity soars, so too does the need for research into the implications (actual and perceived) of this SNS. This study fills this gap by examining undergraduates' uses and perceived effects of Facebook within a third-person effect (TPE) theoretical framework.

Previous research and theoretical linkages

TPE⁶ predicts that people perceive themselves as invulnerable to the negative effects of media and perceive others as more affected (i.e., the perceptual component); further, this perceptual gap can lead to certain outcomes (i.e., the behavioral component), such as stances on regulation and censorship. Research suggests that the third-person gap may be due to “biased optimism” (i.e., one's feeling they are superior to others) and the need for “ego enhancement” (i.e., judging one's self as more competent than others).⁷

Historically, TPE has been applied to perceptions of the effects of traditional media, such as media violence,⁸ pornography,⁹ and rap music.¹⁰ Unlike traditional media, SNS's allow for user-generated content, personalization, and interactivity. Further, unlike most undesirable media typically explored using the TPE framework, Facebook can, theoretically, yield both positive and negative effects. However, few have investigated how people perceive the negative effects of Facebook on themselves versus others, and further, how this perceptual discrepancy relates to pro-regulatory attitudes toward Facebook.

Only recently has TPE been applied to SNS's. Zhang and Daugherty found that undergraduates believed Facebook to have a larger effect on others than themselves; these results supported the “social distance corollary,” the tendency to perceive people close to us as less affected than those perceived as distant. They also found evidence of behavioral consequences of TPE, with those exhibiting a higher degree of TPE less likely to visit SNS's recommended by friends.¹¹

Although the aforementioned study provides preliminary evidence of TPE within SNS's, it is limited in that the authors examined participants' perceived general effects of Facebook, but failed to examine specific types of effects. Research indicates the importance of identifying specific types of “effects” on self and others¹²; we concur, and extend this research by examining specific types of perceived effects of Facebook. Addressing this void in the literature is important, especially given concerns regarding young people's Facebook use in relation to personal relationships, future employment, and privacy. In the following section, we address the research in these areas.

Personal relationships

Much research has explored the impact of SNS's on personal relationships. Tong et al.¹³ emphasize Facebook's role in relational maintenance and relationship seeking. Likewise, Donath and Boyd¹⁴ contend that Facebook helps people maintain personal connections, thus allowing users to follow friends' updates and communicate through wall-posts. Further, Urista et al. posit that people use SNS's to “experience a selective, efficient, and immediate connection with others.”¹⁵^(p226)

Conversely, research indicates that Facebook may damage relationships, including friendships and romantic partnerships. Studies suggest that negative content of one's Facebook profile and the number of friends one has can lead to the formation of negative social judgments.¹⁶ Indeed, these negative social judgments can have detrimental consequences for personal relationships.¹⁷ Based on this literature, it is clear that Facebook use (in conjunction with content shared on one's profile) can result in positive and/or negative consequences for personal relationships. However, little is known regarding how individuals perceive the effects of Facebook on their relationships. Drawing on TPE theory, we advance the following hypothesis:

H1: Participants will rate others' “personal relationships” as more negatively affected by Facebook use than their own.

Employment

Much anecdotal evidence suggests that content posted on Facebook can be hazardous to users' current and future employment. Journalists routinely report cases in which employees have been terminated due to Facebook content deemed inappropriate by work superiors.¹⁸ Further, 45% of hiring managers use SNS's to research job candidates.¹⁹ These statistics are relevant to college students, who are avid Facebook users and will soon be seeking employment. However, gaps remain in our understanding of individuals' beliefs regarding the perceived impact of Facebook use on employment. Based on TPE theory, we predict the following:

H2: Participants will rate others' “future employment opportunities” as more negatively affected by Facebook use than their own.

Privacy concerns

Finally, critics have raised concern over privacy threats faced by SNS users, largely due to the amount of information shared online. Research finds that undergraduates are more likely to disclose personal information on Facebook than in day-to-day conversations,²⁰ and over 70% post demographic data on their profile.²¹ Debatin et al.²² found that people claim to understand SNS privacy threats; at the same time, they share large volumes of personal data, in part due to the belief that the benefits outweigh the risks. Collectively, research indicates that there is little relationship between amounts of personal information disclosed online and concerns over privacy.²³ However, little is known regarding how individuals perceive the effects of Facebook on their privacy versus how they see others' privacy as negatively influenced. Based on TPE, we predict the following:

H3: Participants will rate others' “privacy” as more negatively affected by Facebook use than their own.

Support for regulation: the behavioral component of TPE

Advancing research on the third-person perception (i.e., the perceptual component) is important in its own right; however, we concur with McLeod et al.²⁴ that the behavioral component is the more socially relevant phenomenon. Gunther argues that the importance of TPE lies in its potential to predict people's willingness to advocate action to protect others from perceived negative media influence.²⁵ Research examining third-person perception and the behavioral component has found a statistically significant relationship when the media content in question is viewed as negative.^26,27 However, as noted by Sun et al., within the context of socially desirable and/or ambiguous media, this relationship is unclear.²⁸

To our knowledge, the study at hand is the first to investigate the link between the perceptual component and regulation-related attitudes in the context of SNS's. However, it is difficult to predict whether there will be a positive relationship between the variables. Respondents may view Facebook as risk free, thereby diminishing the predictive value of third-person perceptions on pro-regulatory attitudes. Conversely, if respondents believe Facebook to have a greater negative impact on others, then we may find a relationship between third-person perception and regulatory support. Given this uncertainty, we pose the following question:

RQ1: Will the third-person perceptual gap predict pro-regulation attitudes?

Methods

Participants and sampling procedure

We conducted a survey of 357 undergraduates in the Northeastern U.S. (57.1% female; M=19.95 years, standard deviation [SD]=1.25; 91.6% Caucasian; see Table 1). After gaining Institutional Review Board (IRB) approval, a researcher distributed the 15-minute survey in over 20 classrooms, assuring respondents of anonymity. Classrooms were chosen due to class size and professors' approval.

Table 1.

Descriptive Statistics for Respondents' Characteristics

Item	Attributes	Total	%	Mean (SD)
Gender	Male	153	42.9
	Female	204	57.1
Race	White/Caucasian	327	91.6
	African American	7	2.0
	Latino/Hispanic	12	3.4
	Asian American	2	0.6
	Native American	1	0.3
	Other	7	2.0
Age	18–21	313	87.9	19.95 (1.25)
	22–24	43	12.1
Class year	First year	89	24.9
	Sophomore	118	33.1
	Junior	58	16.2
	Senior	92	25.8
Grade Point Average	3.0–4.0	298	85.1	3.30 (0.39)
	2.0–2.9	51	14.6
	<1.9	1	0.3
Facebook account	Yes	342	95.8
	No	15	4.2
Years with Facebook	<1 year	13	3.8
	1–2 years	122	35.7
	3–4 years	197	57.6
	>5 years	10	2.9
Minutes per day	0–10 minutes	70	20.5	37.44 (35.06)
	11–20 minutes	80	23.4
	21–30 minutes	64	18.7
	31–60 minutes	95	27.8
	>1 hour	33	9.6
No. of days/week	1 day/week	6	1.8
	2–3 days/week	28	8.2
	4–5 days/week	34	9.9
	6–7 days/week	274	80.1
Privacy settings used	Yes	301	88.0
	No	22	6.4
	Not sure/uncertain	19	5.6
Degree of privacy (i.e., who can view profile)	Everyone can view profile	13	3.8
	Only networks and friends lists	78	22.9
	Only friends lists	236	69.2
	Uncertain	14	4.1
Number of Facebook friends	0–200	44	13.6	480.26 (260.45)
	201–400	108	33.3
	401–600	102	31.5
	601–800	45	13.9
	>800	25	7.7

SD, standard deviation.

Measures

Use of Facebook

Items measured participants' Facebook use, including whether they had a profile, and how many days per week and minutes per day they logged on. Additional questions asked whether participants used privacy settings (yes, no, and uncertain), the degree to which privacy settings are used (e.g., the profile is open to everyone, just friends and networks, or just friends), and estimated size of the participant's network of friends.

Third-person perceptual items

Respondents were asked to estimate the negative effect of Facebook on “real-life personal relationships,” “future employment,” and “privacy” for the self and four comparison groups who differed in social distance (e.g., “your closest friends,” “younger people,” “people in your Facebook network of friends,” and “Facebook users in general”). Participants responded to a 5-point scale (1=“no negative effect”; 5=“strong negative effect”).

Third-person behavioral items

Support for enhanced Facebook regulation was measured by four items. For example, using a 5-point scale (strongly disagree to strongly agree), participants responded to the following statement: “More regulations are needed to safeguard people's private information on Facebook.” The Cronbach's alpha for this scale was 0.70.

Demographic information

The final section measured respondents' demographic data, including gender, race/ethnicity, and age.

Results

Describing the sample

Ninety-six percent of respondents had a Facebook account; non-Facebook users were excluded from subsequent analyses. Approximately 64% of participants logged on to Facebook seven days per week. On average, participants logged on to Facebook 3.62 (SD=2.76) times daily for 37.44 (SD=35.06) minutes. Respondents estimated having 480.26 Facebook friends (SD=260.45) and know approximately 86% of these people in real life; 27.5% reported accepting “friend requests” from strangers. Most (88%) used privacy settings, with 69.2% allowing only users on their “friends list” to view their profile (Table 1).

Testing hypotheses and exploring research questions

A series of paired t-tests comparing the raw self versus “other” scores addressed the first three hypotheses (Table 2). H1, that participants would view their personal relationships as less likely to be negatively influenced by Facebook than others, received partial support. Although participants perceived their personal relationships as significantly less negatively influenced than those of younger people, friends in their Facebook network, and Facebook users in general, the perceived difference between self and closest friends was not statistically significant. H2, that participants would view future employment as less likely to be negatively affected by Facebook than others, was supported for all comparison groups. H3, that participants would view their privacy as less likely to be negatively affected by Facebook than others, received support.

Table 2.

Paired t-Tests for Perceived Impact Estimates on Personal Relationships, Employment Opportunities, and Privacy on Self and Comparison Groups

	Comparison group
			Real world				Online (Facebook) world
	Self		Closest friends		Younger people		Network of Facebook friends		Facebook users (in general)
Type of effect	M	SD	M	SD	M	SD	M	SD	M	SD
Personal relationships	2.10	1.07	2.14	1.03	2.98^**	1.11	2.42^**	0.97	2.68^**	0.98
Future Employment	2.26	1.03	2.96^**	1.08	3.27^**	1.08	3.17^**	0.98	3.38^**	0.98
Privacy	2.53	1.11	2.77^**	1.02	3.36^**	0.98	3.02^**	0.95	3.22^**	0.93

Means and standard deviations reported here represent perceived negative effects on self and each target group. Higher means indicate greater perceived negative influence.

p<0.001, two-tailed t-tests between perceived impact on self and each target group.

To explore RQ1, which examined whether a greater third-person perceptual gap between self and others for the three effect types would be more likely to support enhanced Facebook regulations, a linear regression was run, with self-scores in the first block, gender in the second block, other-scores in the third block, and the summed pro-regulation score as the dependent variable. As shown in Table 3, the third-person perceptual gap (represented by other-scores in this analysis, due to our controlling self-scores in the first step of the model) for the perceived effects on privacy for self and closest friends predicted pro-regulatory attitudes. The belief that Facebook use could be hazardous for the future employment of younger people also predicted pro-regulation scores. Interestingly, with regard to personal relationships, perceptions for the self, people in the respondent's Facebook network, and Facebook users in general were found to predict pro-regulatory attitudes. Finally, women were more likely to support regulation of Facebook. The overall model accounted for 16% of variance in pro-regulation scores (F_2, ₃₂₄=5.00, p<0.001).

Table 3.

Standardized Regression Coefficients Showing the Effect of Gender, Self-Scores, and Other Scores on Pro-Regulation Attitudes Toward Facebook

Predictor variables	Pro-regulation attitudes
Demographics
Gender (0=male, 1=female)	0.15^*
Self-scores
Perceived effect on own future employment	−0.05
Perceived effect on own personal relationships	0.19^*
Privacy	−0.23^*
Other-scores
Perceived effect on future employment
For close friends	−0.03
For younger people	0.21^*
For people in your Facebook network	−0.04
For Facebook users in general	−0.09
Perceived effect on personal relationships
For close friends	−0.14
For younger people	0.08
For people in your Facebook network	0.25^*
For Facebook users in general	−0.20^*
Perceived effect on privacy
For close friends	0.25^*
For younger people	0.13
For people in your Facebook network	−0.04
For Facebook users in general	0.08
Total R square	0.20
Adjusted R square	0.16
n	326

The reported beta weights in the regression are final beta weights.

p<0.05.

Conclusion

Congruent with Davidson's theory, the current findings indicate that TPE, within the context of Facebook use, occurs for respondents' perceptions of influence on personal relationships, future employment, and privacy, such that they view themselves as significantly less likely to be influenced than their closest friends, younger people, friends in their Facebook network, and Facebook users in general. The only exception was when it came to participants' perceptions of negative influence on their closest friends' personal relationships; that is, there was no statistically significant perceptual gap between self and closest friends. This may be explained by the fact that one's closest friends are a part of the respondent's personal relationships; thus, if a respondent does not perceive his/her own relationships as negatively influenced by Facebook, then they have less reason to perceive their friends' relationships as negatively affected.

Meanwhile, despite using privacy settings, many respondents reported having expansive friend lists into the hundreds, and more than one-quarter accept “friend requests” from strangers. Indeed, respondents may be over-estimating their social ties to their network of Facebook friends, which helps explain why they perceive themselves to be largely immune, or less immune than others, to the negative effects of Facebook. Even with privacy settings in place, the fact that most undergraduates have such large “friend networks” means that they may be vulnerable to an array of threats (e.g., identify theft, harassment, etc…). Since scholars and practitioners consider these findings, it appears that this research could help educate students (and people in general) about the risks of using Facebook.

When it comes to the question of whether the third-person perceptual gap predicts pro-regulatory attitudes, our findings are rather inconclusive. That is, in a few cases, for some of the effect types, and for some of the comparison groups, the perceptual gap predicted higher pro-regulation scores. Although research has documented a link between third-person perception and regulation of socially undesirable media,^29,30 the lack of a consistent relationship within this study may be due to the fact that respondents likely view Facebook (generally speaking) as a nonharmful or ambiguous activity, relative to other media content (e.g., media violence and pornography). Future research should continue to explore the complex relationship between the perceptual and behavioral components.

Although this study is an important contribution, it is not without limitations. Although we had valid reasons for surveying undergraduates, we concur with Paul et al. that our findings of a third-person perceptual gap may be partly due to student-participants perceiving themselves to be smarter than others.³¹ Future studies should sample populations beyond predominantly Caucasian undergraduates, especially as Facebook spreads to different demographics. Additionally, although this study examined TPE in relation to overall Facebook use, future research should examine the role of specific content shared by the users. That is, is TPE magnified for users who disclose little on their Facebook profiles and/or are vigilant about the content they or others post on Facebook? As such, conducting a TPE survey in conjunction with a content analysis of Facebook profiles provides an innovative direction for research. Finally, scholars should continue to investigate factors, such as self-optimism and ego enhancement, which contribute to the TPE. Despite these limitations, we believe that exploring TPE within the context of SNS's provides important findings that are of scholarly and practical value.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Facebook. 2010. Facebook statistics. www.facebook.com/press/info.php?factsheet#!/press/info.php?statistics. 2011 Jan. 20.

Boyd

. Facebook's privacy trainwreck. Convergence: The Journal of Research into New Media Technologies, 2008; 14:13–20.

Debatin

, Lovejoy

, Horn

et al. Facebook and online privacy: attitudes, behaviors, and unintended consequences. Journal of Computer-Mediated Communication, 2009; 15:83–108.

Wortham

. Facebook and privacy clash again. The New York Times. May 6, 2010; B.1.

Zeng

, Huang

, Dou

. Social factors in user perceptions and responses to advertising in online social networking communities. Journal of Interactive Advertising, 2009; 10:1–13.

Meredith

. Combating cyberbullying: emphasizing education over criminalization. Federal Communications Law Journal, 2010; 63:311–340.

Mesch

. Parental mediation, online activities, and cyberbullying. CyberPsychology and Behavior, 2009; 12:387–393.

Davison

. The third-person effect in communication. Public Opinion Quarterly, 1983; 47:1–15.

Paul

, Salwen

, Dupagne

. The third-person effect: a meta-analysis of the perceptual hypothesis. Mass Communication and Society, 2000; 3:57–85.

10.

Gunther

, Mundy

. Biased optimism and the third-person effect. Journalism Quarterly, 1993; 70:58–67.

11.

Hoffner

, Plotkin

, Buchanan

et al. The third-person effect in perceptions of the influence of television violence. Journal of Communication, 2001; 51:283–299.

12.

Scharrer

. Third-person perception and television violence: the role of out-group stereotyping in perceptions of susceptibility to effects. Communication Research, 2002; 29:681–704.

13.

Gunther

. Overrating the X-rating: the third-person perception and support for censorship of pornography. Journal of Communication, 1995; 45:27–38.

14.

Rojas

, Shah

, Faber

. For the good of others: censorship and the third-person effect. International Journal of Public Opinion Research, 1996; 8:163–185.

15.

McLeod

, Eveland

, Nathanson

. Support for censorship of violent and misogynic rap lyrics: an analysis of the third-person effect. Communication Research, 1997; 24:153–174.

16.

Zhang

, Daugherty

. Third-person effect and social networking: implications for online marketing and word-of-mouth communication. American Journal of Business, 2009; 24:53–63.

17.

Salwen

, Dupagne

. The TPE: perceptions of the media's influence and immoral consequences. Communication Research, 1999; 26:523–550.

18.

Tong

, Van Der Heide

, Langwell

et al. Too much of a good thing? The relationship between number of friends and interpersonal impressions on Facebook. Journal of Computer-Mediated Communication, 2008; 13:531–549.

19.

Donath

, Boyd

. Public displays of connection. BT Technology Journal, 2004; 22:71–82.

20.

Urista

, Dong

, Day

. Explaining why young adults use myspace and Facebook through uses and gratifications theory. Human Communication, 2009; 12:215–229.

21.

Walther

, Van Der Heide

, Kim

et al.

The role of friends' behavior on evaluations of individuals' Facebook profiles: are we known by the company we keep?

Human Communication Research, 2008; 34:28–49.

22.

Muise

, Christofides

, Desmarais

. More information than you ever wanted to know—does Facebook bring out the green-eyed monster of jealousy? Cyber Psychology and Behavior, 2009; 12:441–444.

23.

Weiss

. 2011. Companies face ‘legal potholes’ as they crack down on workers' social media posts. American Bar Association Journal. www.abajournal.com/news/article/companies_face_legal_potholes_as_they_crack_down_on_workers_social_media_po/. 2011 Jan. 20.

24.

Haefner

. 2009. More employers screening candidates via social networking sites. www.careerbuilder.com/Article/CB-1337-Interview-Tips-More-Employers-Screening-Candidates-via-Social-Networking-Sites. 2010 May 15.

25.

Christofides

, Muise

, Desmarais

. Information disclosure and control on Facebook: are they two sides of the same coin or two different processes? CyberPsychology and Behavior, 2009; 12:341–345.

26.

Jones

, Soltren

. 2005. Facebook: threats to privacy. www.swiss.ai.mit.edu/6805/student-papers/fall05-papers/facebook.pdf. 2010 Nov. 5.

27.

Dwyer

, Hiltz

, Passerini

. 2007. Trust and privacy concern within social networking sites: a comparison of Facebook and MySpace. Proceedings of AMCIS, 2007. www.csis.pace.edu/∼dwyer/research/DwyerAMCIS2007.pdf. 2010 Nov. 5.

28.

Tufekci

. Can you see me now? Audience and disclosure management in online social network sites. Bulletin of Science and Technology Studies, 2008; 11:544–564.

29.

McLeod

, Detenber

, Eveland

. Behind the third-person effect: differentiating perceptual processes for self and other. Journal of Communication, 2001; 51:678–695.

30.

Chia

, Lu

, McLeod

. Sex, lies and video compact disc: a case study on third-person perception and motivations for censorship. Communication Research, 2004; 31:109–130.

31.

, Wei

. Third-person effect, gender, and pornography on the Internet. Journal of Broadcasting and Electronic Media, 2002; 46:13–33.

32.

Sun

, Pan

, Shen

. Understanding the third-person perception: evidence from a meta-analysis. Journal of Communication, 2008; 58:280–300.