Sharing,Caring,and Surveilling: An Actor–Partner Interdependence Model Examination of Facebook Relational Maintenance Strategies

Abstract

Relational maintenance is connected to high quality friendships. Friendship maintenance behaviors may occur online via social networking sites. This study utilized an Actor–Partner Interdependence Model to examine how Facebook maintenance and surveillance affect friendship quality. Bryant and Marmo's (2012) Facebook maintenance scale was evaluated, revealing two factors: sharing and caring. Facebook surveillance was also measured. For friendship satisfaction and liking, significant positive actor and partner effects emerged for caring; significant negative actor, partner, and interaction effects emerged for sharing; and significant positive actor effects emerged for surveillance. For friendship closeness, significant positive actor effects emerged for caring and surveillance.

Increasingly, people use technology for social communication. Around 80% of American adults use the Internet, and many Internet users report that their social lives and relationships have improved through computer-mediated communication (CMC).^1,2 CMC tools such as e-mail, instant messaging, texting, and social networking sites (SNS) may allow individuals to supplement their offline maintenance strategies.^3,4

Relational maintenance behaviors are “strategic and routine behaviors that help people keep their relationships in desirable states” (e.g., satisfactory, close, etc.).⁵ Offline behaviors contributing to desired relational states range from communicating positivity⁶ to providing support⁷ to specific discussions regarding the relationship.⁸ While much of the maintenance literature is focused on romantic relationships,⁹ some scholars have focused specifically on friendships.^7,10,11,12

SNS are defined by boyd and Ellison as web-based services allowing individuals to construct profiles and connect with others.¹³ Walther and Ramirez claimed relational maintenance might be the “greatest utility of social networking systems” because people are able to maintain relationships with a much larger pool of people than might be possible without social network sites.^14(p279) Although people state preferences for richer mediums (such as face to face) for interpersonal relational maintenance,¹⁵ when presented with low cost and easily accessible mediums through CMC, people will use lower bandwidth communication technologies for relational communication.³ SNS suitability for relational maintenance derives from how easily such sites allow phatic communication.¹⁶ Phatic communication is communication that functions specifically to establish bonds of personal union, that is, to maintain relationships.¹⁷ Facebook's facilitation of phatic communication allows individuals to maintain friendships via SNS.

Currently, Facebook reports more than one billion active users.¹⁸ The Atlantic reported one in five Internet page views is a Facebook page.¹⁹ Facebook provides users with a variety of ways to share social information, including status updates, wall posts, photographs, and private messages. Given the usefulness of SNS for maintaining relationships and Facebook's popularity, this study focuses specifically on how people use Facebook to maintain social relationships.

Of interest is both how people might use Facebook to maintain relationships as well as what the outcomes are for particular maintenance strategies. To examine Facebook maintenance strategies, Bryant and Marmo²⁰ developed a Facebook friendship maintenance scale from in-depth qualitative analyses of focus group responses.²¹ Bryant and Marmo²⁰ divided the items into Stafford and Canary's²² pre-existing conceptual categories for offline maintenance instead of testing the underlying structure. However, communicative actions that maintain relationships offline are not necessarily replicated in online behaviors.²³ Thus, one of the goals of this study will be to examine the underlying factor structure of Bryant and Marmo's scale.

Facebook maintenance strategies may have positive relational outcomes for friendships. Relational maintenance behaviors are correlated with higher quality friendships.^10,24 Research has found dyads enacting offline relational maintenance tend to perceive their relationships as more satisfying.^8,10,22 Communication via Facebook may also assist friends in maintaining close relationships. Other studies have found SNS use and Facebook maintenance are correlated with greater relational closeness.^20,25 Finally, given the utilization of the term “Like” in Facebook, it seems fruitful to examine Facebook maintenance behaviors impact on perceived liking in friendships. Specifically, it is hypothesized Facebook maintenance strategies are positively associated with relational satisfaction, closeness, and liking.

Friendship is a dyadic relationship.²⁶ While the quality of one's friendship is based in part on one's own behaviors, the interdependent nature of the relationship implies quality results from both friends' behaviors. Although much research has focused on social relationship variables and SNS,^25,27–30 very few have examined dyadic effects. This study seeks to rectify this gap in the literature by examining both the relationship between one's own Facebook behavior and relational quality outcomes and the relationship between one's friend's Facebook behavior and relational quality outcomes. It is expected one's friend's Facebook maintenance strategies will be positively associated with satisfaction, closeness, and liking.

On a more passive level, Facebook allows us to stay informed regarding our friends' activities through interpersonal surveillance.^31,32 Joinson found people ranked finding information regarding their contacts (termed social searching) as an important Facebook component.²⁹ Although surveillance is not terribly interactive, keeping tabs on friends in this manner may allow people to manage relationships by informing other types of communication.^31,33 For example, if Facebook alerts us our friend is having a birthday, we know to wish them Happy Birthday. If our friend posts about trying skydiving, we can ask about skydiving when we see them next. However, surveillance and maintenance may serve different functions. Surveillance may allow individuals to remain aware of connections in their larger network, and provide information that reduces general uncertainty about the network.³⁰ Maintenance behaviors, on the other hand, go beyond mere awareness and affect the overall quality of a particular relationship.⁵ For example, although interpersonal online surveillance may be mutual,³² friends may not be fully aware of each other's surveillance efforts. Thus, surveillance may lack the positive effects of more interactive strategies. Thus, the following research questions are posed:

RQ1: Are one's surveillance behaviors correlated with perceptions of relational quality?

RQ2: Does one's friend's surveillance behaviors contribute to perceptions of relational quality?

Methods

Participants and procedures

A convenience sample of 112 young adults dyads (N=224) responded to questions regarding Facebook maintenance. Young adult college students were selected as a sample, as peer relationships are particularly important to young adults.^33,34 Furthermore, young adults are heavy Facebook users. When these data were collected, 87% of adults aged 18–29 years used an SNS, of which Facebook is the most popular.³⁵ Undergraduates were recruited from undergraduate communication courses to take part in a study on friendship maintenance. Participants were given a three-letter code and asked to recruit a friend. Both dyad members were instructed to complete the online survey separately. Friendship dyad members were later matched based on the three-letter code. This survey was part of a larger data collection. Students were asked if they were Facebook friends with the friend they had recruited. Only 18 participants said that they were not. These participants were redirected to the rest of the survey and did not complete the Facebook measure.

Participants were young adults (M=21.27, SD=3.29, 40.3% male). The gender composition of dyads was 39.22% female–female, 27.45% male–male, and 33.33% cross-sex. A total of 83.9% of the sample reported being white, 14.7% black, 0.7% Asian, and 0.7% Hispanic. Intraclass correlations on the independent variables were significant, indicating the indistinguishable dyads were nonindependent (Table 1).

Table 1.

Intraclass Correlations Between Dyad Members on Independent Variables

	ρ
Caring	0.44
Sharing	0.44
Surveillance	0.31

Note. All intraclass correlations are significant at the p<0.001 level (asymptotic sig. test).

Measures

Grand means, grand standard deviations, and pooled alphas were calculated for all independent and dependent measures (Table 2). All measures except for closeness used 1 to 7 Likert-type scales where a response of 7 indicated strong agreement with the item and a response of 1 indicated strongly disagreement.

Table 2.

Alphas, Grand Means, and Grand Standard Deviations for All Measures

	Alpha (pooled)	M	SD
Caring	0.93	5.23	1.54
Sharing	0.91	4.20	1.81
Surveillance	0.84	4.00	1.79
Satisfaction	0.94	5.96	1.25
Closeness	n/a	5.37	1.56
Liking	0.94	5.73	1.20

Facebook maintenance

To determine the factor structure of Bryant and Marmo's scale, the 16 items were factor analyzed using maximum likelihood extraction and a promax rotation. Factor retention was based on parallel analysis comparison to 95th percentile eigenvalues.^36,37 Items with minimum loadings of 0.32 were retained.^38,39 Two items were dropped due to crossloadings. A two factor solution was obtained (KMO=0.89, see Table 3). The first factor contained items related to the self-disclosure functions of relational maintenance, and and this factor was labeled “sharing.” The items in the second factor represent messages showing care for and interaction with one's friend, and this factor was labeled “caring.”

Table 3.

Pattern Matrix for Facebook Maintenance Scale

Item	Caring	Sharing
Post on my friend's wall to make him/her feel special.	0.695	0.202
Send cheerful messages.	0.460	0.313
Respond in timely manner when my friend sends me a FB message.	0.742	−0.026
Update my profile info/status so my friend will stay up-to-date on my everyday life.	−0.001	0.833
Seek support by posting emotional (sad or exciting) news and hoping my friend will respond.	−0.247	1.04
Make posts to share my thoughts and positions on current events.	−0.142	0.993
Show this person I value our relationship by wishing them a happy birthday.	0.926	−0.195
Offer condolences and support when my friend posts bad news.	0.862	0.027
Offer congratulations when my friend shares good news in a post.	0.997	−0.113
Comment on my friend's profile so other users will see our connection.	0.246	0.643
Post photos so my friend can share my experiences even though s/he was not present.	0.551	0.240
Use FB to reflect on experiences I have shared with my friend.	0.476	0.392
Use FB to coordinate future interactions with this person.	0.087	0.655
Use FB's private message options to send info only to my friend.	0.174	0.532
Communicate with my friend using multiple components of FB.	0.344	0.501
Use public message options (wall/status) to send info I don't mind other people seeing.	0.732	−0.017

Bolded factor loading indicates that the item loaded on that factor.

FB, Facebook.

Facebook surveillance

Facebook surveillance was measured using items developed by Bryant and Marmo.²⁰ These items included, “I view this person's profile to monitor his/her interactions and watch out for his/her best interests,” “I feel like monitoring this person's Facebook page is sufficient to maintain this relationship,” “I pay attention to this person's updates as a way to know what he/she is doing w/o actually paying attention to him/her,” and “I pay attention to this person's updates when I am scrolling through the newsfeed so I feel up-to-date on his/her life.”

Relational quality

Satisfaction

Relational satisfaction was measured using an adapted version (the word marriage was replaced with “relationship”) of Norton's Quality of Marriage Index (QMI).⁴⁰ The QMI was originally developed to measure global satisfaction in marital relationships. However, adapted versions have been used to assess satisfaction in friendships and casual relationships.^41–44

Liking

Liking was measured using nine items from Rubin's measure. Items were adapted to read “my friend” instead of “my partner.”⁴⁵ Rubin's liking scale evaluates the respect and affection one has for one's partner. Previous research has assessed liking in friendships using an adapted version of the measure.^41,46–48

Closeness

Closeness was measured using Aron et al.'s Inclusion of the Other in the Self (IOS) pictorial measure.⁴⁹ Participants respond by choosing one set (out of seven) overlapping circles representing how close they feel to the other person. Circles are then assigned numeric values. A score of 7 indicates extreme closeness; a score of 1 indicates complete separateness.

Results

The hypotheses posited an individual's self-reported Facebook maintenance and surveillance strategies and their partner's Facebook maintenance and surveillance strategies would predict relational satisfaction, liking, and closeness. The hypotheses were tested with an actor–partner independence model (APIM) using a multilevel modeling technique, a preferred choice for testing actor–partner effects in indistinguishable dyads such as friends.^50,51 Predictor variables were centered using the grand mean.^51,52,53 Satterthwaite degrees of freedom are reported for all t tests.^51,53

Facebook maintenance

Satisfaction

Significant actor and partner effects emerged for Facebook maintenance behaviors on satisfaction. However, while caring correlated with satisfaction in the predicted direction for actors (t(173.93)=6.75, p<0.001) and partners (t(173.94)=3.27, p<0.001), both actor (t(174.92)=− 2.16, p<0.05) and partner (t(174.91)=−2.96, p<0.01) sharing were negatively related to satisfaction. In addition, a significant actor–partner interaction was detected (t(85.83)=2.49, p<0.05). Follow-up simple slopes analyses identified the interaction was such that when the actor's sharing behavior is low, the negative effect of the partner's sharing on satisfaction is greater. In addition, a simple slopes analysis of the interaction from the partner perspective identified when the partner's sharing is low, the actor's sharing has a negative effect on satisfaction (see Table 4 for a model summary).

Table 4.

Parameter Estimates for Facebook Sharing and Caring on Relational Outcomes

	b	β	SE_b
Satisfaction (pseudo R²=0.323^***)
Intercept	5.88
Sharing – Actor	−0.12	−0.22	0.07^*
Caring – Actor	0.45	0.70	0.07^***
Sharing – Partner	−0.17	−0.31	0.06^**
Caring – Partner	0.22	0.34	0.07^***
Sharing – Actor×Partner	0.07	0.22	0.03^*
Caring – Actor×Partner	0.03	0.08	0.03
Liking (pseudo R²=0.321^***)
Intercept	5.65
Sharing – Actor	−0.05	−0.10	0.06
Caring – Actor	0.41	0.64	0.07^***
Sharing – Partner	−0.18	−0.32	0.06^**
Caring – Partner	0.17	0.26	0.07^*
Sharing Actor×Partner	0.08	0.25	0.03^**
Caring – Actor×Partner	0.00	0.00	0.03
Closeness (pseudo R²=0.147^*)
Intercept	5.32
Caring – Actor	0.28	0.43	0.07^***
Caring – Partner	0.21	0.32	0.07^**
Caring – Actor×Partner	0.05	0.13	0.05

Note. ^***p<0.001, ^**p<0.001, ^*p<0.05.

Liking

Significant actor and partner effects emerged for Facebook maintenance behaviors on liking. Actor (t(174.89)=6.19, p<0.001) and partner (t(174.90)=2.53, p=0.01) caring positively correlated with liking. A different relationship was found for sharing and liking. Actor sharing (t(174.06)=−0.95, p=0.345) was not related to liking. Partner sharing (t(174.04)=−3.08, p<0.01) was significantly and negatively related to liking. An actor–partner effect on liking was detected (t(86.50)=2.94, p<0.01). A simple slopes analysis revealed a relationship similar to the actor–partner effect found for satisfaction. When actor sharing is low, the negative effect of the partner's sharing is greater. When partner's sharing is low, the effect of the actor's sharing on liking is negative but when partner sharing is high, the actor effect becomes nonsignificant (see Table 4 for a model summary).

Closeness

The initial model for closeness was not significantly different from the null model. Thus, the nonsignificant sharing effects were removed from the model. The model including only the caring variables was significant. Actor (t(169.32)=4.22, p<0.001) and partner (t(169.26)=3.15, p<0.01) caring maintenance behaviors were positively correlated with closeness (see Table 4 for a model summary).

Surveillance

Significant actor effects were found for Facebook surveillance on satisfaction (t(171.25)=3.31, p<0.001), liking (t(176.34)=4.15, p<0.001), and closeness (t(160.88)=3.26, p<0.001). No significant partner or interaction effects emerged (see Table 5 for a model summary).

Table 5.

Parameter Estimates for Facebook Surveillance on Relational Outcomes

	b	β	SE_b
Satisfaction (pseudo R²=0.078^***)
Intercept	5.94
Actor	0.15	0.27	0.05^***
Partner	−0.03	−0.05	0.05
Actor×Partner	0.06	0.19	0.03
Liking (pseudo R²=0.105^***)
Intercept	5.70
Actor	0.19	0.22	0.05^***
Partner	−0.04	−0.07	0.04
Actor×Partner	0.05	0.17	0.03
Closeness (pseudo R²=0.105^***)
Intercept	5.29
Actor	0.19	0.33	0.06^**
Partner	0.11	0.19	0.06
Actor×Partner	0.07	0.23	0.04

Note. ^***p<0.001; ^**p<0.001.

Discussion

In this study, three different functions of Facebook correlated in different ways to relational outcomes. The first function, sharing, is commonly associated with Facebook and SNS. Facebook has appeared to be a vehicle primarily for sharing self-disclosures.^54–57 Self-disclosure, in turn, has long been associated with increased relational development.⁵⁸ However, sharing was negatively related to the relational quality indicators tested here. For example, one's own sharing on Facebook was negatively related to satisfaction, and how much one's friend self-disclosed on Facebook was negatively related to both satisfaction and liking.

As noted, this finding is a departure from previous research connecting self-disclosure with relational development.^58–61 However, much of the previous work on self-disclosure focused primarily on self-disclosure occurring in interactions. Self-disclosure on Facebook may function much more as a broadcast than an interpersonal interaction. O'Sullivan termed this “masspersonal communication” and urged scholars to examine the differences between mass communication, interpersonal communication, and masspersonal communication.⁶² This finding may represent one of those differences; self-disclosure on an SNS may have different effects than reciprocal self-disclosure one would have in an interactive setting. Self-broadcast may have negative relational effects, particularly if receivers find themselves thrust into the role of the reluctant confidant.⁶³ The interaction effect between actor and partner sharing and liking suggests some individuals may find the broadcast nature of Facebook sharing annoying. The less likely they are to engage in Facebook sharing behaviors themselves, the more likely they are to find these broadcasts detrimental to the friendship.

Evidence of this potential explanation exists in other literature regarding communication on Facebook. For example, Carpenter found self-promoting behaviors (e.g., posting status updates, updating profile information) are associated with the grandiose exhibitionism facet of narcissism.⁶⁴ Bryant and Marmo identified the need for response and reciprocity as one of the important rules for Facebook friends.²⁷ This rule may reflect Facebook users' desire to feel like part of an ongoing interaction with their co-Facebookers rather than the reluctant recipient of a one-to-many broadcast.

On the other hand, the findings for Facebook caring were consistently and significantly positive. Individuals felt more relational satisfaction, more closeness, and liked their friend more when both they and their friend take the time to show they care about each other via Facebook. Caring messages may be reserved for those with whom we have close and satisfying relationships. Taking the time to send a targeted relational message may indicate a deeper and closer relationship.

Finally, this study also examined Facebook surveillance. There is a debate in the literature regarding if surveillance is maintenance and the potential effects of surveillance on relationships.^31–33 Bryant and Marmo found surveillance was used more by close friends than causal friends or acquaintances.²⁰ Responding to these results, Bryant and Marmo asked, “Can surveillance adequately maintain relationships without direct communication between partners?”^27(p14)

The emergence of only actor effects suggests the effects of surveillance may be more closely related to uncertainty reduction than maintenance.⁶⁵ Scholars have argued individuals are more likely to attempt to reduce uncertainty about individuals they perceive will offer future rewards.^66,67 This explanation may carry over to Facebook surveillance. People may be more likely to monitor close and high quality friendships due to the perception of future rewards rather than Facebook surveillance creating higher quality relationships. In addition, Tong and Walther argued one reason we may choose SNS for relational maintenance is SNS provide mild entertainment.²³ We may feel more entertained when we view updates and information regarding people we like, feel close to, and with whom we have a satisfying friendship. However, when our friends watch us online, this does not improve our perceptions of the overall quality of the relationship. Although, the dyadic analysis presented here provides some additional direction for understanding Facebook surveillance, future research should specifically test these theoretical explanations for the relationship between actor effects for relational quality and surveillance.

In total, these results may suggest interactivity effects for mediums offering options of both masspersonal broadcasts and unique specific interpersonal messages. Some have suggested Facebook is essentially naval gazing and perhaps even narcissistic.⁶⁴ Others have argued Facebook serves as way for us all to keep tabs on each other.^29,30 Using SNS for one-sided broadcasts may have deleterious relational effects. However, communicating over SNS in ways that show caring about a unique other may tap into the mutuality construct (perceptions of connection, similarity, receptivity, and understanding) of the interactivity principle.⁶⁸ The feelings of interdependence and understanding engendered by mutuality-type interactivity messages over SNS may contribute to overall relational quality in a way broadcast disclosures and entertaining surveillance may not.

Limitations and Future Directions

One limitation is the cross-sectional nature of this study. Although examining dyadic effects of Facebook friendship maintenance is a significant contribution to the literature, the direction of the effects cannot be determined from this study. The maintenance behaviors examined here may lead to greater or less satisfaction, liking, and closeness, or it may be the dyads' levels of relational satisfaction, liking, and closeness result in different types of Facebook communication.

This study also only examined the relationship between Facebook maintenance and relational quality without taking into account the effects of other forms of maintenance. People use a variety of channels to maintain their relationships (for a review of media multiplexity, see Haythornthwaite).⁶⁹ In addition, most people make SNS connections with individuals who they previously knew offline.²⁸ In some cases, Facebook may serve as a supplement for maintaining geographically close relationships. In primarily Internet-based relationships,⁴ Facebook may serve as the primary form of maintenance. Differences between these cases were not analyzed in this study.

Another limitation of this study is the results and interpretations rely on self-report data. Participants may remember or report their actions on Facebook differently from what actually occurred, although work by the Pew Internet and American Life Project has found that self-report survey data generally reflects the data from logs of actual Facebook usage.⁷⁰

In addition, specific a priori hypotheses regarding different types of Facebook maintenance were not posed because of the previous lack of clarity regarding the factor structure of the measure. Future studies should pose and test theoretical hypotheses regarding the differences between broadcast and interactive communicative strategies via CMC.

Conclusion

There has been a debate in both the literature and the popular press regarding whether Facebook has a detrimental influence on our lives and relationships^64,71,72 or whether it is a useful and helpful way to expand our social networks.^23,31,73,74 The results of this study suggest that in terms of maintaining desired states in relationships, communication on Facebook may be both helpful and harmful. The determining factor in whether SNS is helpful or harmful may be the way people use these channels. Individuals who share frequently without being responsive to their Facebook connections may risk aggravating these friends. Individuals who show caring and responsiveness to their friends may find that an SNS provides a convenient way to help keep a friendship satisfying and close.

Footnotes

Acknowledgment

A previous version of this article was presented at the International Association for Relationship Researchers, 2012, Chicago, IL.

Author Disclosure Statement

No competing financial interests exist.

References

Pew Internet and American Life Project. (2012) Trend data (adults). www.pewinternet.org/Trend-Data-(Adults)/Online-Activities-Daily.aspx (accessed Apr. 17, 2012).

Symantec Corporation. (2009) Norton online living report. www.symantec.com/about/news/release/article.jsp?prid=20090317_02 (accessed Mar. 22, 2012).

Rabby

, Walther

. (2003) Computer-mediated communication effects on relationship formation and maintenance. In Canary

, Dainton

, eds. Maintaining relationships through communication. Mahwah, NJ: Lawrence Erlbaum, pp. 141–162.

Wright

. On-line relational maintenance strategies and perceptions of partners within exclusively Internet-based and primarily Internet-based relationships. Communication Studies, 2004; 55:239–253.

Canary

, Stafford

. (1994) Maintaining relationships through strategic and routine interaction. In Canary

, Stafford

. eds. Communication and relational maintenance. San Diego, CA: Academic Press, pp. 3–22.

Canary

, Stafford

, Hause

, et al. An inductive analysis of relational maintenance strategies: comparisons among lovers, relatives, friends, and others. Communication Research Reports, 1993; 10:5–14.

Oswald

, Clark

, Kelly

. Friendship maintenance: an analysis of individual and dyad behaviors. Journal of Social & Clinical Psychology, 2004; 23:413–441.

Stafford

. Measuring relationship maintenance behaviors: critique and development of the revised relationship maintenance behavior scale. Journal of Social & Personal Relationships, 2011; 28:278–303.

Ogolsky

, Bowers

. A meta-analytic review of relationship maintenance and its correlates. Journal of Social & Personal Relationships., 2013; 30:343–367.

10.

McEwan

, Guerrero

. Maintenance behavior and relationship quality as predictors of perceived availability of resources in newly formed college friendship networks. Communication Studies., 2012; 63:421–440.

11.

Hays

. The development and maintenance of friendship. Journal of Social & Personal Relationships, 1984; 1:75–98.

12.

Dainton

, Zelley

, Langan

. (2003) Maintaining friendships throughout the life span. In Canary

, Dainton

, eds. Maintaining relationships through communication. Mahwah, NJ: Lawrence Erlbaum, pp. 79–124.

13.

boyd

, Ellison

. Social network sites: definition, history, and scholarship. Journal of Computer-Mediated Communication, 2007; 13:210–230.

14.

Walther

, Ramirez

. (2010) New technologies and new directions in online relating. In Smith

, Wilson

, eds. New directions in interpersonal communication research. Thousand Oaks, CA: Sage, pp. 264–284.

15.

Rice

. Media appropriateness: using social presence theory to compare traditional and new organizational media. Human Communication Research, 1993; 19:451–484.

16.

Radovanovic

, Regnedda

. (2012) Small talk in the digital age: making sense of phatic posts. Proceedings of the #MSM2012 Workshop. http://ceur-ws.org/Vol-838/paper_18.pdf (accessed Nov. 1, 2012).

17.

Malinowski

. (1936) The problem of meaning in primitive languages. In Ogden

, Richards

, eds. The meaning of meaning. New York: Harcourt Brace, pp. 296–336.

18.

Facebook. (2012) Key facts. http://newsroom.fb.com/content/default.aspx?NewsAreaId=22 (accessed Jun. 1, 2012).

19.

Madrigal

. (2012) The case for Facebook. www.theatlantic.com (accessed Jun. 1, 2012).

20.

Bryant

, Marmo

. Relational maintenance strategies on Facebook. Kentucky Journal of Communication 2010. http://kycommunication.com/jenniferpdf/Bryant.pdf (accessed Jun. 7, 2010).

21.

Bryant

, Marmo

. (2010) Using Facebook to maintain friendships: examining differences between acquaintances, casual friends, and close friends. Paper presented at a meeting of the National Communication Association, San Francisco, CA.

22.

Stafford

, Canary

. Maintenance strategies and romantic relationship type, gender, and relational characteristics. Journal of Social & Personal Relationships, 1991; 8:217–242.

23.

Tong

, Walther

. (2011) Relational maintenance and CMC. In Wright

, Webb

, eds. Computer-mediated communication in personal relationships. New York: Peter Lang, pp. 98–118.

24.

Oswald

, Clark

. How do friendship maintenance behaviors and problem-solving styles function at the individual and dyadic levels. Personal Relationships, 2006; 3:333–348.

25.

Ledbetter

, Mazer

, DeGroot

, et al. Attitudes toward online social connection and self-disclosure as predictors of Facebook communication and relational closeness. Communication Research, 2011; 38:27–53.

26.

Ledbetter

. Family communication patterns and relational maintenance behavior: direct and mediated associations with friendship closeness. Human Communication Research, 2009; 35:130–147.

27.

Bryant

, Marmo

. The rules of Facebook friendship. A two-stage examination of interaction rules in close, casual, and acquaintance friendships. Journal of Social & Personal Relationships, 2012; 29:1013–1035.

28.

Ellison

, Steinfield

, Lampe

. The benefits of Facebook “friends”: social capital and college students' use of online social network sites. Journal of Computer Mediated Communication, 2007. http://jcmc.indiana.edu/vol12/issue4/ellison.html (accessed Mar. 2, 2010).

29.

Joinson

. (2008) “Looking at,” “looking up,” or “keeping up with” people? Motives and uses of Facebook. Paper presented at the SIGCHI Conference on Human Factors in Computing Systems, Florence, Italy. New York: ACM Press/Addison-Wesley Publishing Co., pp. 1027–1036.

30.

Lampe

, Ellison

, Steinfield

. A face(book) in the crowd: social searching as social browsing. Proceedings of CSCW 2006. New York: ACM Press, pp. 167–170.

31.

Bryant

, Marmo

, Ramirez

. (2011) A functional approach to social networking sites. In Wright

, Webb

, eds. Computer-mediated communication in personal relationships. New York: Peter Lang, pp. 3–20.

32.

Trottier

. Interpersonal surveillance on social media. Canadian Journal of Communication, 2012; 37:319–332.

33.

Bliezner

, Adams

. (1992). Adult friendship. Newbury Park, CA: Sage.

34.

Chow

, Roelse

, Buhrmester

, et al. (2012) Transformations in friend relationships across the transition into adulthood. In Laursen

, Collins

, eds. Relationship pathways. Thousand Oaks, CA: Sage, pp. 91–111.

35.

Duggan

, Brenner

. (2013) The demographics of social media users – 2012. http://www.pewinternet.org/Reports/2013/Social-media-users.aspx (accessed Feb. 20, 2013).

36.

Hayton

, Allen

, Scarpello

. Factor retention decisions in exploratory factor analysis: a tutorial on parallel analysis. Organizational Research Methods, 2004; 7:191–205.

37.

Patil

, Singh

, Mishra

, et al. Parallel analysis engine to aid determining number of factors to retain 2007. http://ires.ku.edu/∼smishra/parallelengine.htm (accessed Dec. 10, 2012).

38.

Costello

, Osborne

. (2005) Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Practical Assessment, Research, & Evaluation, 2005; 10.

39.

Tabachnick

, Fidell

. (2001) Using multivariate statistics. Boston: Allyn & Bacon.

40.

Norton

. Measuring marital quality: a critical look at the dependent variable. Journal of Marriage & Family, 1983; 45:141–151.

41.

Goodboy

, Myers

. Relational maintenance behaviors of friends with benefits: investigating equity and relational characteristics. Human Communication, 2008; 11:71–86.

42.

McEwan

, Zanolla

. When online meets offline: A field investigation of modality switching. Computers in Human Behavior, 2013; 29:1565–1571.

43.

Peters

, Rowatt

, Johnson

. Associations between dispositional humility and social relationship quality. Psychology, 2011; 2:155–161.

44.

Yum

. The relationships among loneliness, self/partner constructive maintenance behavior, and relational satisfaction in two cultures. Communication Studies, 2003; 54:451–467.

45.

Rubin

. (1973) Liking and loving. New York: Holt, Rinehart, & Winston.

46.

Hess

. Measuring distance in personal relationships: the relational distance index. Personal Relationships, 2003; 10:197–215.

47.

Kaplan

, Keys

. Sex and relationship variables as predictors of sexual attraction in cross-sex platonic friendships between young heterosexual adults. Journal of Social & Personal Relationships, 1997; 14:191–206.

48.

Pontari

. Appearing socially competent: the effects of a friend's presences on the socially anxious. Personality & Social Psychology Bulletin, 2009; 25:283–294.

49.

Aron

, Aron

, Smollan

. Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality & Social Psychology, 1992; 63:596–612.

50.

Kenny

. Models on nonindependence in dyadic research. Journal of Social & Personal Relationships, 1996; 13:279–294.

51.

Kenny

, Kashy

, Cook

. (2006). Dyadic data analysis. New York: Guilford.

52.

Aiken

, West

. (1991). Multiple regression: testing and interpreting interactions. London: Sage.

53.

Hayes

. A primer on multilevel modeling. Human Communication Research, 2006; 32:385–410.

54.

Christofides

, Muise

, Desmarais

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

55.

Kolek

, Saunders

. Online disclosure: an empirical examination of undergraduate Facebook profiles. Journal of Student Affairs Research & Practice, 2008; 45:1–25.

56.

John

. Sharing and Web 2.0: the emergence of a keyword. New Media & Society Online First, 2012.

57.

Nosko

, Wood

, Molema

. All about me: disclosure in online social networking profiles: the case of Facebook. Computers in Human Behavior, 2010; 26:406–418.

58.

Altman

, Taylor

. (1973) Social penetration: the development of interpersonal relationships. New York: Holt, Rinehart, and Winston.

59.

Hendrick

. Self-disclosure and marital satisfaction. Journal of Personality & Social Psychology, 1981; 40:1150–1159.

60.

Meeks

, Hendrick

. Communication, love, and relationship satisfaction. Journal of Social & Personal Relationships, 1998; 15:775–773.

61.

Weidler

, Clark

. A distinct association: inclusion of other in the self and self-disclosure. The New School Psychology Bulletin, 2011; 9:34–44.

62.

O'Sullivan

. (2005) Masspersonal communication: rethinking the mass-interpersonal divide. Paper presented at a meeting of the International Communication Association, New York.

63.

Petronio

. (2002) Boundaries of privacy. Albany: State University of New York Press.

64.

Carpenter

. Narcissism on Facebook: self-promotional and anti-social behavior. Personality & Individual Differences, 2012; 52:482–486.

65.

Ramirez

, Walther

, Burgoon

, et al. Information-seeking strategies, uncertainty, and computer-mediated communication: toward a conceptual model. Human Communication Research, 2002; 28:213–228.

66.

Berger

, Bradac

. (1982) Language and social knowledge. London: Edward Arnold.

67.

Sunnafrank

. Predicted outcome value during initial interactions: a reformulation of uncertainty reduction theory. Human Communication Research, 1986; 13:3–33.

68.

Burgoon

, Bonito

, Ramirez

, et al. Testing the interactivity principle: effects of mediation, propinquity, and verbal and nonverbal modalities in interpersonal interaction. Journal of Communication, 2002; 52:657–677.

69.

Haythornthwaite

. Social networks and internet connectivity effects. Information, Communication, & Society, 2005; 8:125–147.

70.

Hampton

, Goulet

, Marlow

, et al. (2012) Why most Facebook users get more than they give: the effect of Facebook power users on everybody else. http://pewinterent.org/reports/2012/Facebook-users.aspx (accessed Apr. 17, 2012).

71.

Marche

. (2012) Is Facebook making us lonely?. www.theatlantic.com (accessed Jun. 1, 2012).

72.

Turkle

. (2011) Alone together: why we expect more from technology and less from each other. New York: Basic Books.

73.

Rainie

, Wellman

. (2012) Networked: the new social operating system. Cambridge, MA: MIT Press.

74.

Tufekci

. (2012) Social media's small, positive role in human relationships. www.theatlantic.com (accessed Jun. 1, 2012).