Contextualizing how teens manage personal and interpersonal privacy on social media

Abstract

Many researchers have been studying teens’ privacy management on social media, and how they individually control information. Employing the theoretical framework of communication privacy management (CPM) theory, I argue that individual information control in itself is desirable but insufficient, giving only a limited understanding of teens’ privacy practices. Instead, I argue that research should focus on both personal and interpersonal privacy management to ultimately understand teens’ privacy practices. Using a survey study (n = 2000), I investigated the predictors of teens’ personal and interpersonal privacy management on social media and compared different types of boundary coordination. The results demonstrate that feelings of fatalism regarding individual control in a networked social environment, which I call networked defeatism, are positively related with interpersonal privacy management. Also, interpersonal privacy management is less important when coordinating boundaries with peers than it is when coordinating sexual materials, and dealing with personal information shared by parents.

Keywords

communication privacy management networked defeatism privacy management sexting sharenting social media teens

Advances in information and communication technologies and the widespread use of social media directed scholarly attention toward investigating users’ privacy and control over personal information. Over the years, much research has looked at how social media users employed or disregarded the service providers’ privacy settings and other ways to individually control privacy (see Baruh et al., 2017). Marwick and boyd (2014) and boyd (2008, 2011, 2014) have argued for moving beyond personal privacy management or individually controlling information: “any model of privacy that focuses on the control of information will fail [. . .] Control assumes that people have agency, or the power to assert control within a particular situation” (boyd, 2011: 349). Privacy is not fixed, nor achieved; rather, it should be seen as a process that is dependent on others and on the context wherein one is active.

In contemporary scholarship, a growing number of authors investigate privacy as boundary coordination, taking into account both the networked context and other actors in the privacy negotiation process (e.g. De Wolf, 2016; Hargittai and Marwick, 2016; Lampinen, 2016; Litt, 2013; Marwick and boyd, 2014; Quinn and Papacharissi, 2018). Because of its attention to the specific context of privacy, much research in this area is qualitative in nature. For example, in an ethnographical study boyd (2014) found how one of the teens she interviewed shared much personal information online and argued how “the appearance of unlimited sharing [on social media] allows them [referring to young users] to achieve privacy meaningfully” (p. 75). Indeed, by sharing information herself, the teenager negotiated the social context and made sure that people were less attentive to the personal information she evaluated as deeply private.

To the best of my knowledge, a quantitative study focused on analyzing and comparing the predictors of personal privacy management (individually managing privacy) and interpersonal privacy management (managing privacy together with others) on social media has not been done. Such a study, however, is important to offer a more holistic and complete view on privacy and to challenge the myth of teenagers being reckless and unconcerned. As a result of these factors, “What are the predictors of personal and interpersonal privacy management on social media?” (RQ1) is the first research question in this study. To take into account the networked context, I also introduce the concept of networked defeatism—that is, being fatalistic toward individually controlling information in a networked social environment—and study its relationship with privacy management. Finally, to further our understanding of teenagers’ privacy management, I also look into different types of boundary coordination. Specifically, I investigate how teenagers manage interpersonal privacy when sexting (RQ2), and how they negotiate parent’s sharenting behavior (RQ3). Sexting describes mediated sexual practices and when performed by teens ignites concerns among parents and the broader society. De Ridder (2019: 3) previously argued, echoing Goffman (1963), how “well-intentioned pedagogical advice” from educators, such as making oneself unrecognizable when sharing sexually explicit photos or refraining from the behavior altogether, also has the side effect of contributing to a discourse of sexting as a “stigma symbol.” In addition, I would like to add how this also reinforces privacy management as an individual responsibility rather than underlining the communicative system of privacy management and making agreements with those who send/receive sexual materials. Sharenting, a practice where parents share information about their children on social media, concerns another type of boundary coordination that is understudied. Whereas much attention has been devoted to how and why parents decide to share the information of their children (Blum-Ross and Livingstone, 2017; Brosch, 2016), the voices of teenagers in this process are mostly absent (Ouvrein and Verswijvel, 2019).

Before formulating the specific hypotheses connected to these research questions, I first further substantiate my position that privacy management on social media should be seen as a negotiation process dependent on others, information type, and the networked context of social media.

Boundary coordination and contextual privacy

Altman (1975) underlined privacy management as a social process and focused on how people manage personal, interpersonal, and group privacy boundaries. People are understood to have recourse to behavioral mechanisms, such as closing a door, agreeing on subjects that can be discussed in various contexts, and so forth, in order to regulate self-disclosure. Petronio (2002) built upon Altman’s framework and developed the communication privacy management (CPM) theory. This theory suggests that privacy management is a boundary coordination process of private information. Once private information is shared, others become co-owners of that information. Hence, personal and interpersonal boundaries need to be negotiated, and mutually agreed-upon privacy rules are required. In the context of hospitality exchange services (specifically couchsurfing), Lampinen (2016) researched how various privacy rules were negotiated. For example, hosts waited to discuss private issues when guests were out or at night. Similarly, De Wolf (2016) examined how various rules were discussed and negotiated in the context of youth organizations. For example, before posting pictures publicly on Facebook, some youth organizations organized a slide show evening with parents and children to collaboratively agree which pictures could and could not be posted.

Considering the dynamics and affordances of social media, personal privacy management in itself seems pointless, constructs a false sense of security, and even contributes to a process of individual responsibilization. Although individual control is important, the affordances and the open nature of social media challenge the privacy management of users. boyd (2011) distinguishes four affordances of social media: persistence, replicability, scalability, and searchability. Online content sticks, is easy to copy, has the potential to be seen by many, and is easy to search through. Papacharissi and Gibson (2011) added a fifth affordance: shareability, which affords the social action of sharing instead of withholding personal information. Together, these affordances introduce new online dynamics, such as imagined audiences and a context collapse (boyd, 2008; Litt, 2012; Vitak, 2012). Instead of labeling the privacy practices of individuals as paradoxical—because of a discrepancy between privacy attitudes and behavior—recent work indicates the existence of fatigue when it comes to online privacy management because “[. . .] the ability of individuals to control the spread of their personal information is compromised by both technological and social violations of privacy” (Hargittai and Marwick, 2016: 3572). The study of fatigue or defeatism, however, is very rare. Its role, however, should not be underestimated. Choi et al. (2018) found a stronger impact of privacy fatigue on privacy behaviors than privacy concern “which is widely regarded as the dominant factor in explaining online privacy behavior” (p. 42).

Personal and interpersonal privacy management is not only dependent on the networked context but also on the type of information that is disclosed and on the relationships that are involved (Petronio, 2002). According to a narrow interpretation, sharing any personal information on social media could be equated with giving up one’s privacy. However, this view impedes the development of a networked self and underestimates how individuals can achieve privacy meaningfully when negotiating the social context (see the example in the ‘Introduction’ section). Besides investigating the management of personal information in general, I also find it necessary to clearly delineate the type of information and relationships involved. We do not only manage privacy with peers but manage multiple boundaries around the self and the groups we belong to (Altman, 1975). In this study, I do not tackle all the information types and relationships; instead, I focus on interpersonal privacy management when distributing sexual materials and on how teens negotiate parents sharing information about them in the context of social media. Arguably, sexting and sharenting have caught the attention of scholars in the last few years. However, a systematic study of privacy management with respect to sexting and sharenting, both on a personal and an interpersonal level, is missing.

Predictors and hypotheses

Throughout this section, I will substantiate the different hypotheses while relying on the CPM theory. This theory was developed with face-to-face interactions in mind, but fundamental principles also apply in the context of social media (see Child and et al., 2011; De Wolf et al., 2014; Lampinen, 2016; Petronio, 2016).

Various criteria influence the privacy rules that are developed. Age and gender can be considered core criteria that remain more or less stable as opposed to catalyst criteria (e.g. emotional mood or sharing less information when being angry or disappointed). The first hypothesis is focused on age. According to the CPM theory, the privacy management process becomes more complex as individuals grow older (Petronio, 2002). Through socialization processes, an individual gradually learns a code of behavior as well as the act to balance the disclosing and withdrawing of private information. In the context of social media, young people are often accused of not caring about their privacy. Barnes (2006), for example, argued how “adults are concerned about invasion of privacy, while teens freely give up personal information.” This statement, however, has been debunked by highlighting the role of digital skills (Hargittai and Litt, 2013); the role of service providers that challenge the privacy management process by frequently changing their policies (Stutzman et al., 2013); and the importance of sharing intimate and personal information for teens’ development of the self (boyd, 2014). Hence, “freely giving up information” is exaggerated and over-simplified. Contrarily, research has also found that younger people are even more likely as opposed to older generations to protect their privacy on social media (Blank et al., 2014; Litt, 2013; Madden et al., 2013). Discussing the management of boundaries together with others, De Wolf et al. (2014) found how in youth organizations younger people were more likely to apply group privacy management strategies in Facebook than older people. However, when controlled for the role, one is attributed within one’s youth organization, this relationship disappeared. Hence, group variables were more important than age in explaining variance in group privacy. Because of conflicting evidence on how age is related to privacy management and the scant research on interpersonal privacy management, I do not specify their relationship. Based on the scholarship discussed, I present the following two hypotheses:

H1a. Age is related to personal privacy management.

H1b. Age is related to interpersonal privacy management.

Gender is another core criterion that defines the development of privacy management. In the process of developing the self, people learn the social world and the culture they are part of (Mead, 1934). One of the basic things a child learns is that our society is gendered, and that men and women are treated differently. In turn, this differentiation also determines the process of boundary coordination according to Petronio (2002). More specifically, sex-role identity and gender expectations are found to define the development of privacy rules (Petronio, 2002). The literature on privacy management and gender in the context of social media mostly indicates that women are stricter in their privacy management (Child and Starcher, 2017; De Wolf et al., 2014; Hoy and Milne, 2010; Litt, 2013; Quinn and Papacharissi, 2018). I hypothesize that girls are stricter in both their personal and interpersonal privacy management than boys as the following:

H2a. Girls apply more personal privacy management strategies than boys.

H2b. Girls apply more interpersonal privacy management strategies than boys.

When people have experienced privacy turbulence or when privacy boundaries are violated, the CPM theory argues that people will re-negotiate privacy rules and recalibrate boundaries. Privacy turbulence can range from minor issues to full breakdowns. Because of the open and fluid social environment of social media, including context collapses (Vitak, 2012) and invisible audiences (Litt, 2012), the occurrence of a breakdown or turbulence seems likely. DeGroot and Vik (2017) previously found that the lack of explicit privacy rules was likely to cause privacy turbulence. However, previous research also found behaviors of correcting, updating, and fine-tuning privacy rules after experiencing privacy disturbances (Child and Starcher, 2017; Child et al., 2011). I expect a positive relationship between experienced privacy turbulence and both personal and interpersonal privacy management.

H3a. Teens who have experienced privacy turbulence apply more personal privacy management strategies.

H3b. Teens who have experienced privacy turbulence apply more interpersonal privacy management strategies.

Besides turbulence, privacy concern is another important privacy predictor and can be defined as an individual’s “belief about the risks and potential negative consequences associated with sharing information” (Baruh et al., 2017: 27). The CPM theory indicates that stricter privacy rules are employed when having higher concerns (Petronio, 2002). Initial research indicated a discrepancy between privacy concerns and behaviors (Acquisti and Gross, 2006), which has been labeled as the privacy paradox (Barnes, 2006). However, most research nowadays seems to agree that a higher privacy concern often results in stricter usage of protective measures (Baruh et al., 2017; Millham and Atkin, 2018). Previous research, for example, has found how privacy concerns are positively related to removing data from commercial databases (Son and Kim, 2008) or deleting cookies (Lutz and Strathoff, 2014). Therefore, as a fourth hypothesis, I assume that privacy concerns are positively related to both personal and interpersonal privacy management strategies:

H4a. Teens with a greater concern for privacy apply more personal privacy management strategies.

H4b. Teens with a greater concern for privacy apply more interpersonal privacy management strategies.

Few studies have explored the role of privacy fatigue or defeatism. The CPM theory discusses different reasons why boundary coordination might not work (e.g. intentional rule violations or fuzzy boundaries). Because CPM was developed with face-to-face interactions in mind, the role of privacy fatigue or defeatism is obviously not included. Recent scholarship distinguishes different types of fatigue, such as security (becoming exhausted with security options), consent (becoming exhausted with reading privacy policies), and breach fatigue (becoming exhausted with data breaches) (Choi et al., 2018). In this particular study, I investigate the role of being fatalistic toward individually controlling personal information in a networked social environment, which I call networked defeatism, and assume that the latter phenomenon influences the privacy management process. Previous research on the topic found that users with higher privacy fatigue tend to put less effort into personal privacy management (Choi et al., 2018; Stanton et al., 2016). Therefore, this study hypothesizes its relationship with personal privacy management as follows:

H5a. Networked defeatism is negatively related to personal privacy management.

An inverse relationship, however, is expected with interpersonal privacy management. I consider networked defeatism as an obstacle for developing and exercising privacy on an individual level. However, I expect teens’ feelings of networked defeatism to positively influence the establishment of collective borders. Interpersonal privacy management is not only dependent on the choices of an individual alone but also assumes collaboration and negotiation. When people perceive that privacy boundaries are difficult to maintain on an individual level, I expect the social component to take the upper hand, and I assume that teens who have a higher score on networked defeatism to be active and rely on one another to manage boundaries as follows:

H5b. Networked defeatism is positively related to interpersonal privacy management.

The way privacy rules develop varies depending on the kind of relationship and information that is coordinated (Petronio, 2002). Teens’ producing, consuming, and distributing of intimate and sexual or erotic materials via their mobile phone, often called sexting, has caught the attention of scholars over the past few years. Most research found sexting to be a marginal phenomenon, with no major differences between boys and girls with regard to the likelihood to send sexts (Lenhart, 2009; Van Ouytsel et al., 2014; Vanden Abeele et al., 2014). Sending sexual materials is often confined within the boundaries of a romantic relationship (Vanden Abeele et al., 2014) and, in the context of social media, should be regarded as a negotiation of values and norms about sexuality (De Ridder, 2017). Hasinoff and Shepherd (2014) further show how respecting another’s privacy when sexting is seen as the expected social norm. On the contrary, what is worrisome, and a clear violation of one’s privacy for that matter, is the nonconsensual forwarding of sexts to the third parties. A recent meta-analysis by Madigan et al. (2018) found how up to 12% of teens forwarded a sext without the consent of the original sender, and 8.4% received a forwarded sext without consent. A systematic study of the privacy management strategies for sexual materials remains, I believe, to be carried out.

Based on previous literature and the CPM theory, I assume that teens are strict in their negotiation of sexual materials and privacy management. To use the CPM vocabulary, the levels of boundary permeability will be more closed, and the information is more likely to be considered secret. Because of the sensitivity of sexual content and the limited efficacy of individual control once information is disclosed on social media, I also expect interpersonal privacy management to be higher for sexual material compared to personal privacy management:

H6a. Teens’ personal privacy management when sending sexually explicit information is stricter than their general personal privacy management on social media.

H6b. Teens’ interpersonal privacy management when sending sexually explicit information is stricter than their general interpersonal privacy management on social media.

H6c. Teens rely more on interpersonal privacy management when sexting than on personal privacy management.

In the last few years, the phenomenon of sharenting has caught the attention of scholars. Blum-Ross and Livingstone (2017: 111) indicate that parents do not freely give up just about everything about their children but carefully craft “their story.” Parents not only share for reasons of self-presentation but also feel responsible toward their children (e.g. showing that they are committed parents) and other parents (e.g. showing best practices). However, research also indicates that sometimes inappropriate or too personal content is shared (Brosch, 2016). Other research shows how different preventive and corrective strategies are used to control information that concerns the child, such as asking fellow family members not to post baby pictures without their consent or asking them to delete certain information (Ammari et al., 2015). That said, the knowledge on sharenting and the negotiating process between teens and parents is limited. Most research is focused on parental motivations to share posts and pictures of their children and only covers the parents’ perspective on the matter (Blum-Ross and Livingstone, 2017; Brosch, 2016; Ouvrein and Verswijvel, 2019). I assume that the interpersonal privacy management between parents and teens concerning parents’ sharenting behavior is stricter than the interpersonal privacy management with peers on social media. Employing the CPM theory, the former can be labeled as a type of inclusive boundary coordination, where person A (here: the parent) manages more of person B’s (here: the teenager’s) privacy than person B does for person A. Arguably, in the context of social media, considering the crucial role that social media plays in the lives of teens since the early 2000s (boyd, 2014), privacy management between peers is more routinized than between parents and teens:

H7. Teens’ interpersonal privacy management concerning parents’ sharenting behavior is stricter than the general interpersonal privacy management with peers on social media.

Method

Procedure

This study was part of a larger survey that focused on teens’ media ownership and media usage in Flanders (northern part of Belgium). In addition, the study also raised questions about news literacy, digital citizenship, privacy management, and other societal themes. The data were collected in November 2017 by a non-profit association dedicated to letting young people engage with digital media. Various schools were recruited and asked to distribute an online survey to their students. In total, 12 schools participated in this study across all the regions in Flanders. The survey was filled out during class hours. After data cleaning, 2681 valid responses remained. I deleted the cases with an unusual duration (duration > 100 minutes; duration < 5 minutes), as well as the responses by those who did not complete the survey or were of a higher age than 21. To increase the representativeness of the sample, I weighted the cases by gender, age, and education.¹ In the end, I had a weighted sample of 2000 cases. Participants ranged from 11 to 21 (M = 14.94, SD = 1.85), with 1017 boys (51%) and 983 girls (49%).

To answer research question 1 and hypotheses 1 to 5, I employed hierarchical multiple regression, which allows to test the effects of predictors after accounting for all other variables. I differentiate between three sets of predictor variables: demographics, privacy, and networked variables. No violations were found when testing the assumptions of linear regression. Specifically, I checked for linear relationships, multivariate normality, and the absence of multicollinearity, autocorrelation, and heteroscedasticity. To answer research questions 2 and 3 and the subsequent hypotheses, I made use of paired sample t-tests.

Measures

In all the scales mentioned, the respondents were asked to rate items on a 7-point Likert-type scale. The scales ranged from “strongly disagree” to “strongly agree.” I used an exploratory factor analysis (with varimax rotations) in the construction of the scales to test the internal structure. For each scale I mention the reliability (Cronbach’s alpha, α), mean (M), and standard deviation (SD). All items, including those that were left out after the dimension reduction, are listed in the Appendix 1.

Personal privacy management represents one of the two main dependent measures in this study. To operationalize this measure, I used a modified version of De Wolf et al.’s (2014) individual privacy management on Facebook. For example, I asked the respondents if they were careful when accepting friend requests (e.g. “I am careful when accepting friend requests”). The analysis extracted only one factor, consisting of five items, which accounts for 46.92% of the total variance with factor loadings from .5 to .798 (α = .76, M = 4.81, SD = 1.29).

Interpersonal privacy management represents the other main dependent measure. I adapted and included items from the scales developed by Stutzman and Kramer-Duffield (2010) and De Wolf et al. (2014). For example, I asked the respondents if they talk with their friends about what can and cannot be shared (e.g. “My friends and I talk about what can and cannot be shared”). The factor analysis differentiates two factors, which together account for 76.65% of the variance. Factor 1 accounts for 58.73% of the variance with factor loadings from .734 to .8345. Factor 2 only accounts for 17.92% of the variance and was left out from further analyses. After the dimension reduction, the scale consists of three items, with α = .87, M = 3.71, and SD = 1.51.

Privacy concern and privacy turbulence are the other two privacy-related variables and are included as independent measures in the regression models. To measure privacy concern, I formulated items similar to the privacy concern scale of Xu et al. (2011) (e.g. “I am worried that online information will be misused”). In the factor analysis, only one factor was extracted for privacy concern, which accounts for a total variance of 77.42% with factor loadings from .76 to .777 (α = .90, M = 4.23, SD = 1.44).

DeGroot and Vik (2017) conducted a qualitative study and categorized different types of privacy turbulence (e.g. pre-emptive disclosure violations and discrepancy breaches). Here I asked the respondents to indicate whether or not they experienced one of these turbulent events. A minority mentioned not having experienced any turbulence (46%). In the regression models, this variable is included as a dummy variable (experienced turbulence = 1, no previous experience of turbulence = 0).

To measure networked defeatism, I asked about the respondents’ sense of individual control over their privacy on social media. For example, I asked the respondents whether or not it is impossible, as an individual, to control your personal information by yourself (e.g. “As an individual you have little control over online privacy”). One factor was extracted and accounts for 63.98% of the variance with factor loadings from .637 to .799. The scale consists of five items and has high internal reliability (α = .86, M = 3.65, SD = 1.22).

To answer research questions 2 and 3, I asked about the interpersonal and personal privacy management regarding sexually explicit information and whether and how teenagers negotiate parents’ sharenting behaviors with their parents. For example, I included statements such as “I have agreements with those who receive my sexual materials” for interpersonal privacy management regarding sexually explicit information and statements such as “My parents and I have agreements on what information can be shared about me online” for sharenting interpersonal privacy management. A total of 235 respondents, 12.10% of the sample, indicated to have sexted at least once during the past 2 months. A total of 885 respondents, 44.25% of the sample, indicated their parents to be active on social media and share personal information that concerns them. For sexting personal privacy management, only one factor was extracted, comprised of three items, which explained 69.73% of the variance with factor loadings from .773 to .852 (α = .85, M = 4.5, SD = 1.74). The factor analysis revealed a one-factor solution for sexting interpersonal privacy management consisting of three items, which explained 81.32% of the variance with factor loadings from .795 to .917 (α = .88, M = 4.50, SD = 1.75). For sharenting interpersonal privacy management, only one factor was extracted, with an explained variance of 74.42% and with factor loadings from .701 to .848, consisting of three items (α = .83, M = 4.02, SD = 1.59).

Results

Table 1 and Table 2 present a summary of the hierarchical regression analyses for variables predicting personal and interpersonal privacy management on social media. H1a, b asked about the relationship between age and privacy management. A positive relationship was found between age and personal privacy management (β = .101, p < .001) (H1a). An inverse relationship, however, was found for interpersonal privacy management (β = −.07, p < .01) (H1b). Older teens are thus less likely to make use of interpersonal privacy management strategies than younger ones, whereas younger teens are less likely to make use of personal privacy management strategies. The positive gender coefficient in both tables indicates that girls apply more personal and interpersonal privacy management strategies than boys (β = .211, p < .001; β = .085, p < .001) (H2a, b). When considering the privacy variables in the models, I see an inverse relationship between the experience of privacy turbulence and personal privacy management (β = −.082, p < .001) (H3a). No significant relationship was found between the experience of privacy turbulence and interpersonal privacy management (β = −.03, p > .05) (H3b). Privacy concern is the strongest predictor in the models for both personal (β = .26, p < .001) and interpersonal privacy management (β = .23, p < .001) (H4a, b). Finally, no negative relationship was found between networked defeatism and privacy management on a personal level (β = .013, p > .05) (H5a). However, respondents who were more fatalistic toward individually controlling information in a networked social environment are more likely to negotiate privacy boundaries with others (β = .153, p < .001) (H5b).

Table 1.

Summary of hierarchical regression analysis for variables predicting personal privacy management on social media.

	Model 1		Model 2		Model 3
	B (SE)	β	B (SE)	β	B (SE)	β
Demographics
Women	.66 (.056)***	.256***	.541 (.055)***	.21***	.544 (.055)***	.211***
Age	.062 (.015)***	.089***	.071 (.015)***	.10***	.07 (.015)***	.101***
Privacy variables
Turbulence			−.21 (.55)***	−.081***	−.212 (.055)***	−.082***
Concern			.231 (.019)***	.259***	.228 (.02)***	.255***
Networked variables
Networked defeatism					.014 (.023)	.013
Constant	3.560 (.229)***		2.619 (.233)***		2.591 (.238)***
R ²	.07		.141		.141

SE: Standard Error.

***

p < .001.

Table 2.

Summary of hierarchical regression analysis for variables predicting interpersonal privacy management on social media.

	Model 1		Model 2		Model 3
	B (SE)	β	B (SE)	β	B (SE)	β
Demographics
Women	.371 (.067)***	.123***	.214(.066)**	.071**	.256 (.065)***	.085***
Age	−.052 (.018)**	−.064**	−.049(.018) **	−.060**	−.057 (.017)**	−.07**
Privacy variables
Turbulence			−.065(.066)	−.021	−.089 (.065)	−.03
Concern			.288 (.023)***	.276***	.241 (.024)***	.231***
Networked variables
Networked defeatism					.191 (.028)***	.153***
Constant	4.314 (.276)***		3.159 (.281)***		2.777 (.283)***
R ²	.02		.092		.112

SE: Standard Error.

p < .01; ***p < .001.

Only 235 respondents, or 12.10% of the sample, sent at least one sexually explicit image to others in the last 2 months. A total of 62.80% claimed to be unrecognizable when sending sexts. A total of 30.20% girls and 41.20% boys were mentioned being naked in those pictures. A significant difference was found between personal privacy management in general (M = 4.73, SE = .09) and personal privacy management for sexts (M = 4.50, SE = .11), t (234) = −2.016, p = .045 (H6a). A strong difference was found between interpersonal privacy management in general (M = 3.68, SE = .10) and interpersonal privacy management for sexting (M = 4.53, SE = .11); t (232) = 6.592, p = .000 (H6b). Finally, no significant difference was found between personal (M = 4.50, SE = .11) and interpersonal privacy management (M = 4.53, SE = .11) when sexting; t (232) = .242, p = .809 (H6c).

A total of 885 of the respondents, 44.25% of the sample, indicated their parents to be on social media and share information about them. A total of 79.80% stated that Facebook was the prime medium to do this. Most teens responded negatively to their parents asking them for permission (58.40%). A paired sample t-test was conducted to compare interpersonal privacy management in general and interpersonal privacy management when negotiating parents’ sharenting behaviors. In line with previous results on sexting, I found negotiating parents’ sharenting behaviors to be higher (M = 4.03, SE = .05) than general interpersonal privacy management with peers (M = 3.74, SE = .05); t (884) = 4.663, p < .0001 (H7).

Discussion

In everyday discourse, privacy on social media is often equated with individual information control and access restriction. Although personal control is desirable and suggested, it can also be misleading and one-sided. In a networked social environment, there is only so much an individual can do to preserve their privacy (Hargittai and Marwick, 2016). Building further on the CPM theory, I conceptualized privacy as a boundary coordination process in which privacy rules or strategies are developed to coordinate private information with others. In this study, I investigated and compared the predictors of interpersonal and personal privacy management on social media and studied the boundary coordination processes of teens with friends, with parents concerning their sharenting behavior, and with persons they exchange sexual materials with. Overall, my study contributes to the scholarship on online privacy management by contextualizing the privacy management process, and, specifically, by taking into account the networked context and certain types of information and relationships. To the best of my knowledge, no previous study has compared the predictors of personal and interpersonal privacy management, looked at the relationship between networked defeatism and interpersonal privacy management, or studied interpersonal privacy management when sexting or sharenting.

Similarly to existing research, I found that girls are stricter in their privacy management than boys (Child and Starcher, 2017; De Wolf et al., 2014; Quinn and Papacharissi, 2018). In addition, age was positively related to managing privacy individually. However, a negative relationship was found between age and interpersonal privacy management. In general, the CPM theory argues that teens’ “boundaries expand to accommodate the increasing privacy needs that he or she develops” (Petronio, 2002: 8). The results indicate that in this process teens increasingly rely on individual privacy management strategies rather than collaboratively managing privacy. Furthermore, I found evidence for privacy concern to be the most important factor in explaining the privacy management of teens, both at the personal and the interpersonal level, in line with the meta-analysis of Baruh et al. (2017). Recently, various authors have suggested the networked social environment to be influential in the privacy management of teens (Choi et al., 2018; Hargittai and Marwick, 2016). I took this point into account by including a variable that measured feelings of fatalism toward individually controlling personal information on social media. No significant relationship was found with personal privacy management. Hence, no proof was found that being more or less fatalistic is explanatory for managing privacy on an individual level, as opposed to the results of Choi et al. (2018) and Stanton et al. (2016). However, I did find a positive relationship with interpersonal privacy management. These results indicate that when feeling more fatalistic toward individual control in a networked social environment, teenagers are more likely to negotiate and make agreements with others in their network.

Looking closer at teens’ interpersonal privacy management, it is especially noticeable that teens are more prone to managing privacy with others when sharing sexually explicit information and when parents share information that is of their concern as opposed to interpersonal privacy management with friends in general. I assume that general privacy management processes with friends are more routinized and implicit. Unfortunately, in everyday discourse sexting is regularly labeled as risky and/or “unnatural sexualized passions of the young” (see De Ridder, 2019: 2). Previous research showed how sexting is often confined within the boundaries of romantic relationships (Vanden Abeele et al., 2014) and how the practice can be seen as an exploration and development of one’s sexuality (De Ridder, 2017), thereby questioning the supposedly risky and harmful behavior. In addition, this research shows how teens are rational in their sexting behavior and develop thicker boundaries when managing sexually explicit information with others. As opposed to the formulated hypothesis, no significant difference was found between interpersonal and personal privacy management when sexting, which shows the importance of both individual control and negotiating boundaries.

Similar results were found for teens’ negotiation of parents’ sharenting behavior: teens are stricter when managing boundaries with parents than with peers. From a CPM theory perspective, I consider interpersonal privacy management between teens and parents to be a specific type of boundary coordination that is understudied. Most research concerning sharenting investigates the motivations of parents for sharenting, and how they consider (or not) teens’ privacy in this process (Blum-Ross and Livingstone, 2017; Brosch, 2016). Indeed, the voice of teenagers is mostly missing. By investigating the negotiation of privacy between teens and parents, I acknowledge the agency of teens in the privacy management process.

The theoretical contribution of this study is threefold. First, in agreement with the CPM theory, core and catalyst criteria were found to influence the development of privacy rules. Second, the results show how privacy-rules development varies depending on the relationship and the information type. Third, by taking into account networked defeatism and its relationship with interpersonal privacy management, I show the importance of social media in the development of privacy rules, including interpersonal privacy management as part of the larger privacy management process. Specifically, when narrowly conceptualizing privacy as individual control and exploring the role of privacy fatigue or networked defeatism, it could be argued that young people are fatalistic in their privacy management when finding significant negative relationships (e.g. Choi et al., 2018; Stanton et al., 2016). However, as the results of my study show, this is not necessarily true when considering privacy management as managing privacy together with others. Overall, I provide empirical support for Petronio’s CPM theory and the communicative system of privacy management.

As opposed to the hypothesis and in disagreement with the CPM theory, the data indicate a negative relationship between experiences of turbulence and personal privacy management. No significant relationship was found with interpersonal privacy management. Building further on DeGroot and Vik (2017), I presented the respondents with many different types of privacy turbulences. In addition to whether or not the respondents encountered one of these turbulent events, future research might also focus on the severity and recency of the event declared by the respondent to further examine the relationship between turbulence and privacy management. Researchers typically include privacy-related variables (such as experience of privacy turbulence or privacy concern) when investigating privacy management. In my research, I found networked defeatism to be an import predictor for interpersonal privacy management. I recommend future research to take into account other context-related variables such as security, consent, or breach fatigues (see Choi et al., 2018). The focus in this research was on interpersonal privacy management and the negotiation of personal information. Future research, however, might also look at the predictors of interpersonal privacy management for sexting or sharenting to further understand what explains teens’ privacy behaviors. Finally, the research method allowed me to only measure an individual teen’s point of view. To have more valid results and acknowledge the communicative system of privacy management, other actors (e.g. parents and romantic partners) should be included as well. For example, many teens indicated that their parents do not ask for permission when sharenting (58.4%). To have a more holistic perspective, a similar question could be asked of their parents.

Footnotes

Appendix 1.

Items (7-point Likert-type scale)	M	SD
Personal privacy management (The statements below inquire about your personal privacy management in the context of social media.)
- I adjusted my settings, so others have to ask permission when tagging me in a picture- I am careful when accepting friend requests- In general, I make use of privacy settings to manage my privacy- I untag myself from photos I do not find appropriate- I make sure only friends can see my profile on social mediaExcluded from further analyses after measuring internal consistency:- I make use of private messages when I talk about sensitive information	3.975.445.024.794.824.68	1.9621.6031.6491.8571.8021.817
Interpersonal privacy management(The statements below inquire how you negotiate personal information with others on social media.)
- In my friends group we have agreements on what can and cannot be shared- My friends and I talk about what can and cannot be shared- Before I share information, my friends and I discuss the appropriate privacy settingsExcluded from further analyses after dimension reduction:- I ask permission of my friends before I post information- My friends ask permission before they share information about me- I do not share information about others when they do not want this	3.783.933.434.524.235.35	1.731.7261.6351.7261.7261.745
Networked defeatism(The statements below inquire about individual control over privacy on social media.)
- As an individual you have little control over online privacy- Adjusting your personal privacy settings has little impact- It is impossible, as an individual, to control your personal information by yourself- It is impossible to control who shares information about you- In general, I have the feeling that I can exert little control over my personal information	3.623.43.693.963.59	1.611.4851.4811.531.501
Privacy concern(The statements below inquire how concerned you are about your online privacy.)
- I am worried that online information will be misused- I am worried that others will find private information about me- I am worried that online information might be used against me in ways I did not foresee- I am worried that others will post embarrassing information about me online	4.454.14.264.11	1.6541.6451.581.706
Sexting personal privacy management (The statements below inquire how you handle sexual tinted photos you have sent.)
- I secure my sexts so others cannot further distribute them without my permission- I make sure my sexts disappear automatically after a few seconds (e.g. selecting time frame in Snapchat)- In general, I control my privacy settings when sending sextsExcluded from further analyses after measuring internal consistency:- I make use of private messages when I send sexual materials about myself	4.34.934.364.62	1.9341.8721.9312.04
Sexting interpersonal privacy management (The statements below inquire about how you handle sexual tinted photos you have sent.)
- I have agreements with those who receive my sexual material- I talk with others who receive my sexual material on what can and cannot be shared- Those who receive my sexual material ask for my permission before forwarding this information	4.714.434.51	1.9091.9111.983
Sharenting interpersonal privacy management (To what extent do you agree with the statements below about your parents’ sharing behavior?)
- My parents ask for my permission before sharing information about me- My parents and I have agreements on what information can be shared about me online- My parents and I talk about the information they post about me	4.043.914.14	1.9141.8371.77

Acknowledgements

I wish to thank Bart Vanhaelewyn, Bas Baccarne and Florian Vanlee for their diligent proofreading. I also express my appreciation to the anonymous reviewers, whose constructive feedback greatly improved the article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Ralf De Wolf

Notes

Author biography

Ralf De Wolf (PhD) is a postdoctoral researcher at the Department of Communication Sciences and connected to the research group for Media, Innovation and Communication Technologies (imec-MICT), Ghent University, Belgium. His research focuses on the social aspects of online privacy and security, the presentation of self, and the social construction of algorithms.

References

Acquisti

Gross

(2006) Imagined communities: Awareness, information sharing, and privacy on the Facebook. In: Danezis

Golle

(eds) Privacy Enhancing Technologies. Berlin: Springer, pp. 36–58.

Altman

(1975) The Environment and Social Behavior: Privacy, Personal Space, Territory, Crowding. Monterey, CA: Brooks/Cole.

Ammari

Kumar

Lampe

, et al. (2015) Managing children’s online identities: how parents decide what to disclose about their children online. In: proceedings of the 33rd annual ACM conference on human factors in computing systems, Seoul, South Korea, 18–23 April, 2015.

Barnes

(2006) A privacy paradox: social networking in the Unites States. First Monday 11(9): 1–10.

Baruh

Secinti

Cemalcilar

(2017) Online privacy concerns and privacy management: a meta-analytical review. Journal of Communication 67(1): 26–53.

Blank

Bolsover

Dubois

(2014) A new privacy paradox: young people and privacy on social network sites. In: American sociological association annual meeting, San Francisco, CA, 17 August. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2479938

Blum-Ross

Livingstone

(2017) “Sharenting,” parent blogging, and the boundaries of the digital self. Popular Communication 15(2): 110–125.

boyd

(2008) Facebook’s privacy trainwreck: exposure, invasion, and social convergence. Convergence: The International Journal of Research into New Media Technologies 14(1): 13–20.

boyd

(2011). Networked privacy personal democracy forum. Available at: https://personaldemocracy.com/media/networked-privacy

10.

boyd

(2014) It’s Complicated: The Social Lives of Networked Teens. New Haven, CT: Yale University Press.

11.

Brosch

(2016) When the child is born into the internet: sharenting as a growing trend among parents on Facebook. The New Educational Review 43(1): 225–235.

12.

Child

Starcher

(2017) Fuzzy Facebook privacy boundaries: exploring mediated lurking, vague-booking, and Facebook privacy management. Computers in Human Behavior 54: 483–490.

13.

Child

Petronio

Agyeman-Budu

, et al. (2011) Blog scrubbing: exploring triggers that change privacy rules. Computers in Human Behavior 27: 2017–2027.

14.

Choi

Park

Jung

(2018) The role of privacy fatigue in online privacy behavior. Computers in Human Behavior 81: 42–51.

15.

De Ridder

(2017) Social media and young people’s sexualities: values, norms and battlegrounds. Social Media + Society 3: 1–11.

16.

De Ridder

(2019) Sexting as sexual stigma: the paradox of sexual self-representation in digital youth cultures. European Journal of Cultural Studies 1–16. DOI: 10.1177/1367549418810080.

17.

De Wolf

(2016) Group privacy management strategies and challenges in Facebook: a focus group study among Flemish youth organizations. Cyberpsychology: Journal of Psychosocial Research on Cyberspace 10(1): 5.

18.

De Wolf

Willaert

Pierson

(2014) Managing privacy boundaries together: exploring individual and group privacy management strategies in Facebook. Computers in Human Behavior 35: 444–454.

19.

DeGroot

Vik

(2017) “We were not prepared to tell people yet”: confidentiality breaches and boundary turbulence on Facebook. Computers in Human Behavior 70: 351–359.

20.

Goffman

(1963) Stigma: Notes on the Management of Spoiled Identity. London: Simon & Schuster.

21.

Hargittai

Litt

(2013) New strategies for employment? Internet skills and online privacy practices during people’s job search. IEEE Security & Privacy 11(3): 38–45.

22.

Hargittai

Marwick

(2016) “What can I really do?” Explaining the privacy paradox with online apathy. International Journal of Communication 10: 3737–3757.

23.

Hasinoff

Shepherd

(2014) Sexting in context: privacy norms and expectations. International Journal of Communication 8: 2932–2955.

24.

Hoy

Milne

(2010) Gender differences in privacy-related measures for young adult Facebook users. Journal of Interactive Advertising 10(2): 22–45.

25.

Lampinen

(2016) Hosting together via couchsurfing: privacy management in the context of network hospitality. International Journal of Communication 10: 1581–1600.

26.

Lenhart

(2009) Teens and Sexting: How and Why Minor Teens Are Sending Sexually Suggestive Nude or Nearly Nude Images Via Text Messaging. Washington, DC: Pew Internet & American Life Project. Available at: www.pewinternet.org/2009/12/15/teens-and-sexting/

27.

Litt

(2012) Knock, knock. Who’s there? The imagined audience. Journal of Broadcasting & Electronic Media 56(3): 330–345.

28.

Litt

(2013) Understanding social network site users’ privacy tool use. Computers in Human Behavior 29(4): 1649–1656.

29.

Lutz

Strathoff

(2014) Privacy concerns and online behavior—not so paradoxical after all? Viewing the privacy paradox through different theoretical lenses. Available at: https://ssrn.com/abstract=2425132

30.

Madden

Lenhart

Cortesi

, et al. (2013) Teens and Mobile Apps Privacy. Washington, DC: Pew Research Center. Available at: www.pewinternet.org/2013/08/22/teens-and-mobile-apps-privacy/

31.

Madigan

Rash

, et al. (2018) Prevalence of multiple forms of sexting behavior among youth: a systematic review and meta-analysis. JAMA Pediatrics 172(4): 327–335.

32.

Marwick

boyd

(2014) Networked privacy: how teenagers negotiate context in social media. New Media & Society 16(7): 1051–1067.

33.

Mead

(1934) Mind, Self and Society from the Standpoint of a Social Behaviorist. Chicago, IL: University of Chicago.

34.

Millham

Atkin

(2018) Managing the virtual boundaries: online social networks, disclosure, and privacy behaviors. New Media & Society 20(1): 50–67.

35.

Ouvrein

Verswijvel

(2019) Sharenting: parental adoration or public humiliation? A focus group study on adolescents’ experiences with sharenting against the background of their own impression management. Children and Youth Services Review 99: 319–327.

36.

Papacharissi

Gibson

(2011) 15 minutes of privacy: Privacy, sociality and publicity on social network sites. In: Reinecke

Trepte

(eds) Privacy Online Theoretical Approaches and Research Perspectives on the Role of Privacy in the Social Web. New York, NY: Springer, pp. 75–89.

37.

Petronio

(2002) Boundaries of Privacy: Dialectics of Disclosure. Albany, NY: State University of New York Press.

38.

Petronio

(2016) Communication privacy management theory. In: Jensen

Craig

Pooley

, et al. (eds) The International Encyclopedia of Communication Theory and Philosophy. Hoboken, NJ: Wiley-Blackwell, pp. 278–286.

39.

Quinn

Papacharissi

(2018) The contextual accomplishment of privacy. International Journal of Communication 12: 45–67.

40.

Son

Kim

(2008) Internet users’ information privacy-protective responses: a taxonomy and a nomological model. MIS Quarterly 32(3): 503–529.

41.

Stanton

Theofanos

Prettyman

, et al. (2016) Security fatigue. IT Professional 18(5): 26–32.

42.

Stutzman

Kramer-Duffield

(2010) Friends only: examining a privacy-enhancing behavior in Facebook. In Proceedings of the SIGCHI conference on human factors in computing systems, Atlanta, GA, 10–15 April. New York: ACM, pp. 1553–1562.

43.

Stutzman

Gross

Acquisti

(2013) Silent listeners: the evolution of privacy and disclosure on Facebook. Journal of Privacy and Confidentiality 4(2): 7–41.

44.

Van Ouytsel

Van Gool

Ponnet

, et al. (2014) Brief report: the association between adolescents’ characteristics and engagement in sexting. Journal of Adolescence 37(8): 1387–1391.

45.

Vanden Abeele

Campbell

Eggermont

, et al. (2014) Sexting, mobile porn use, and peer group dynamics: boys’ and girls’ self-perceived popularity, need for popularity, and perceived peer pressure. Media Psychology 17(1): 6–33.

46.

Vitak

(2012) The impact of context collapse and privacy on social network site disclosures. Journal of Broadcasting & Electronic Media 56(4): 451–470.

47.

Dinev

Smith

, et al. (2011) Information privacy concerns: linking individual perceptions with institutional privacy assurances. Journal of the Association for Information Systems 12(2): 798–824.