“Should I Stay or Should I Leave?”: Exploring (Dis)continued Facebook Use After the Cambridge Analytica Scandal

Abstract

Social media platforms bring both benefits and risks which have been documented copiously in extant academic literature. A range of issues related to privacy and trust inhibit the fulsome enjoyment of social media by users. In 2018, several news sources documented that Cambridge Analytica acquired psychographic data for Facebook users and used that data to target ads for the November 2016 US election. Although none of the news reports indicated that Facebook was complicit in this matter, some Facebook users publicly announced they would leave Facebook and encouraged others to do so. Using in-depth interviews with 10 undergraduate and graduate college students aged 18–29 years, this research study explores decisions to stay with or leave Facebook following the Cambridge Analytica case as such decisions intersect with privacy concerns. While all the respondents were concerned about their privacy, many of them believed that participation in social media requires an exchange of personal data for the use of the service. None of the respondents left Facebook permanently because of the Cambridge Analytica incident. But several reported non-use and reduced use prompted by privacy concerns and other social concerns associated with the use of Facebook. Although these research interviews are centered on a very specific event, they are instructive on the various approaches to privacy patterns and trust in social media. Greater transparency, advocacy, and transnational cooperation would be critical interventions to inspire greater trust in social media platforms.

Globally, 43% of the world’s population use social media (Kemp, 2018) and 69% of Americans use social media (Smith & Anderson, 2018). Increasingly, individuals around the world participate actively in political discourse and maintain their social connections using social media. With the spread of news and information, social media has also become a part of the global public sphere, a space where information-sharing and public debate occur (Carlsson, 2016, p. 12).

Given these characteristics of social media, it was disturbing for many users that social media was being used to influence public opinion in the 2016 elections (Hsu, 2018). In March 2018, a personality quiz app was used by Aleksandr Kogan to collect private data of the 300,000 users who chose to use the quiz app. The app also collected the Facebook data of an estimated 87 million users connected to those who used the app. This developer then sold the data that was mined to Cambridge Analytica. Cambridge Analytica employed this data in the delivery of targeted political advertisements on Facebook (US Senate, 2018).

Facebook CEO, Mark Zuckerberg, described the Cambridge Analytica situation as a “breach of trust” (US Senate, 2018). Facebook came under fire from the US government for not being proactive enough in stemming this privacy breach. Zuckerberg was required to testify before Congress on April 18, 2018 (US Senate, 2018). Although Facebook did not appear to be complicit with Cambridge Analytica, many Facebook users blamed the social media organization for the compromise of their data since Facebook was the social media platform that was associated with the breach (Lapowsky, 2018). Beyond that, the Cambridge Analytica Scandal demonstrated that the stakes of social media malpractice go beyond individual privacy invasion, to threats to citizen rights in a democratic political system (Hsu, 2018).

Some users reacted to the Cambridge Analytica scandal by stating they would leave Facebook and by encouraging others to do so. Many joined a Twitter campaign #deletefacebook. #Deletefacebook is not new. A Twitter search reveals the hashtag in tweets from as early as 2009. An upsurge in tweets containing the #deletefacebook hashtag in the wake of the Cambridge Analytica scandal represented a backlash to what some felt was the inability of Facebook to prevent privacy invasions (Hsu, 2018). In the months that followed the initial news reports of the Cambridge Analytica scandal, about 74% of users made some adjustments to their use of Facebook by adjusting privacy settings, taking a break from the service or deleting the app from their phones (Perrin, 2018).

Using in-depth interviews with 10 undergraduate and graduate students in an American college aged 18–29 years, this research study explores decisions to stay with or leave Facebook following the Cambridge Analytica case. Among the research questions are: What do respondents think about the Cambridge Analytica Scandal? What factors influenced the decision to stay with Facebook or leave Facebook following this event? What are the possible consequences of closing a Facebook account?

Self-Disclosure and Privacy on Social Media

Social media platforms allow individuals to selectively disclose information to online contacts ranging from general biographical information to photos, daily activities, feelings, opinions, and location. Self-disclosure is “any message about the self that an individual communicates to another” (Gibbs et al., 2006, p. 155). Not all social media users are inclined to disclose details about their lives. The literature on online self-disclosure shows that persons wanting to establish and build relationships will disclose more information than those who do not (Gibbs et al., 2006; Krasnova et al., 2010) and that disclosure is often to more distant contacts (Waters & Ackerman, 2011). There is an understanding among social media users that disclosure will lead to some anticipated payoffs such as relationship building, social support, and satisfaction of life (Huang, 2016).

Users who are preoccupied with privacy generally disclose less information (Krasnova et al., 2010). Such individuals are likely to limit self-disclosure on social media and only disclose very superficial information (Attrill & Jalil, 2011). Common privacy concerns include identity theft and deception, misuse of personal information, and government or corporate surveillance (Quinn, 2016). The most common privacy behavior used in response to these concerns is to engage system controls, including privacy settings (Quinn, 2016). When users have control over information with the use of privacy settings, they feel more secure and trust online platforms more (Taddei & Contena, 2013). Users who experience privacy invasions, such as being hacked on social media, getting unwanted advances, and being subjected to online harassment, are more likely to modify their privacy settings (Debatin et al., 2009). But there is some evidence of a “privacy paradox” in social media use where the people who are most concerned about privacy do not always utilize the privacy settings available on social media (Barnes, 2006).

Social media scholarship has considered the issues of self-disclosure and privacy at length. This study explores perspectives on privacy among young adults within the context of a concrete instance of a documented privacy breach. The study identifies specific privacy issues and concerns that are being raised by young persons in the aftermath of this event. While there have been several studies that consider privacy and social media practice, none of the studies identified have gathered perceptions on a specific incident that was widely discussed in the media. The context of a concrete privacy breach can illustrate the vulnerabilities for social media users that may be more abstract when presented with hypothetical situations. As such, this study addresses a gap in the literature, since it elucidates perceptions of young persons in relation to a specific social media privacy-related event.

Intentional Disconnection From Social Media

Challenges related to privacy, misinformation, and antisocial behavior sometimes cause dissatisfaction for social media users. A limited number of studies consider intentional non-use of social media and deliberate disconnection of services. There are many persons who choose to be Facebookless (Ongun & Güder, 2013). Facebooklessness is a new form of isolation caused by not having a Facebook account. The most common reason for Facebooklessness is lack of trust in Facebook (Ongun & Güder, 2013). Persons who actively leave Facebook do so for a range of reasons such as time-wasting, to reduce distraction, non-interest, privacy concerns, coping with a breakup, and to build a romantic partner’s trust (Dindar & Akbulut, 2014; Stieger et al., 2013).

The distinction between use and non-use of Facebook is murky and there are instead different “degrees of engagement” with the platform (Baumer et al., 2013). Some users choose to leave temporarily by deactivating their account (where the account can be later retrieved). Others leave permanently by deleting their account (where the account is no longer recoverable). Users often discontinue use because their motivations for joining are not met by the platform or because the platform is unable to meet user expectations and needs (Alam & Wagner, 2013). This “use-based distress” is directly linked to the specific gratifications sought by the user. For example, if a user wants to use Facebook to build relationships, they may want to leave if they experience privacy violations which would be a use-based distress in that instance (Alam & Wagner, 2013). The research on intentional disconnection from social media is limited. This study advances the literature on this topic by exploring the decisions to stay with or leave a social media service after a privacy breach.

Methods and Procedures

Interview Procedure

I gathered data using in-depth interviews which are useful for capturing detailed descriptions and respondent interpretations (Weiss, 1995). I employed convenience sampling and identified a “sample of representatives” (Weiss, 1995, Chapter 2, Section 2, para. 4). In particular, I targeted five male and five female participants so as to have equal representation across genders. I intended to have an even split between current and former Facebook users. However, it was difficult to find former Facebook users and so only one former user was included in this study.

I recruited participants at a medium-sized midwestern university through undergraduate and graduate class announcements, flyers, and referrals. Interested persons received a recruitment email. The recruitment email expressed an interest in understanding “how people use social media, their thoughts about privacy, and decisions they make about privacy settings.” The email communicated the selection criteria as follows: “You must be between 18 and 29 years old to participate in this study. Also, you must be a current or former Facebook user.” Respondents indicated their interest in completing the interview by sending a return email indicating “Yes” to the contents of the recruitment email.

I conducted interviews in a private interview room located on the university campus. I audio-recorded all interviews with permission and I took notes for each interview. I used Audacity on a password-protected laptop as the primary recording. Back-up recordings were done using a handheld digital voice recorder or the researcher’s cell phone. All participants received an information sheet with additional details on the study objectives, risks, and confidentiality measures. The relevant institutional review board approved these recruitment and data-gathering procedures.

The 10 interviews covered in this report took place in October and November 2018. The duration of interviews ranged from 28 to 58 min. The schedule of interview questions was developed in line with Gorden’s (1992) suggestions including avoiding “leading questions” and employing “motivating questions.” The interview guide consisted of five sections: Background, Social Media Use, Privacy Attitudes, Facebook Privacy Breach, and Wrapping Up. Interviews included the following summary of the Cambridge Analytica scandal:

Earlier this year Facebook was implicated in the breach of privacy for millions of users. Several news companies reported that Cambridge Analytica used Facebook to gather voting profiles on millions of Facebook users leading up to the November 2016 elections. Social media users were concerned about their privacy and many opted to close their Facebook accounts.

The interview schedule was finalized after it was reviewed by two colleagues and after a practice interview was conducted with another colleague who met the study’s eligibility criteria.

What It Means to #deletefacebook

One specific area of probing was whether the respondent had closed or deleted their Facebook account. There are different ways that users can understand the term “delete Facebook” or “close” a Facebook account. Users may think this can be accomplished by just deleting the app from their smartphone. In such cases, the user’s account will remain active. Users have the option of deactivating their account. This allows the user to take a break from the platform. While deactivated, that user’s profile and timelines are not visible to anyone. Users who deactivate can reactivate anytime by logging in to Facebook (2019). Deactivated accounts were not categorized in this study as closed accounts.

Users have the option of closing their accounts completely by using the “delete account” option. In these cases, accounts are no longer retrievable by the user. This study considered interviewees to have “deleted” Facebook or closed their account only if they had deleted their accounts from Facebook. No respondent had deleted their account at the time of these interviews.

Insider versus Outsider Positionality

Dwyer and Buckle (2009) argue that often researchers find themselves in the space between outsider and insider. This is what I experienced in the interviews. In some cases, I was an insider, and in others I was an outsider. There were two axes for eligibility in this study’s sample: one axis was that respondents had to be a current or former Facebook user, and the second axis for participation was that the respondent had to be in the age group of 18–29 years. I am an insider since I am a Facebook user. But I have become disenchanted with the service. As the interviews progressed, I could more easily relate to the interviewees who used Facebook occasionally or did not use it at all. Although I did not reveal that I am a Facebook user to my respondents, they could possibly tell from my knowledge level that I am a Facebook user or former user. Among the drawbacks of being an insider is that the context may be too familiar so that elaborations may not be gathered from respondents. This was addressed by asking respondents to “elaborate”, “walk me through”, to “help me understand” or to “make the connection for me” in cases where it appeared the respondent appeared to assume prior knowledge.

I am outside of the age group of my interviewees. I am 8 years older than the oldest person I interviewed and 17 years older than the youngest. I found it difficult to relate to my younger participants, especially those who are younger than 25. The challenge of being an outsider is often in the social distance between the researcher and the respondents. In all cases, I sought to gain rapport by taking on a congenial, non-judgmental disposition. I found that probing questions often led to further clarifications.

In terms of my other characteristics, I am an African Jamaican woman. Only one of the participants was of the same race. None was of the same ethnicity. And so, all the interviews required communication across several social boundaries: cultural, racial, gender, and age. I speak with an accent which may also have caused an interviewer effect. This could have elicited more socially desirable responses to political and race-related topics. I am a PhD student and that may also impact my relatability for the younger participants. I used probing to avoid any misunderstandings that could have emerged from any social distance with respondents.

Reactions to Cambridge Analytica Scandal

What Do Respondents Think About the Cambridge Analytica Scandal?

All 10 respondents knew about the Cambridge Analytica scandal before the interview and 3 of them introduced the Cambridge Analytica scandal without my prompting. Of the 10 respondents, 7 were not surprised that Facebook was implicated in the misuse of user data:

. . . it was not a shocker for me because that’s how Facebook makes money . . . it did not bother me at all, because it is not a big issue with me . . . I feel like that’s the way they make money. If you’re concerned, don’t use it. (Maxwell, 26 years)

I thought, well, that sounds about right. (I: Okay). Honestly, I wasn’t surprised at all. (Chris, 29 years)

I am not surprised (I: Okay) at all that there was a breach of some of these data. I think that happens a lot more than we actually know. (Andrea, 29 years)

Three respondents, however, were very surprised by the Cambridge Analytica matter. In the words of one respondent,

It was only then, I think, that many users, including myself, realized how Facebook has been facilitating the collection of information by others. (Mark, 28 years)

What Factors Influenced the Decision to Stay With Facebook or Leave Facebook Following This Event?

Deciding to Leave

Only nine participants had active accounts at the time of the interview (all but Kevin, 22 years, who had his account deactivated). Mark, 28 years, was the only respondent who made formal changes to his account specifically in response to the Cambridge Analytica scandal. He was the respondent who showed the most surprise about the situation and his responses showed that he felt betrayed:

It certainly felt exploitative because Facebook always presents itself as this friendly neighborhood community-facilitating platform. But it’s not that. It’s really harvesting our data. That’s what I realized more than I did before. And so it certainly made me feel quite exposed and vulnerable.

Although Mark deactivated his account, he reactivated it later:

. . . for a long time I kept it deactivated. But then I reactivated it . . . because it was quite helpful in networking, in getting in touch with people you had never met who can give you information about things like college applications, or living in a particular country, or this, or that.

Mark’s responses suggest that he had only deactivated his account as a reaction to Facebook’s misuse of data. Social needs introduced the imperative to reactivate.

Deciding to Stay

All the other participants did not make any formal changes to their Facebook account in response to the Cambridge Analytica scandal. Of the eight participants who kept their Facebook accounts active, six of these did not experience any change in their perception of Facebook. Three of these respondents did not think the specific privacy issues involved with Cambridge Analytica could really affect them:

. . . there’s the weird hypocritical sense in me is that I don’t think there’s enough self-generated data on my own Facebook to be concerned about leaking too much. I only ever posted anything that I do not mind everybody seeing. I’ve never really posted anything controversial at all, um, or too, too personal. (Andrea, 29 years)

. . . the only harm that can come to me that I was aware of [would be] from it posting and using, like, my data to target specific ads from Cambridge Analytica or whatever they were doing with the data to try to sway the election. I guess in that way it would be bad. But I never considered closing my account because of it. (Joseph, 29 years)

Three of the respondents who stayed with Facebook considered themselves to be rare users of Facebook. By maintaining low levels of engagement, these users felt they were not using the platform enough for the privacy breaches to matter:

I don’t get on my news feed enough for that fake news stuff to ever have really affected me that much. (Chris, 29 years)

It depends on your situation. For me, because I didn’t put anything on Facebook, the privacy concerns do not bother me that much. (Maxwell, 26 years)

Also, rare use and low levels of engagement was seen as a strategy that could be used to mitigate privacy concerns:

the more shitty stuff Facebook does, and the more insecure I feel with my data, then the less I’m gonna post, the more I’m gonna restrict what I share, and the more cautious I’m gonna be with my use of Facebook. (Maggie, 29 years)

I don’t trust Facebook as a company, data-wise. And so the less I can use it the better. Because obviously, I mean, you don’t get subpoenaed to Congress over and over again if you don’t have privacy issues. (Chris, 29 years)

Finally, a few respondents shared that the social benefits of Facebook would keep them there in spite of the Cambridge Analytica Scandal and other privacy challenges:

I know there’s many issues with Facebook. But I keep it because I’m someone who has 85% of the people—more, like—90% of the people I love are far away. So it’s a very easy tool to keep in touch with them. (Maggie, 29 years)

It would definitely be an inconvenience for reaching those particular two friends [who have poor phone service]. I would not be able to reach my boyfriend at all. And I would not be able to reach Jen at all, which would be problematic. (Victoria, 23 years)

Participants who were less surprised by Facebook privacy issues were resigned to the view that social media and similar online services provide value as a product of exchange. To reap the benefits, one must be willing to reveal preferences, social connections, and other personal data. Kevin, the only user whose account was deactivated at the time of the interview, said,

Part of this whole Cambridge Analytica and other Facebook issues is that, you know, at some point you get discouraged, you know. You think that this is the way it is . . . If you want to use those platforms, you have to give up some of your data. (Kevin, 22 years)

Victoria echoed similar sentiments:

Whenever you’re not paying for a product, you are the product. And I’m familiar with that concept. And (I: Okay) I think that’s a fair deal. At least in the case of Facebook, I’m getting a lot of value. (Victoria, 23 years)

What are the Possible Consequences of Closing a Facebook Account?

Not Quite Closed—Deactivating Facebook

While none of the respondents have closed their Facebook account, three have deactivated their accounts at least once. Joseph and Maggie deactivated temporarily to avoid distractions, while Mark did so in response to privacy concerns following the Cambridge Analytica situation. Kevin, 22 years, is the only respondent who had his account deactivated at the time of the interview. He does not have any active social media accounts.

Kevin’s Facebook account was inactive for at least 2 months at the time of the interview. He had been re-activating and deactivating it several times in the 12 months that preceded the interview. However, for most of that time it had been deactivated. He considers himself to be a non-user of Facebook. The main reason Kevin deactivated his account is that he did not believe his social media image was consistent with his real image:

I never felt true to myself on social media. I always felt like I was trying to be someone else. I would look at my profile and I wouldn’t recognize myself . . . And then I would see my friends, and they wouldn’t seem to me on Facebook and on Instagram (I: Mm-hmm) and on Snapchat the way that they seem to me in real life. (I: Right). And that dissonance, uh, just kind of freaked me out a little bit. (Kevin, 22 years)

Kevin proposed a reason why he thought others don’t leave Facebook:

I had thought about leaving for a year beforehand (I: Mm-hmm). But it always came back to the reason that I joined which is that things were happening on Facebook . . . That I would be missing out on a part of our, like, experience as, like, my friend group? Our age? Our culture? I don’t know what. What it was was always a little bit unclear to me. But there was this feeling that I would be missing out. (Kevin, 22 years)

Kevin’s concerns about fakeness on social media (for himself and others) were enough to prompt him to deactivate his account in spite of his concerns for missing out on aspects of the social environment. Based on Kevin’s language, he believes he has left Facebook. He is anomalous among the respondents because he keeps his account deactivated most of the time. His experience is the closest approximation among respondents to having a closed account.

Losing Touch

All respondents associated Facebook with keeping in touch with contacts. This was especially true for respondents who were living in a State that is not their home or those who were international students. All participants indicated that Facebook was a useful tool to get updates on friends and to view photos of others. Losing touch with friends, family, professional, and other contacts is a consequence of closing Facebook that was reported by the respondents.

Facebook Log in

Facebook serves as a log in for several other apps and online accounts. Leaving Facebook is associated with losing access to these other online accounts:

There are a bunch o’ other apps that you can use your Facebook account to sign in on. And it is much easier to do it than create a new account through that place every time, which I do that. I don’t use Facebook to sign in. But sometimes I think about it. (Chris, 29 years)

When asked if he thinks that is a reason people stay on Facebook, Chris responded,

Yeah. I didn’t think about that earlier, but yes I do. Especially ‘cause they’re probably already in depth into some other app that they’re using Facebook in. And you would lose all of your history in that by deleting your Facebook. (Chris, 29 years)

The voluntary use of Facebook credentials for other online services makes it difficult to delete Facebook.

Discussion

The Cambridge Analytica revelation did not cause any of these respondents to leave Facebook permanently, although one of them did leave Facebook temporarily by deactivating his account. That one person returned to his deactivated account a few months later when trying to contact a remote acquaintance. More than half of the respondents were not surprised by the Cambridge Analytica story, and the story was consistent with their impression of how their data were already being used. Prior information about Facebook’s data practices might have mitigated the public outrage about Cambridge Analytica.

Access to social contacts is the main reason people stay on Facebook. Furthermore, Facebook’s mechanisms encourage staying rather than leaving. The cross-platform use of Facebook logins to access other online services suggests that some of these respondents could be connected to Facebook perpetually, even if Facebook services are no longer useful to them.

Deactivation is a compromise for people who believe they may need the data or connections sometime in the future. They can reactivate anytime without difficulty to re-engage completely, or to re-engage selected services for a limited time. The deactivation and reactivation cycle is consistent with Baumer et al.’s (2013) finding that there are different levels of social media engagement. Indeed, among respondents there were no distinct categories of users and non-users, only differing levels of engagement.

Privacy issues are a concern, but most respondents have resigned themselves to this as the status quo. Most respondents are not overly concerned about the privacy issues. Some, however, state that they are scared, concerned, or worried about how Facebook treats their data. A reaction to these fears is to use the platform less frequently and with lower levels of engagement. Findings here do suggest that Facebook is becoming a secondary social media platform with more than half the respondents noting that it is not their preferred social media platform. For some respondents it was not an age-appropriate method of communicating, but it was usually used to maintain transnational and cross-generational connections, particularly with older family members. Due to rare use or lower levels of engagement, the privacy issues associated with Facebook were not a foremost concern. In general, findings suggest that there is some mistrust among respondents toward Facebook. For the majority of active users, Facebook is used in spite of the privacy challenges because it provides useful resources. The benefits of having Facebook outweigh the privacy challenges, that is, the social benefits particularly. But respondents would adjust their levels of disclosure and would limit their activities on Facebook if they were concerned about privacy.

This research study conducted in-depth interviews on only 10 respondents and so the results cannot be generalized statistically. The results can, however, be generalized theoretically and emergent concepts can be tested. Social desirability bias, interviewer bias, and interviewer effects are possible. But I worked to mitigate these issues by establishing good rapport and by probing where necessary. The results are not replicable. However, these localized findings can be applied to research questions for more generalizable surveys. The intentional non-use of social media is fertile ground for further study of social media. This would include further work to understand the impact of social media departure on individuals and the impact on the social media platform that the user leaves.

Re-imagining the Future of Social Media

Although social media has become a taken-for-granted aspect of many people’s lives globally, its history is short in relation to other media forms. Facebook is less than two decades old and much of the work it does is unprecedented. The Cambridge Analytica crisis is one of several moments where one must question how the operation of Facebook and other social media companies will progress in the future. Transparency, advocacy, and transborder cooperation are some mechanisms that could help with the progress of social media.

Facebook and other social media companies need to be more transparent about the business model that they are using. The findings suggest that many users do not really know how Facebook makes money and what aspects of their Facebook use is monetized. They do not always know what information is being shared and which parties their data are being shared with. Information on these matters needs to be more clearly articulated in a way that can be communicated to even the most inexperienced or youngest user. Specific information on privacy breaches must be shared. Users need to be advised on what breaches have taken place, how breaches occurred, and how their individual accounts have been affected.

There is a need for advocacy outside of the private sector for citizen’s social media rights. Governments, consumer groups, and private citizens should join forces to educate and advocate for social media users. Such groups should be involved in policy and legislative discussions. They could also be responsible for designing social media rights education in schools for children and on media for adults.

Although the majority of interviewees did not use Facebook as their primary social media, the platform continues to grow across the world. New users in developing countries are getting Facebook accounts as one of their first steps as digital citizens. Facebook, and social media by extension, is not dying. The future of social media, no doubt, involves further threats to citizens’ rights. Social media organizations serve citizens across the world. Transnational cooperation is needed as new types of threats to citizen rights are enacted on social media. Multi-stakeholder participation from a range of countries should be considered for the development of protocols and procedures around social media regulatory and policy issues.