Internet Privacy Costs of User-Generated Content
Jason M. Weimer*
School of Media Arts and Studies
- *Corresponding Author:
- Jason M. Weimer
School of Media Arts and Studies
Visit for more related articles at
Global Media Journal
The growing universe of user-generated content (UGC) accessible through social media platforms is raising new questions about the role and nature of privacy in the modern media environment. While preliminary studies on UGC have concerned themselves with identifying potential consequences to individual privacy as a result of user-generated content production, relatively little is currently known about the types of privacy violations actually being experienced by UGC creators. This study supports previous research indicating an ever-increasing number of UGC producers. Findings of a survey conducted at a large Mid-western university indicated that UGC creation behaviors, at least among college students, is virtually ubiquitous. The findings support the conclusion that creators of UGC are willing to pay privacy costs such as unsolicited marketing communications, and unwanted advances from acquaintances for the gains in social capital made possible by creating personally identifiable UGC online. Analysis revealed that privacy violations are related to the extent of time individuals have been contributing UGC to social media platforms. Additionally, UGC creators perceive privacy violations experienced with greater regularity more severely than those experienced less frequently.
Internet, social networks, user generated content, privacy costs
Over the past fifteen years, traditional understandings of the roles occupied by audiences and producers of mass communications have been challenged in light of the Internet and advancements in information and communication technologies (ICTs) that increasingly allow users to create their own content online. Many have cited the variety of opportunities for audience interactivity in web based media and the growing universe of user-generated content (UGC) online as evidence of a fundamental alteration to the way messages are created and valued within societies across the globe (Acar 2008; Boyd and Ellison 2007; van Dijck 2009; Gonzales and Hancock 2008; Walther 1996).
As a consequence of the relatively recent rise in UGC, a great deal of concern is being expressed about the consequences to individual privacy in the new media landscape. Many of these perspectives paint UGC as potentially dangerous to its creators given risks for experiencing identity theft, job loss/rejection, or simply being embarrassed by content personally uploaded to online networks. Modern researchers are addressing these risks through exploring how creators of UGC perceive the risks and benefits of their behavior. While these investigations are important for generating understandings about the UGC phenomenon, this line of inquiry leaves a fundamental question pertaining to individuals’ actual experience with different privacy violations as a result of creating personally identifiable UGC largely unexplored. Are UGC creators really sacrificing their personal lives in order to like each other’s photographs, comment on each other’s posts, and make connections with those of similar interests online?
Courtois, Mechant, De Marez, and Verleye (2009) defined UGC as “content made publically available on the Internet, reflecting a certain amount of creative effort. UGC is created outside of professional routines and practices and exists in different shapes and sizes” (p.111). The rapid rise in UGC entering the media landscape can be seen through the proliferation of social networking platforms, blog forums, and online dating sites (Debatin, Lovejoy, Horn, and Hughes 2009; Ellison, Heino, and Gibbs 2006; Leung 2009; Weisbuch Ivcevic and Ambady 2009). In only seven years, the social networking site Facebook has grown into a network consisting of over 750 million global users, half of which access the site everyday. The average Facebook user contributes 90 pieces of content (web links, news stories, blog posts, notes, photo albums, etc.) each month, which network wide translates into more than 30 billion pieces of content being contributed every month (“Facebook | Statistics” 2011). Users of the video sharing website YouTube upload 24 hours of video every minute, meaning over 1 million hours of video content is being contributed every month (“YouTube - YouTube Fact Sheet” 2011). Similar, although slightly more modest, stories can be told about networks such as MySpace, Twitter, and Foursqaure.
The uses and gratifications (U&G) framework developed by Katz, Blumler, and Gurevitch (1974) has been a popular approach to studying the attraction of audiences to both consume and create the UGC illustrated in these examples (Courtois et al. 2009; Debatin et al. 2009; Guosong 2009; Leung 2009). In no small part this is due to uses and gratifications’ conception of the audience in an active role, displaying goal directed media consumption (creation) behaviors in response to expected need gratifications (Katz et al. 1974). It makes sense to use such a framework for explaining the motivations driving the popularity of social media and other UGC forums that can only be sustained through continued user contributions within them.
Some researchers have shown similar functions being served by UGC and traditional media content such as identity signaling, surveillance, social relations, escapism, and entertainment (Courtois et al. 2009). While valid, these functions aren’t overly helpful in understanding how UGC messages differ from those in traditional media environments. The building and maintaining social capital functions put forth by Ellison, Steinfeld, and Lampe (2007) are helpful in making this distinction. The authors’ findings show social networking sites functioning as a means for maintaining or creating “weak (non-emotional) and strong (emotional/access to scarce resources) ties” (p.1146) with others thorough a cheap and effective medium. Given the social needs often being gratified through UGC, it isn’t surprising to find many of these messages taking the form of personal self-disclosures.
The tendency for some forms of UGC to contain accurate and identifiable information (e.g. Facebook entries) generates concerns around privacy implications to users. Central to these concerns is the issue of users’ control over how their personal information flows into and out of the social networks they participate in. Tufekci (2008) drew attention to the permanent and searchable nature of online disclosures providing the potential for long-term privacy consequences. Ibrahim (2008), citing Gross and Acquisti (2005), asserted “most social networking sites make it easy for third parties from hackers to government agencies, to access participants’ data without the site’s direct collaboration, thereby exposing users to risks ranging from identity theft to online and physical stalking and blackmailing” (p.247). Debatin et al. (2009) identify inadvertent disclosure of personal information, damaged reputation due to rumor and gossip, unwanted contact, harassment or stalking, and uses of personal data by third parties all as potentially unintended and uncontrollable consequences to individuals participating in online social networks.
The tradeoff between gains in social capital and losses in users’ control over their personal information represents what Gershon (2011) has termed the “permeability predicament” (p.9). In relation to social networks, the permeability or accessibility of information contained on a social network will constrain the degree to which users can maintain and build social capital with other users in their networks. If users’ contributions cannot be easily consumed by others in their social networks the potential for building social ties or capital is diminished if not exhausted. As a result, social networks have incorporated structural elements into their designs that allow for networked individuals to readily access the contributions of others in their networks (e.g. newsfeeds, following, and tagging functionalities). At the same time, increasing the accessibility of socially networked information enhances the potential for unintended or unwanted abuses of personal information such as the myriad of privacy invasions discussed above.
Despite this apparent tradeoff, (Govani & Pashley, 2005) showed Facebook users having low levels of awareness about Facebook’s privacy policies. Furthermore, their study showed users’ awareness of Facebook’s privacy options/policy didn’t have a significant effect on the utilization of privacy tools. Tufecki’s (2008) study showed users’ perceived likelihood that their profiles on Facebook and MySpace would be seen by future employers, government agencies, corporations, or romantic partners did not have an impact on their profile visibility. Generally speaking, the willingness of users to openly share personal information through creating their own content online has been explained by the perceived gratifications (e.g. gains in social capital) outweighing the cost associated with invasions of privacy (Debatin et al., 2009; Youn, 2005).
Important to this discussion is realizing the majority of the work to date has been concerned with the potential rather than realized consequences of creating personally identifiable UGC (Govani & Pashley, 2005; Ibrahim, 2008; Tufekci, 2008; Youn, 2005). The work of Debatin et al. (2009) is among the few empirical research studies that has addressed realized consequences of online personal disclosure. Their conclusions showed individuals who reported a personal invasion of privacy were more likely to change their Facebook privacy settings than those who heard a third person account of a personal invasion of privacy. However, the gravity of these personal invasions was not explored.
This study directly addresses the realized privacy violations experienced by users of social networks. Rather than investigating how users perceived the potential risks of creating UGC that contains personally identifiable information, the analysis that follows concerned itself with identifying users’ perceived experiences with a variety of privacy violations. In addition to measuring the types of violations experienced by users, data concerning the frequency of experience and the perceived severity of these violations was also collected.
Collectively, these elements form the basis of what this inquiry terms “privacy costs.” Privacy costs are violations of individuals’ privacy that users have personally experienced as a result of creating UGC. As each individual participating in a social network is likely to have a somewhat different perspective of how different types of privacy violations affect him or her, one must be careful not to think of privacy costs as having universal values. Instead, each user will evaluate, presumably upon the basis of the frequency of experiencing a specific privacy violation and how severely they perceive each violation, what types of privacy violations have negative consequences for them and how severe or costly each consequences is.
This seems to be an important area to explore given the extensive work concerning the realized benefits of consuming/creating UGC and the relative absence of research on realized costs. This study investigated this gap through the following research questions: 1) Are producers of UGC paying privacy costs? 2) What is the relationship between individuals’ participation in UGC forums and realized violations to their privacy? 3) How do UGC producers evaluate the potentials for different privacy costs?
A 16-question survey was prepared and pretested on a group of 11 producers of UGC. From pretest responses, privacy cost categories were refined to differentiate between categories and eliminate the potential for redundancy. The finalized survey was voluntarily completed by students enrolled in an undergraduate media studies course at a large university in the Midwestern United States.
The class surveyed had an enrollment of over 200 students, and was exclusively offered to communication majors with concentrations in media studies or production. While convenient in this respect, the sample may provide a general understanding of how US college students who create UGC experience privacy costs, given the expectation for media students to be active in social networking, blogging, and audio/video uploading behaviors.
The survey was conducted in person during a regularly scheduled class meeting, and was not announced to students before the class began. After agreeing to participate and identifying their demographic information, students were asked to characterize their level of participation in creating UGC. Data was collected on whether or not participants currently had personal content they created online (yes/no); how often they contributed UGC (less than 3 times p/month, 3-7 times p/month, 2-3 times p/week, 4-5 times p/week, daily or almost daily), and how often they accessed UGC (same scale as contribution).
Different types of potential privacy costs were generated using the privacy risks identified by Debatin et al. (2009), but additional categories were added from suggestions provided in the pretest. In all, nine privacy cost categories were measured with respondents being asked to identify perceived experiences with each category as a result of UGC they created, how severe or important they found each category to be, and how many times they had experienced each consequence category. A sample of survey questions is displayed in Appendix 1.
An initial statistical analysis was performed on survey responses to generate an overall understanding of respondents’ experience with UGC. Secondly, regression and one-way ANOVA testing was performed on the responses to ascertain relationships between individuals’ experience with UGC and their experience with privacy costs. Privacy costs experienced as a result of UGC were coded (yes/no) and aggregated for all respondents.
It is important to note that the self-reporting measures used in this study do assume individuals are capable of accurately reporting their experiences of privacy violations. This assumption is in line with the work of Debatin et al. (2009) and is also consistent with previous work on perceived costs of UGC (Govani & Pashley, 2005; Livingstone, 2008; Tufekci, 2008; Youn, 2005).
190 surveys were obtained, with 17 of these containing omissions or incompleteness deemed too severe to analyze. Incomplete responses were discarded leaving 173 for analysis. Demographic information for the sample is presented in Table 1. The sample was predominately male, which was expected given a higher proportion of males being enrolled in the school’s media production/studies major sequence. A mean respondent age of 19 years indicates that most respondents were first or second year college students.
Table 1: Sample Demographics & Proportion of UGC Creators.
Tables 2, 3, and 4 report findings on respondents’ experience with UGC creation and consumption behaviors. A striking statistic is seen in 82% of subjects reporting they access UGC every day. This finding coupled with almost half of respondents indicating they contribute UGC to the web every day speaks to the preeminence of UGC consumption and creation in the lives of modern youth. It is also interesting to note that more than 75% of respondents indicated producing UGC for three or more years, with almost 38% reporting they have been creating UGC for more than four years. Given the average age of respondents, over one-third have been creating UGC for over 20% of their lifetimes.
Table 2: Duration of UGC Production.
Table 3: Access Frequency.
Table 4: Contribution Frequency.
Another striking statistic is seen in 94% of the respondents indicating participation in some form of online UGC creation. While this seems astoundingly high at first, these findings are more or less in line with previous findings (Debatin et al. 2009; Govani and Pashley 2005; Tufekci 2008). Regrettably, this finding also implies that very few non-producers of UGC were surveyed. One of the study’s goals had been to examine the relationship between producers and non-producers of UGC in terms of their experience with privacy violations; however, the low N size of non-producers made such a comparison impossible. General findings on the most common and frequently reported privacy cost categories are presented in Tables 5 and 6. General analysis on each category’s perceived severity is presented in Table 7.
Table 5: Privacy Cost Experience by Category.
Table 6: Frequency of Experiencing Privacy Costs by Category.
Table 7: Privacy Cost Severity Rankings by Category.
The most commonly reported privacy costs being experienced as a result of creating UGC were receiving unsolicited marketing communications and unwanted advances from acquaintances. Having personal information made available to third parties and having personal information made available to undesired others were also somewhat common with 37% and 22% of respondents indicating experience with each category respectively. However less than one-third of respondents indicated experience with six of the nine privacy cost categories analyzed. What’s more, the most commonly experienced privacy costs were those that many experienced frequently long before the advent of the Internet (junk mail, unwanted conversations or correspondences, embarrassing photographs on display, etc.).
This may help explain the generally apathetic responses concerning the severity or seriousness of privacy costs. Mean responses on a 0-5 scale (low to high severity) for each category were generally around 2.5 indicating mild perceptions of privacy costs up to and including such costs as identity theft, harassment, and stalking. As unsolicited marketing communications were outranked only by identity theft in terms of the most severely perceived privacy costs, the results are largely consistent with previous research indicating low awareness and concern for the privacy risks associated with creating UGC online that creators haven’t personally experienced (Debatin et al. 2009). Nevertheless, one needs to be careful about generalizing all UGC creators as being unconcerned about experiencing privacy violations as a result of sharing personal information online. The high variance in responses reported for each privacy cost category’s severity indicates that perceptions about privacy consequences are far from universal.
Research Questions Revisited
The first research question asked whether or not UGC producers are paying privacy costs as a result of their productions. While the limitations identified above still apply, the results indicate UGC producers do in fact pay privacy costs as a result of the UGC they produce. Over 2,000 experiences were reported with unsolicited marketing messages and 62% of respondents indicated at least one experience with unsolicited marketing messages as a result of UGC they created. Despite respondents indicating such messages being relatively severe invasions of privacy in comparison to the other costs explored here, one should question whether this can truly be conceived as a cost of producing UGC. Since such experiences are so common, it seems reasonable to assume that there could have been confusion concerning which messages were the result of UGC and which might be the result of other online behaviors (e.g. web-browsing and online shopping). However, costs such as unwanted advances from acquaintances, information uploads without permission, and personal information alteration without permission are all costs that are reported with some regularity and seem to have increased likelihoods of occurrence as a direct result of producing UGC.
The second research question was concerned with the relationship between individuals’ level of participation in UGC forums and their realized privacy violations as a result. Regression and one-way ANOVA testing was performed to identify relationships between users’ experiences with UGC and privacy costs. Step-wise multiple regressions were performed to understand if the duration of individuals’ involvement with UGC productions, frequency of accessing UGC, and/or frequency of contributing UGC were predictive variables for individuals’ experience with privacy violations, how severely they perceived privacy violations, and/or how frequently they experienced privacy violations. Testing revealed the duration of UGC production or the amount of time individuals have been participating in UGC production behaviors to be a significant predictor of the types of privacy costs experienced, and the perceived severity of these costs. The duration of an individual’s participation in UGC productions behaviors explained approximately 12% of the variance in the types of privacy costs experienced by participants (R2 = .128, p < .01) (Table 8). The duration of an individual’s participation on UGC productions behaviors explained approximately 4% of the variance in respondents’ perceived severity of each privacy costs category (R2 = .042, p < .01) (Table 9). None of the variables identified in the study were significant predictors of how frequently individuals experienced privacy violations. Collectively these results leave a great deal to be discovered about the relationship between producing UGC and experience privacy violations as a result. With more 85% of the variance in the types of privacy costs experienced and over 95% of the variance in how severely these costs are perceived unexplained, it is clear that significant influences have not been accounted for here. It is also important to keep in mind that the makeup of this study did not allow for comparing UGC producers with non-producers, a comparison that could be very useful for distinguishing the degree to which UGC productions contributes to privacy violations.
Table 8: Regression Duration of UGC Production and Privacy Costs.
Table 9: Regression Duration of UGC Production and Perceived Severity of Privacy Costs.
The third research question addressed how severely individuals perceived the different categories of privacy costs identified. While Table 4 lists the most severely perceived costs in order, a couple of inferences will be drawn out here. Primarily, it should be acknowledged that unsolicited marketing messages were only outranked by identity theft in terms of severity. Consequently, receiving unsolicited marketing is the only privacy cost category that is both being experienced by the majority of users and evaluated as having above average severity.
As mentioned before, this finding seems to be marginalized by knowing unsolicited marketing messages were commonplace long before UGC, and while UGC may be enhancing the frequency or identifiably of these messages, it is dangerous to assume that users are truly evaluating marketing messages as a more severe consequence than stalking or harassment.
Generally speaking, the privacy cost categories identified in this study did not receive above average rankings of severity from respondents. Since most individuals also had little experience with the majority of the privacy costs explored, this finding points toward the interpretation that individuals are not perceiving the privacy violations they experience as a result of their producing UGC as severe, but these perceptions would likely change if the frequency of experiencing privacy violations increases over time (e.g. unsolicited marketing messages).
This study supports previous research indicating an ever-increasing number of UGC producers. Indeed, with 94% of respondents indicating participation in some form of UGC creation, such behaviors, at least among college students, is virtually ubiquitous. These findings support the conclusion that creators of UGC are willing to pay privacy costs such as unsolicited marketing communications, and unwanted advances from acquaintances for the gains in social capital made possible by creating personally identifiable UGC online. However, it should be noted that very few potential costs had been realized by most participants in this study, and those costs experienced with some regularity were perceived much more severely than those that had not been experienced.
While making this acknowledgement is important, the tolerance for unwanted communications that seem to result from unintended uses of personal information disclosed in online UGC points to the potential for an evolving conceptualization of privacy in the modern era. To date, individuals cannot “opt out” of unwanted communications from others in their social networks or unwanted marketing communications mediated through social networking sites. Despite this lack of control, respondents were apathetic about such violations and this apathy may indicate a relaxation of individuals’ definitions and desires for privacy. Boyd (2008) asserts that “information is not private because no one knows it; it is private because the knowing is limited and controlled” (p.18). While the preceding definition still seems valid, this study points to the level or degree of exclusivity and control young media consumers desire over their personal information may be evolving.
These findings also challenge assertions that individuals are unaware of the risks posed by uploading personal information to the web. Instead, many have some personal experience with privacy violations that result from their participation in social networks, and in many respects are indifferent to the consequences. In an attempt to understand this indifference, it can be asserted that many social network users are interpreting certain privacy costs as being fixed. In other words, the findings indicated above may be indicative of individuals consciously incurring privacy costs as a necessary hurdle for participation in a social network, and while tangible to users, privacy costs are viewed as the cost of doing business on social networks.
Since the vast majority of respondents were creators of UGC, further research is needed to more firmly establish the role privacy concerns play in individuals deciding whether or not to create personally identifiable content online. Also the large variance observed in both the perceptions concerning the severity of different privacy cost categories, and the frequency of experiencing each category seems to indicate a wide range of users being characterized by the term “creator of UGC”. Future studies could work to differentiate the behaviors and attitudes associated with various producers of UGC and such definitions may help to understand why the various creators of UGC view privacy implications so differently. It is also possible that UGC producers are preemptively disclosing personal information about themselves to avoid potential embarrassments or misunderstandings in future face-to-face or interpersonal communications. While such a conclusion is beyond the scope of this study, it would be interesting for a future study to see if there are preemptive motives behind individuals sacrificing control over their personal information. Finally, it has already been said that this study had hoped to evaluate relationships between UGC producers and non-producers. Understanding how UGC producers and non-producers contrast in terms of their experiences with privacy costs would be a valuable continuation of this study.
- Acar, A. (2008). Antecedents and Consequences of Online Social Networking Behavior: The Case of Facebook. Journal of Website Promotion, 3(1/2), 62-83.
- Boyd, D.M., & Ellison, N.B. (2007). Social network sites: definition, history, and scholarship. Journal of Computer-Mediated Communication, 13(1), 210-230.
- Boyd, D. (2008). Facebookâ€™s Privacy Trainwreck. Convergence: The Journal of Research into New Media Technologies, 14(1), 13-20. doi:10.1177/1354856507084416
- Courtois, C., Mechant, P., De Marez, L., &Verleye, G. (2009). Gratifications and Seeding Behavior of Online Adolescents. Journal of Computer-Mediated Communication, 15(1), 109-137.
- Debatin, B., Lovejoy, J., Horn, A., & Hughes, B.N. (2009). Facebook and online privacy: attitudes, behaviors, and unintended consequences. Journal of Computer-Mediated Communication, 15(1), 83-108.
- vanDijck, J. (2009). Users like you? Theorizing agency in user-generated content. Media, Culture & Society, 31(1), 41-58.
- Ellison, N., Heino, R., & Gibbs, J. (2006). Managing Impressions Online: Self-Presentation Processes in the Online Dating Environment. Journal of Computer-Mediated Communication, 11(2), 415-441.
- Weimer Internet Privacy Costs of User-Generated Content GMJ
- Ellison, N.B., Steinfeld, C., & Lampe, C. (2007). The benefits of Facebook â€œfriends:â€ÃÂ social capital and college studentsâ€™ use of online social network sites. Journal of Computer-Mediated Communication, 12(4), 1143-1168.
- Facebook | Statistics. (2011). Facebook. Retrieved August 9, 2010, from http://www.facebook.com/press/info.php?factsheet#!/press/info.php?statistics
- Gershon, R. (2011). Redefining Business Space: Intelligent Networking and the Transnational Corporation (pp. 1-29). Presented at the Annual International Association for Media and Communication Research Conference, Istanbul, Turkey.
- Gonzales, A.L., & Hancock, J.T. (2008). Identity Shift in Computer-Mediated Environments. Media Psychology, 11(2), 167-185.
- Govani, T., &Pashley, H. (2005). Student Awareness of the privacy implications when using Facebook. Retrieved October 14, 2010, from lorrie.cranor.org/courses/fa05/tubzhlp.pdf
- Guosong, S. (2009). Understanding the appeal of user-generated media: a uses and gratification perspective. Internet Research, 19(1), 7-25.
- Ibrahim, Y. (2008). The new risk communities: social networking sites and risk. International Journal of Media & Cultural Politics, 4(2), 245-253.
- Katz, E., Blumler, J.G., &Gurevitch, M. (1974). Uses and gratifications research. Public Opinion Quarterly, 37(4), 509-524.
- Leung, L. (2009). User-generated content on the internet: an examination of gratifications, civic engagement and psychological empowerment. New Media & Society, 11(8), 1327-1347.
- Livingstone, S. (2008). Taking risky opportunities in youthful content creation: teenagersâ€™ use of social networking sites for intimacy, privacy and self-expression. New Media & Society, 10(3), 393-411.
- Tufekci, Z. (2008). Can you see me now? audience and disclosure regulation in online social network sites. Bulletin of Science, Technology & Society, 28(1), 20-36.
- Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23(1), 3.
- Weisbuch, M., Ivcevic, Z., &Ambady, N. (2009). On being liked on the web and in the â€œreal worldâ€ÃÂ: Consistency in first impressions across personal webpages and spontaneous behavior. Journal of Experimental Social Psychology, 45(3), 573-576.
- Youn, S. (2005). Teenagersâ€™ perceptions of online privacy and coping behaviors: a risk-benefit appraisal approach. Journal of Broadcasting & Electronic Media, 49(1), 86-110.
- YouTube - YouTube Fact Sheet. (2011). You Tube. Retrieved July 23, 2011, from http://www.youtube.com/t/fact_sheet
About the Author
Jason Weimer is a Ph.D. candidate in the School of Media Arts and Studies at Ohio University. His work is primarily focused in the area of mobile communication technologies and social networks. Mr. Weimer received his MBA from the University of Cincinnati.