Gender Differences in Privacy-Related Measures for Young Adult Facebook Users
Mariea Grubbs Hoy
University of Tennessee-Knoxville
George Milne
University of Massachusetts Amherst
Abstract
This study examines gender differences in young adults' privacy beliefs, their reactions to behavioral advertising, personal information-sharing behaviors, and privacy protection behaviors on social networks. This investigation uses a large-scale survey of college students based on a social networked sampling technique facilitated through a Facebook group. Results reveal several gender differences in these areas. Third-party data usage beyond the original purpose and behavioral advertising techniques are of concern to both genders but more to women. In addition, women engage in noticeably more proactive privacy protection behavior compared with a decade ago. The authors conclude with a discussion of implications for behavioral advertising.
Keywords: online privacy, behavioral advertising, social networking sites, gender differences
Introduction
Social networking sites (SNSs) such as Facebook and MySpace have become primary gathering places for young adults. With an estimated population of 18.6 million (Rapleaf.com 2008), adults ages 18-24 years are the largest age cohort on these sites. College students are particularly attracted to Facebook, with more than one-quarter naming it their most visited Web site (emarketer.com 2009b). Here users can connect with others, "hang out," and otherwise feel part of a community (emarketer.com 2007). This connection is expressed by creating a profile comparable to a personal home page (boyd 2007) that usually contains an array of personally identifiable information, including name, contact information, and demographic data. To create the connectedness of the online community, participants invite others to become their "friends" and respond to others' requests to "friend" them. Creating this link involves approval using a request by one party that is confirmed by the other (boyd 2007). Users can post pictures and videos of themselves and their friends and "tag," or identify by name, each person. In addition, activity on the site can be communicated to friends using the "news feed" feature (Stone 2008). Thus, SNSs' original purpose is to facilitate social interaction with others of one's choice.
Advertisers have taken note of SNSs and are using the content provided in user profiles to target consumers with individually tailored ads. These tailored ads can be based on a general profile demographic, such as sending local bridal shop ads to women whose relationship status is "engaged." Using the "friends of connections" tool, advertisers can target Facebook users whose friends are connected to specific pages, groups, or applications. These tactics represent a few of the many forms of behavioral tracking of consumers' activities online, which also involve "searches the consumer has conducted, the Web pages visited, and the content viewed-in order to deliver advertising targeted to the individual consumer's interests" (Federal Trade Commission [FTC] 2007, p. 2). This activity, known as behavioral marketing or behavioral advertising, typically takes place without users' awareness. Many consumers consider this behavior unacceptable and a violation of their privacy (Turow et al. 2009).
When advertisers use the personal information found on SNS profiles to deliver personalized ads, the usage of that information extends beyond what the SNS user originally intended: to develop and maintain social connections. As such, users may experience heightened privacy concerns when they become aware of the practice. The FTC has closely scrutinized the general practice of behavioral tracking and targeting recently, as evidenced by its November 2007 town hall and subsequent FTC Staff Report: Self-Regulatory Principles for Online Behavioral Advertising (FTC 2009). The focus of this concern is on behavioral advertising that involves third parties, not first-party activities, such as use of cookies for maintaining a shopping cart.
In such an environment, it is important for advertisers and policymakers to get a better understanding of SNS users' beliefs and behaviors so that advertising campaigns do not violate consumer privacy expectations. Of particular interest to this investigation are the awareness, beliefs, and behaviors of women, who use this medium more than men (Rapleaf.com 2008). Given the privacy risks that exist on social networking sites, the focus on women is warranted because previous research has found women to be more concerned about online privacy (Sheehan 1999) and at greater risk for online harassment and stalking (Working to Halt Online Abuse [WHOA] 2009). Moreover, though previous research shows women to be more protective of their online privacy than men (Sheehan 1999), it does not address the current Web 2.0 environment and its behavioral advertising strategies. Thus, the purpose of this study is to examine gender differences among 18- to 24-year-old people regarding online privacy and the use of personal information beyond its original intent in the context of the SNS Facebook. Although we examine several aspects of third-party usage of Facebook profile data beyond the original purpose of social connection, we specifically address the issue of behavioral advertising.
This research differs from prior research in several ways. In general, limited work has addressed SNSs and privacy, beyond describing the extent and type of personal information disclosed on the profiles (e.g., Acquisiti and Gross 2006; Gross and Acquisiti 2005). Previous survey research predated Facebook's Beacon behavioral advertising program and therefore did not address this emerging public policy issue. The data for this study were collected in April and May 2008 and expressly included questions related to this topic. Most prior research has consisted of in-class, paper-and-pencil surveys, resulting in data exclusively from one university. This study instead used an online survey that recruited through a Facebook group and thereby attracted users from a variety of universities and states. In essence, the very instrument studied (an SNS) was the mechanism used to recruit respondents and direct them to the online means of collecting data.
Literature Review
Social Networking and Privacy
In their original conception, SNSs represented an evolution of online communities that formed around a common interest. Although such sites still exist, SNSs have shifted to converge around people instead. They have an egocentric structure, which primarily facilitates people's communication with others who are already part of their offline social network (boyd and Ellison 2007). Facebook identifies members of an online extended social community as "friends," though often the social connection may be rather loose-such as a classmate or "friend of a friend." A social graph can be formed around the individual member, which shows all the online relationships he or she has (Interactive Advertising Bureau [IAB] 2009a). Thus, a Facebook profile, which makes the user's friend base visible to others, provides a means of creating and displaying the person's social sphere. Facebook also links its users to others in the same "network"-for example, a university or place of employment. Social networking platforms encourage users to post and share personal information as part of their online social interactions.
The nature of what is considered personal information in a Web 2.0 environment (i.e., SNSs) is changing, such that "The more content we contribute voluntarily to the public or semi-public corners of the Web, the more we are not only findable, but also knowable" (Madden et al. 2007, p. i). The online information trail that people leave, especially on SNSs, "creates a central repository of personal information" (Barnes 2006, p. 4) "that is both persistent and cumulative" (Viegas 2005, p. 3). Although more adults are aware of their "digital footprints," few are actively monitoring their online persona by conducting online searches of their names, most are not concerned about personal information that might be available to others online, and most do not take precautions to limit their personal information online (Madden et al. 2007).
Recent studies document the extent to which college students reveal personal information on their SNS profiles. Some of these studies have been based on self-reported information disclosures using surveys (e.g., Dwyer, Hiltz, and Passerini 2007; Stutzman 2006; Tufekci 2008), all of which report a high level of personal information disclosure on SNS profiles. Others have conducted content analysis studies on downloaded profiles from a network (e.g., Gross and Acquisiti 2005; Kolek and Saunders 2008). In their examination of more than 4500 Facebook profiles, Gross and Acquisiti (2005) find that an extensive amount of personally identifiable information posted, including real names, images that enable identification of the person, AOL instant messenger screen names, birth dates, hometowns, and high schools. Furthermore, access to a person's profile provided access to others through a listing of Facebook friends. They suggest several factors influence information revelation: the perceived benefits of selectively revealing information to strangers outweighing the costs of potential privacy invasion, the influence of peer pressure and "herding" behavior, a casual attitude or apathy regarding personal privacy, ignorance about the possible implications of personal information disclosure, trust in the participants or host site, and required acceptance of the site's default privacy settings. In a follow-up study, Acquisiti and Gross (2006) find that though most Facebook users were aware of their profile visibility, a sizeable portion was not. In addition, though the respondents expressed general online privacy concern, they were not noticeably concerned about their privacy on Facebook. Finally, those students who were aware of the visibility of their personal information appeared to rely on their own ability to manage their personal information and privacy, even though 77% reported that they had not read Facebook's privacy policy.
The data-rich environment of social networks and the willingness with which people post information has attracted advertisers' attention. For example, SNSs have used consumer profile information, their social relationships, and their behavior to upload ads on the social network platform and elsewhere on the Web. Demographic information serves to segment people and, coupled with behaviors, tailor advertising. When people use applications or join groups or fan pages, this information can be shared with third parties. More intrusive forms of behavioral tracking have been introduced and found controversial.
Facebook's efforts to introduce behavioral targeting through its Beacon advertising program, and the subsequent consumer backlash, suggest that such covert information gathering may not be welcomed by social network participants. Inaugurated in fall 2007, Beacon advertising was envisioned as an extension of the natural word of mouth occurring among profile users. However, with Beacon advertising, users' purchase behavior with "partners" who had signed up with the program and profile picture got sent to their friends, along with a commercial message about the purchase. The friends had no choice but to receive the message (Story 2007). Controversy erupted when Facebook did not allow members to opt out or provide easy privacy protection features. Facebook subsequently made adjustments to allow users to opt in before their purchases were shared (Story and Stone 2007). In September 2009, a class-action suit against Facebook for using Beacon advertising resulted in the program being shut down permanently (Perez 2009).
Other Facebook offerings have expanded access to user information and generated profile privacy concerns (Ragan 2009). In their study of 150 popular Facebook applications, Felt and Evans (2008) find that more than 90% of the applications access personal information that is unnecessary to deliver the application. Another concern is Facebook Connect (Vascellaro 2008), which allows members to use a single sign-in (Facebook's) to access other sites. When Facebook members use this single login, the visited Web site gets one-day access to that member's public profile information. Unlike Beacon advertising, users must explicitly grant third-party access to their profile data. Marsden (2009) notes that Facebook's insistence on using a person's real name, rather than a screen name or alias, to create an account links an inordinate amount of personal profile information to a specific person. In February 2009, Facebook created another controversy by "quietly" changing its terms of service. Of concern was the clause that stated that Facebook retained the license to the information-including the ability to use, retain, and display posted content (MacMillan 2009). Most recently, in November 2009, Facebook added "friends of connections," which allows advertisers to send ads to the friends of users who have "fanned" their group pages. Inside Facebook blogger Justin Smith (2009) notes that "this feature should lead to increased conversion on Facebook ads because users will find the social context and implied endorsement more interesting."
Gender, Social Networking Site Use, and Information Disclosure
Although the early days of Internet adoption indicated a gender divide, 73% of both men and women currently are online (Fallows 2005; Pew Internet & American Life Project 2008). However, their usage motivations differ. Noting that communication and information are the dominant motivators driving Internet usage, Jackson and colleagues (2001) find that college women were more likely to use e-mail, whereas college men were more likely to engage in Web searches. Fallows (2005, p. 2) finds that even when women conduct searches, they "tend to treat information gathering online as a more textured and interactive process-one that includes gathering and exchanging information through support groups and personal email exchanges."
Seventy-four percent of women name Facebook as their favorite Web site, compared with 60% of men (emarketer.com 2008a). Not surprisingly, women use SNSs more than men (Rapleaf.com 2008). Although both genders equally use social media to stay in contact with friends and family (79%), college women update their profiles (82%) and post pictures (46%) more frequently than do their male counterparts (65% and 35%, respectively). They are also more likely than men to comment on someone else's blog (27% versus 18%; emarketer.com 2008a).
With regard to gender differences in information disclosure, men provide their telephone numbers and addresses on their SNS profiles more often than women (Tufekci 2008). However, women post their preferences about movies, books, and religion more often. In addition, college women disclose personal information on Facebook at a greater level than do men across several areas, including the "About Me" section; photo albums; mean number of photographs; images of the self in a central photograph; any reference to alcohol, drugs, and partying; and photographs of alcohol (Kolek and Saunders 2008).
Several studies have identified gender differences regarding online privacy concerns and subsequent privacy protection behaviors. In general, women perceive more risk online and report more privacy concerns than men (Fogel and Nehmad 2009; Sheehan 1999). In her analysis of multiple Web scenarios, Sheehan (1999) reports three areas in which women expressed more online privacy concern than men: receiving notice that personal information is being used by other divisions of the patronized company; receiving e-mail from an unfamiliar company; and receiving e-mail but having no idea how the company got the address. In terms of privacy protection, women were more likely than men to provide incomplete information when registering for a Web site and notify their Internet service provider about unsolicited e-mail. However, they also read unsolicited e-mail and registered for Web sites (i.e., provided personal information) with greater frequency than men. Sheehan (1999) observes that men instead tended to display confrontational or aggressive behavior and adopted a wider range of behaviors in response to privacy concerns than women, who employed more passive behaviors. Similarly, Milne and Culnan (2004) find that men were less likely than women to read privacy notices and trust Web site notices. A recent identity fraud survey has observed than "men were more likely to use tools that help detect fraud more quickly, such as email or mobile alerts," whereas women took twice as long as men to catch identity fraud (Javelin Strategy and Research 2009, p. 4).
The attention paid to gender differences is important, because women tend to be more likely than men to be victims of online abuse. The organization WHOA (2009) reports that of the more than 2500 cases it has profiled from 2000 to 2008, women constitute 72.5% of the victims of online harassment and stalking. In addition, Javelin Strategy and Research (2009) finds an "unexpected trend," in that women were 26% more likely to be identity fraud victims than men. The extensive use of Facebook and the amount of personal information being shared put all users at privacy risk for the use of their personal information beyond the user's original intent. However, women may be at a greater risk for negative consequences because of their participation rates, their frequency of posting, and the extent of personal information they provide, compared with men.
Research Questions
More than a decade ago, Sheehan (1999) indicated that women express more concern about online privacy compared with men and engage in noticeably different self-protective behaviors. We extend this research by examining gender differences relative to privacy concern, awareness of and beliefs about use of SNS profile information beyond the original purpose (including behavioral advertising), and subsequent self-protecting behaviors on the Web. 2.0. Therefore, we investigate the following research questions:
RQ1: How do men and women differ in terms of privacy concerns and beliefs with respect to SNSs?
RQ2: How do men and women differ in terms of awareness and beliefs regarding usage of their SNS profile information beyond its original purpose (including behavioral advertising)?
RQ3: How do men and women differ in terms of privacy-related behaviors, including SNS privacy policy readership, proactive self-protective behaviors, and self-protective behaviors that require others' compliance?
Method
The study method involved an online survey of 18- to 24-year-old adults, who were recruited with a social networked approach using Facebook. This section describes the recruitment and administration, resulting sample, and survey questions.
Survey Recruitment and Administration
To solicit young adult survey participants who had a Facebook profile, we created a Facebook group that briefly described the purpose (Hey-That's My Profile!), presented the informed consent form, and invited them to link to the online survey (using surveymonkey.com). As part of an undergraduate research class, during a lab, all students sent out an invitation to their Facebook friends to join the group and changed the status on their own profiles to request and encourage their friends to visit the group page and link to the survey. These 29 students cumulatively had a total of 19,368 Facebook friends (not accounting for duplication). The number of friends for each student ranged from 142 to 1659.
To encourage participation further and give the group legitimacy, the research students first joined the group themselves (which resulted in all of their friends being notified through their profile newsfeed). The group page solicitation also encouraged respondents to invite their Facebook friends to invite their friends to join. In addition, if a respondent chose to join the group (which was not necessary to click on the survey link), that action would show up on his or her newsfeed, and that person's friends could see the group. More than 700 users officially joined the group, which allows anyone examining the group page to see group members' names and profile pictures. This action further enhanced perceived legitimacy and stimulated a herding mentality.
During the data collection period (April 8-May 2, 2008), 1030 people clicked the link to begin the survey. From among the completed surveys (689), we qualified the responses by eliminating outliers who took less than three minutes or more than one hour to complete the survey. We also restricted the sample to respondents 18 to 24 years of age and those residing in the United States. Therefore, 589 sufficiently completed surveys remain for analysis.
Sample Profile
Of the 589 respondents, 72.7% were women and 27.3% were men. Nearly three-fourths of the sample was 19-21 years of age. The sample was overwhelmingly Caucasian (90.7%). Although the respondents were concentrated in Southeastern states, they represented 31 states and the District of Columbia. Specifically, 76% of the respondents were from the Southeast/South, and 63% attended the focal university. Persons from the Northeast and Southwest/West accounted for 3% and 2.6%, respectively. Two percent came from the Midwest, and approximately 16% did not report their state.
Table 1 provides an overview of the respondents' participation in SNSs. In addition to Facebook, the respondents had profiles on other SNSs, primarily MySpace. Although more women had a MySpace account than did men (62.9% versus 54.7%), this difference is not significant. However, women did check their Facebook profiles more frequently than men, with 78.6% of them checking it several times per day compared with 67.7% of men (χ2 = 11.965, 4 d.f., p = .018). There were no significant gender differences in the average number of friends.
Table 1. Participation in Social Networking Sites
| Measure | Men
(Percentage) |
Women
(Percentage) |
Chi-Square, Significance Level |
|
Social Networking Sites on Which Participants Have Profile |
|||
|
39.1 | 31.8 | 2.890, N.S. |
|
54.7 | 62.9 | 3.289, N.S. |
|
7.5 | 11.7 | 2.221, N.S. |
|
7.5 | 11.0 | 1.615, N.S. |
| Frequency of Checking Facebook Profile | |||
|
67.7 | 78.6 | |
|
21.1 | 6.9 | |
|
7.5 | 12.7 | |
|
1.9 | 1.6 | |
|
1.3 | 0.2 | |
| 11.965(4), .018 | |||
| T, Significance Level | |||
| Average Number of Facebook Friends | 474.7 | 523.5 | -1.609, N.S. |
|
402.0 | 472.75 | |
|
27-1500 | 28-2000 |
| Note: N.S. = not significant. |
Survey Measures
Privacy Beliefs. The survey measured respondents' participation in SNSs, privacy concerns and beliefs related to SNSs, and beliefs about behavioral advertising and other profile information usage beyond original purpose. We assessed privacy concern on a five-point scale (1 = "Very Unconcerned" to 5 = "Very Concerned"). We assessed beliefs using statements with a companion five-point Likert scale (1 = "Strongly Disagree" to 5 = "Strongly Agree").
Privacy Protection Behaviors. The survey also assessed subsequent privacy-related behaviors in which the respondent engaged to protect and manage his or her online identity or persona. Most of these behaviors were autonomous; the person could engage in these behaviors independent of others. Respondents indicated the frequency in which they engaged these behaviors on a 1 ("I never do this") to 5 ("I always do this") scale.
However, of specific interest were the behaviors in which the respondent engages when "you find something that personally identifies you online that you wish weren't posted, and you can't take care of it yourself." In this scenario, the user often needs another's cooperation, and perhaps even needs to be confrontational, to protect and manage his or her profile privacy. Respondents noted the likelihood they would engage in these behaviors on a 1 ("Very Unlikely") to 5 ("Very Likely") scale.
Results
We used independent sample t-tests to compare men and women on a series of related questions. For each area of behavior analyzed, we Bonferroni-corrected the significance level by dividing an α of .05 by the number of comparisons made. Thus, with five contrasts, the significance level would be α < .01 (.05/5).
RQ1: Privacy Concern and Beliefs
As Table 2 shows, women were significantly more concerned than men about "the privacy of the information about (them) that's posted on Facebook" (3.04 versus 2.61, respectively; t = -4.12, p < .001). Although this difference is statistically significant, from a practical stance, the men were unconcerned, and the women were ambivalent. Although none of the other measures presented in Table 2 revealed significant gender differences, it is worth noting that men and women neither agreed nor disagreed that SNSs do a good job of protecting their privacy, are aware of how their SNS may use their profile information, know that others have posted information about them they wish had not been posted, and realize that others have seen information they have regretted posting. Regarding behavioral advertising, both men and women tend to agree that they wish they "were alerted every time a company was tracking my online behavior."
Table 2. Social Networking Sites: Privacy Concern and Beliefs
| Measure | Men | Women | T, Significance Level |
|
Privacy Concern for Information Posted on Facebook1 |
|||
| How concerned are you about the privacy of the information about you that's posted on Facebook? | 2.61 | 3.04 | -4.120, .0001 |
|
Awareness and Beliefs2 |
|||
| Social networking sites do a good job of protecting my privacy. | 2.97 | 3.03 | -.755, N.S. |
| It is very important to me that I am aware and knowledgeable about how my personal information will be used when I submit/post it online. | 3.82 | 3.99 | -2.192, N.S. |
| I'm not aware of how my social network may use information I post on my profile. | 3.07 | 3.02 | .608, N.S. |
| I wish I were alerted every time a company was tracking my online behavior. | 4.00 | 4.17 | -1.856, N.S. |
| In the past, others have posted information about me that I wish hadn't been posted. | 2.98 | 2.94 | .352, N.S. |
| In the past, people have seen information on my profile that I regret I posted. | 2.56 | 2.43 | 1.417, N.S. |
| Notes: Significance levels < .008 (.05/6) were significant due to Bonferroni corrections.
1On a 1 = "Very Unconcerned" to 5 = "Very Concerned" scale. 2On a 1 = "Strongly Disagree" to 5 = "Strongly Agree" scale. |
RQ2: Awareness of and Beliefs Regarding Profile Usage Beyond Original Purpose
With respect to behavioral tracking, there were no significant differences in terms of agreement that "advertisers track the online content I view" and "advertisers use my social network profile information to send me ads targeted to me." In general, both men and women moderately agreed, thereby demonstrating a modest awareness of this practice. However, women were significantly more likely to disagree with the statement "I wouldn't mind if my social network allowed advertisers to use information on my profile to send me ads about things I might be interested in" (1.76 versus 2.06; t = 3.366, p = .001; see Table 3).
The other gender difference of note is that women agreed more than men that "based on what I have posted on my profile, it would be difficult for a stranger to locate me in real life" (2.96 versus 2.62; t = -3.284, p = .001).
Table 3. Beliefs About Behavioral Tracking and Usage Beyond Original Purpose
| Measure1 | Men | Women | T, Significance Level |
| Advertisers track the online content I view. | 3.50 | 3.49 | .131, N.S. |
| Advertisers use my social network profile information to send me ads targeted to me. | 3.36 | 3.41 | -.489, N.S. |
| I wouldn't mind if my social network allowed advertisers to use information on my profile to send me ads about things I might be interested in. | 2.06 | 1.76 | 3.366, .001 |
| Nonprofit organizations use profile information as a way to identify prospective donors. | 2.89 | 2.91 | -.197, N.S. |
| Based on what I have posted on my profile, it would be difficult for a stranger to locate me in real life. | 2.62 | 2.96 | -3.284, .001 |
| I would not mind if an employer saw what I posted on my profile. | 3.20 | 3.16 | .394, N.S. |
| Sometimes people browse others' profile information as a form of "entertainment." | 4.33 | 4.48 | -2.372, NS |
| The government uses social network profiles as part of background checks. | 2.91 | 3.08 | -1.794, N.S. |
|
Teachers/professors use profile information in grade decisions. |
1.78 | 1.85 | -.900, N.S. |
| Note: Significance levels < .005 (.05/9) were significant due to Bonferroni correction.
1On a 1 = "Strongly Disagree" to 5 = "Strongly Agree" scale |
Differences in Privacy Concerns and Behavioral Advertising Measures. To gain further insight into gender differences and behavioral advertising, we examined the correlation between concern "about the privacy of the information about me that's posted on Facebook" and relevant measures (see Table 4). For men, the greater the privacy concern, the more they agreed that "advertisers use my social network profile information to send me ads targeted to me" (r = .175, p < .05); "it is very important to me that I am aware and knowledgeable about how my personal information will be used when I submit/post it online" (r = .411, p < .05); and "I wish I were alerted every time a company was tracking my online behavior" (r = .295, p < .05). The women's data revealed similar findings, except that women agreed less that "social networking sites do a good job of protecting my privacy" as their privacy concerns increased (r = -.098, p < .05).
Table 4. Correlations between Privacy Concerns1 and Behavioral Advertising Measures
| Measures2 | Men | Women |
|
Awareness |
||
| Advertisers track the online content I view. | .132 | .080 |
| Advertisers use my social network profile information to send me ads targeted to me. | .175* | .126* |
| I'm not aware of how my social network may use information I post on my profile. | .099 | -.003 |
|
Beliefs |
||
| Social networking sites do a good job of protecting my privacy. | -.126 | -.098* |
| It is very important to me that I am aware and knowledgeable about how my personal information will be used when I submit/post it online. | .411* | .132* |
| I wish I were alerted every time a company was tracking my online behavior. | .295* | .173* |
| I wouldn't mind if my social network allowed advertisers to use information on my profile to send me ads about things I might be interested in. | -.087 | -.068 |
| 1On a 1 = "Very Unconcerned" to 5 = "Very Concerned" scale.
2On a 1 = "Strongly Disagree" to 5 = "Strongly Agree" scale. *Significant correlation with privacy concerns about the information posted on Facebook at the .05 level (two-tailed). |
RQ3: Privacy-Related Behaviors
Table 5 displays several significant differences between men and women and their privacy-related behaviors. In every instance, women reported a greater frequency or likelihood of engaging in the behavior than men did. Privacy protection behaviors consist of two categories: (1) those that the person could do autonomously, primarily because the personal information was posted or under the control of the user him- or herself, including privacy policy readership, and (2) those behaviors that required cooperation of, and potentially confronting, others because the information was under the control of another. The results of both privacy behavior categories are in Table 5.
Table 5. Social Networking Sites and Privacy-Related Behaviors
| Measure | Men
(Percentage |
Women
(Percentage) |
Chi-Square, Significance Level |
|
Privacy Policy Readership |
|||
| Read privacy policy of any SNS before joining | 28.5 | 44.9 | 12.890, .000 |
|
|
28.0 | 41.8 | 9.533, .002 |
|
11.8 | 18.5 | 3.7838, .053 |
|
3.1 | 2.8 | .038, N.S. |
| Last time read Facebook's privacy policy | |||
|
61.5 | 44.2 | |
|
3.1 | 9.3 | |
|
9.3 | 14.0 | |
|
4.3 | 6.1 | |
|
21.7 | 26.4 | |
| 16.8(4), .002 | |||
|
Privacy Protection Behaviors Controlled by Self1 |
Men | Women | T, Significance Level |
| I provide some false personal information to set up accounts. | 1.87 | 1.96 | -1.010, N.S. |
| I provide some false personal information on my profile. | 1.74 | 1.65 | 1.258, N.S. |
| I regularly review Facebook's personal settings. | 2.35 | 2.81 | -4.625, .0001 |
| I monitor my profile(s). | 3.80 | 4.10 | -3.322, .001 |
| I'm careful about the pictures I post of myself on my profile. | 3.80 | 4.14 | -3.317, .001 |
| I un-tag pictures. | 2.80 | 3.49 | -6.120, .0001 |
| I'm careful about whom I friend. | 3.50 | 4.05 | -5.052, .0001 |
| I'm careful about what groups I join. | 3.68 | 3.97 | -2.831, N.S. |
| I "google" myself regularly. | 2.29 | 2.43 | -1.215, N.S. |
| I control my privacy settings so that only my friends can see my profile. | 3.20 | 4.21 | -7.610, .0001 |
| I control my privacy settings so that what I do on Facebook doesn't show up on my newsfeed. | 3.33 | 3.79 | -3.960, .0001 |
| I delete messages from my wall. | 2.18 | 2.29 | -1.139, N.S. |
| I use Facebook's new "friends list" privacy control to allow me to filter which friends group sees different details of my profile. | 2.05 | 2.41 | -3.101, .002 |
| Privacy Protection Behaviors Dependent on Others2 | |||
| Find out where the information originated from. | 3.80 | 4.05 | -2.819,.005 |
| Ask the person who posted it to remove it. | 3.72 | 4.13 | -4.037,.0001 |
| Send a highly negative message to the person who posted it. | 2.18 | 2.15 | .29, N.S. |
| Ask the person to delete a comment I made on his/her wall. | 3.34 | 3.71 | -3.480,.001 |
| Ask the network administrator to remove it. | 2.55 | 2.84 | -2.352,.N.S. |
| 1On a 1 = "I never do this" to 5 = "I always do this" scale. Significance levels < .003 (.05/13) were significant due to Bonferroni correction.
2On a 1 = "Very Unlikely" to 5 = "Very Likely" scale. Significance levels < .01 (.05/5) were significant due to Bonferroni correction. |
Behaviors Controlled by Self. More women (44.9%) than men (28.5%) read the privacy policy of any SNS prior to joining (χ2 = 12.89, p = .0001). We found similar results when examining only Facebook. Furthermore, significantly more men have never read Facebook's privacy policy (61.5% versus 44.2% of women; χ2 = 16.8, 4 d.f., p = .002). More troubling perhaps is that if they had read the privacy policy, it was more than six months ago, with only 3.1% of the men and 9.3% of the women having read the notice within the past month.
With respect to privacy protection behaviors controlled by the user, women engaged in the following behaviors more frequently than men: reviewing Facebook's personal settings (2.81 versus 2.35; t = -4.625, p = .0001); monitoring their profile (4.1 versus 3.8; t = -3.322, p = .001); being careful about pictures they post of themselves (4.14 versus 3.8; t = -3.317, p = .001); untagging pictures (3.49 versus 2.80; t = -6.12, p = .0001); being careful about whom they friend (4.05 versus 3.50; t = -5.052, p = .0001); controlling privacy settings so that only friends can see their profile (4.21 versus 3.2; t = -7.61, p = .0001); controlling privacy settings so that what they do on Facebook does not show up on their newsfeed (3.79 versus 3.33; t = -3.96, p = .0001); and using Facebook's new "friends list" privacy control feature (2.41 versus 2.05; t = -3.101, p = .002).
Behaviors Dependent on Others. If the respondents found something personally identifiable about themselves online that they wished was not posted, women displayed a greater likelihood of finding the origin of the information (4.05 versus 3.8; t = -2.819, p = .005); asking the person who posted it to remove it (4.13 versus 3.72; t = -4.037, p = .0001); and asking the person to delete a comment made on their person's wall (3.71 versus 3.34; t = -3.480, p = .001).
Differences in Privacy Protection Strategies. To further investigate differences between men and women in their privacy protection behaviors, we factor analyzed the 13 protection items for men and women separately. Table 6 shows the rotated factor pattern for men, and Table 7 shows the rotated factor pattern for women. The strategies are reflected by the loading structure of the items. Men used the strategies of lying, technology/tools, image management, and being careful. Women used the strategies of lying, post hoc control, being careful, and review. Although lying and being careful were similar between genders, there were differences in how the other strategies were grouped for women versus men.
Table 6. Data Factor Analysis Rotated Component Matrix for Men1
| Component | |||||
| Lying | Technology/tools | Image Management | Careful | ||
| I provide some false personal information on my profile. | .881 | .102 | .045 | .052 | |
| I post inaccurate information about myself on my profile. | .801 | .030 | .209 | .041 | |
| I provide some false personal information to set up my social network accounts. | .793 | .087 | -.012 | .094 | |
| I use Facebook's new "friends list" privacy control to allow me to filter which friends group sees different details of my profile. | .168 | .714 | .051 | .010 | |
| I regularly review Facebook's personal settings. | .226 | .712 | -.010 | .090 | |
| I monitor my profile(s). | -.015 | .626 | .258 | .166 | |
| I control my privacy settings so that only my friends can see my profile. | -.022 | .588 | .250 | .192 | |
| I control my privacy settings so that what I do on Facebook doesn't show up on my newsfeed. | -.255 | .510 | .381 | .152 | |
| I un-tag pictures. | -.015 | .069 | .802 | .025 | |
| I delete messages from my wall. | .197 | .143 | .704 | .177 | |
| I'm careful about the pictures I post of myself on my profile. | -.012 | .208 | .529 | .393 | |
| I "google" myself regularly. | .264 | .315 | .501 | .024 | |
| I'm careful about what groups I join. | .030 | .135 | .139 | .877 | |
| I'm careful about whom I friend. | .160 | .172 | .118 | .820 | |
| Percent variance explained | 16.6% | 16.6% | 14.5% | 12.4% | |
| Notes: Extraction method: principal component analysis; rotation method: Varimax with Kaiser normalization. | |||||
| 1Rotation converged in five iterations. | |||||
Table 7. Data Factor Analysis Rotated Component Matrix for Women1
| Component | |||||
| Lying | Post Hoc Control | Careful | Review | ||
| I provide some false personal information on my profile. | .908 | -.033 | .015 | .044 | |
| I provide some false personal information to set up my social network accounts. | .828 | -.152 | -.011 | .177 | |
| I post inaccurate information about myself on my profile. | .785 | .247 | -.059 | -.054 | |
| I un-tag pictures. | -.105 | .688 | .211 | -.017 | |
| I delete messages from my wall. | .269 | .641 | .307 | -.088 | |
| I control my privacy settings so that only my friends can see my profile. | -.081 | .554 | -.124 | .321 | |
| I control my privacy settings so that what I do on Facebook doesn't show up on my newsfeed. | -.025 | .554 | -.094 | .433 | |
| I use Facebook's new "friends list" privacy control to allow me to filter which friends group sees different details of my profile. | .203 | .402 | .012 | .279 | |
| I'm careful about what groups I join. | -.087 | .043 | .804 | .224 | |
| I'm careful about whom I friend. | .097 | .027 | .796 | -.004 | |
| I'm careful about the pictures I post of myself on my profile. | -.132 | .339 | .469 | .226 | |
| I regularly review Facebook's personal settings. | .052 | .117 | .070 | .723 | |
| I monitor my profile(s). | -.063 | .002 | .286 | .676 | |
| I "google" myself regularly. | .126 | .140 | .051 | .483 | |
| Percent variance explained | 16.5% | 13.5% | 12.5% | 12.2% | |
| Notes: Extraction method: principal component analysis; rotation method: Varimax with Kaiser normalization. | |||||
| 1Rotation converged in six iterations. | |||||
Discussion
Behaviorally targeted online advertising spending in the United States is projected to reach $4.4 billion by 2012 (emarketer.com 2008b). Of that, a little more than $2 billion is allocated to social network advertising (emarketer.com 2008c). However, this growth may be hampered if privacy concerns expressed by regulators, advocacy groups, and consumers are not addressed. This study has examined young adults' awareness of and beliefs regarding usage of their personal profile information beyond its intended use. Women are ambivalent regarding privacy concerns, compared with men, who are unconcerned. Yet women and, to a lesser extent, men have concerns about advertisers' and marketers' use of personal information. These identified gender differences offer insights to advertisers that want to use behavioral advertising.
Privacy Concerns and Implications for Behavioral Advertising
Previous research has found that women disclose more personal information than men on their SNS profiles across a variety of categories (Koleck and Saunders 2008). Thus, women are at more risk for privacy violations by virtue of having their personal information stored online. It is interesting that this greater exposed risk by women is, at first glance, at odds with their higher levels of online privacy concern. However, for this study's sample of young adults aged 18-24 years, the general concern for both sexes is limited (i.e., women = 3.04, men = 2.61 on a five-point scale). Overall, women are ambivalent about privacy and men unconcerned.
The lack of concern is driven in part by the benefits gained from participating on a social network format. With respect to usage, both genders agree that browsing others' profile information is a form of "entertainment"; however, women agree more. This finding is consistent with the general purpose of SNSs: to be findable and knowable. Also, browsing others' profile information seems similar to "searching," and Fallows (2005) indicates that women view online information gathering as a supportive experience and a form of personal exchange.
However, as young adults' concerns about the privacy of the information posted on Facebook increase, so does their awareness that advertisers are using their profile information to send them targeted ads. It is important to these people in particular to be aware and knowledgeable about how their posted personal information will be used. The FTC (2009) admonished the industry to enhance the visibility of this practice. One idea proffered through public comments is to provide "‘just-in-time' notice at the point at which a consumer's action triggers data collection" (FTC 2009, p. 33). This study's data result in a similar suggestion: Advertisers should alert users when they are tracking their online behavior. This approach serves as a viable way to provide notice and, coupled with the opportunity to opt in, control.
The advocacy group, Future of Privacy Forum, announced in January 2010 that the industry has agreed on a standardized logo that will alert users when their personal data are being used to deliver online ads. The organization states that most major companies will begin including the icon, along with the phrase "Why did I get this ad?" in summer 2010. The link would take users to a separate page that explains behavioral targeting rather than including it in the standard privacy notice. This self-regulatory effort represents a response to FTC admonitions (Clifford 2010; Future of Privacy Forum 2010).
In addition, women's privacy concerns about their Facebook profile information are significantly correlated with their assessment of SNSs' efforts to protect their privacy. The more concern they expressed, the less they felt that their SNS, namely Facebook, was doing a good job protecting their privacy. Thus, providing effective means for creating awareness and subsequent control of profile usage has ramifications for both the advertiser and the SNS. In general, female social network users are "not comfortable" with SNSs "selling data to advertisers," and 36% report that they would not use the site if it engaged in this practice (ShesConnected.com, 2009).
Self-Protection and Implications for Behavioral Advertising
The high SNS usage and relatively low concern women showed in this study may be explained somewhat through the self-protective privacy behaviors they undertake. This study has identified several privacy protection behaviors that women engage in with more frequency or likelihood than men, such as taking care in posting pictures, adding friends and joining groups, untagging pictures, and so forth. This finding is in opposition to research a decade prior, which notes that "women are adopting few behaviors that protect or safeguard their privacy online" (Sheehan 1999, p. 36). In the decade that has passed between studies, it is possible that the risks of online activity have been conveyed to women as part of their socialization and perhaps underscored by the publicized stalking and kidnappings in the real world that started from online encounters.
Along these lines, our findings show that women proactively engage in behaviors that they themselves can control or implement. With respect to behavioral targeting, women are less likely than men to be in favor of tailored advertising based on their profile information. Thus, even when they must rely on someone else to assist in protecting their privacy (e.g., find something posted about themselves and cannot remove it), they are assertive in seeking out the source and asking someone (the poster or network administrator) to remove it. Such findings conflict with the decade-old findings that women tended to adopt "primarily passive, non-confrontational behaviors when faced with privacy concerns" (Sheehan 1999, pp. 36-37).
The factor structures for privacy protection behaviors showed that the first factor for both genders was lying, that is, providing false information about themselves while setting up their accounts and on their profiles. Advertisers rely on accurate profile information to use behavioral advertising and send tailored messages. To the extent that SNS users engage in this behavior to protect their privacy, this form of advertising becomes ineffective and wasteful.
As part of its Self-Regulatory Principles for Online Behavioral Advertising, the IAB (2009b) encourages transparency for data collection and usage and the provision of consumer control over whether information is collected, used, or transferred to a third party. The IAB's principles also advocate obtaining consumer consent prior to behavioral targeting. If SNS users thought they had more control over their data usage, perhaps providing inaccurate information as a self-protection strategy would be deemed less necessary.
Advertisers thus face the dilemma of communicating privacy practices and other terms of conditions to users who are not actively seeking the information. Prior studies have called for ways to make the language more readable (Milne and Culnan 2004) or the links to such information more clear and conspicuous (Hoy and Lwin 2007). Going beyond stating these practices in the privacy policy to proactively alerting users prior about the tracking and requesting their permission would be a useful strategy, according to this study's findings. In addition, the SNS might enable users to opt-in to receive a mobile alert when it changes its privacy policy. This strategy might be particularly useful for young adults, who text at a much higher rate than older age groups (Rainie 2006) and increasingly access their SNS using their mobile devices (emarketer.com 2009a).
The remaining factors revealed gender differences. For the second factor, both genders expressed their control through various privacy settings. This result is interesting, in that these consumers do not read privacy policies. Perhaps the immediacy of information sharing in SNSs makes the use of privacy controls more salient. Also, stories in the press pertaining to the dangers of privacy invasions might be influencing the use of privacy settings. Finally, setting privacy controls may be appealing because it does not require reading long, verbose texts. Our findings complement a study by ShesConnected.com (2009), in which 93% of the women reported that the ability to control their privacy settings was "very important."
Beyond privacy setting controls, women extended their control by untagging pictures and deleting messages from their profile wall. In contrast, men's expression of control was supplemented by regularly reviewing Facebook's personal settings and monitoring their profiles. This difference may reflect women's more immediate perceptions of privacy risk or concern, and subsequent action, compared with men.
Women perceive more severe consequences of privacy loss than men, which contributes to their reluctance to buy online. Thus, to the extent behavioral targeting enhances perceived privacy losses, this practice could be counterproductive. However, women's perceived risk declines when they receive recommendations from friends. This reduction in perceived risk also strongly influences women to buy online (Garbarino and Strahilevitz 2004). A March 2009 iVillage study finds that women 18-43 years of age use social media and connect extensively with online communities to share a variety of information (Irwin 2009). To exploit a similar strategy, sites could encourage women to extend their proclivity for privacy protection behaviors to telling other women-especially within their existing online communities. The SNSs, Internet service providers, Direct Marketing Association, IAB, Ad Council, and individual marketers could act separately or as partners to develop campaigns and events designed to stimulate and model woman-to-woman communication about privacy protection. In addition to advocating proactive behaviors of employing SNSs' privacy control features, untagging pictures, and deleting messages from the profile wall that may present too much personal information, women could be encouraged to "be careful" in terms of the groups joined, people "friended," and pictures posted. A similar approach could be taken to explain the benefits of behavioral advertising. Both these actions would address the IAB's (2009b) first principle of education.
Conclusion
This study has examined Facebook privacy-related awareness, beliefs, and behaviors among 18- to 24-year-old people residing in the United States. We implemented a novel, social networked sampling technique using Facebook as a recruiting tool. The study provides insight into three questions. First, though women and men have similar numbers of friends and amounts of usage of SNSs, women are more concerned about their privacy being invaded (though concern by both genders is low). Furthermore, women believe it is important that they be knowledgeable about how personal information they posted and submitted would be used. Second, with respect to using SNS profiles for behavioral advertising, both women and men reject this practice, but women more so. However, almost half of both genders are not aware how this information is being used. Third, women are more likely than men to take proactive self-protective behaviors, and the underlying structure of self-protective behavior differs between men and women. These results provide an updated and extended view of how women protect their privacy. In a Web 2.0 environment, women are much more proactive in protecting their privacy than was reported a decade ago. We also have identified additional privacy protection strategies relevant to SNSs.
With respect to the method, the social networked approach seems to have the advantage of relying on an initial sample's social connections to build a final sample quickly. The creation of a Facebook group page, and asking participants to do so, engendered authenticity and enhanced the recruiting efforts. The sample is a self-selected convenience set of respondents who agreed to join a Facebook group because of their social connections. This sampling approach, though more efficient and potentially more generalizable than most convenience samples, still has limitations. For example, the range of demographics is narrow, due to the reliance on social connections, which limits generalizability. Spending more time selecting the initial groups could improve the usefulness of this method. If a sample with demographic diversity were initially specified, it might help ensure a more diverse sample in the end.
Further research examining a broader set of demographics, including a wider race profile and a non-college student-dominated sample, would improve current efforts. Given Facebook's expansion into older segments (IAB 2009b), this research could be replicated to compare differences across age groups. In addition, though most of the respondents had profiles on SNSs other than Facebook, the sample was distinctly Facebook users. Hargittai (2008) cautions researchers not to generalize results based on users of one site to others.
Finally, we find that privacy concerns need not be high to trigger concerns about the use of SNS profile data beyond its original purpose-especially for advertising/marketing uses of that information. Perhaps something in addition to privacy concerns is prompting this response. Further research could explore the underlying antecedents.
References
Acquisti, Alessandro and Ralph Gross (2006), "Imagined Communities: Awareness, Information Sharing, and Privacy on the Facebook," in Proceedings of 6th Workshop on Privacy Enhancing Technologies, P. Golle & G. Danezis, eds. Cambridge, MA: Robinson College, 36-58.
Barnes, Susan B. (2006), "A Privacy Paradox: Social Networking in the United States," First Monday, 11 (9), available at http://www.firstmonday.org/issues/issue11_9/barnes/index.html, (accessed September 8, 2008).
boyd, danah m. (2007), "Why Youth (heart) Social Network Sites: The Role of Networked Publics in Teenage Social Life," in MacArthur Foundation Series on Digital Learning-Youth, Identity, and Digital Media Volume, David Buckingham, ed. Cambridge, MA: MIT Press.
--- and Nicole B. Ellison (2007), "Social Network Sites: Definition, History and Scholarship," Journal of Computer-Mediate Communication, 13 (1), available at http://www.jcmc.indiana.edu/vol13/issue1/boyd.ellison.html (accessed September 6, 2008).
Clifford, Stephanie (2010), "A Little ‘I' to Teach About Online Privacy," The New York Times, January 27, available at http://www.nytimes.com/2010/01/27/business/media/27adco.html (accessed January 29, 2010).
Dwyer, Catherine, Starr Roxanne Hiltz, and Katia Passerini (2007), "Trust and Privacy Concern Within Social Networking Sites: A Comparison of Facebook and MySpace," in Proceedings of the Thirteenth Americas Conference on Information Systems, Keystone, Colorado (August 9-12, 2007).
Emarketer.com (2007), "Search Marketers Seed Social Networks," May 18, available at http://www.emarketer.com/Article.aspx?id=1004937&src=article1_newsltr (accessed May 18, 2007).
--- (2008a), "College Students' Social Networking," September 22, available at http://www.emarketer.com/Article.aspx?id=1006557 (accessed September 22, 2008).
--- (2008b), "Behavioral Targeting: Marketing Trends," June, available at http://www.emarketer.com/Report.aspx?code=2000487 (accessed November 16, 2009).
--- (2008c), "Facebook Corrals User Comments in Ads," November, available at http://www.emarketer.com/Article.aspx?R=1006700 (accessed November 6, 2008).
--- (2009a), "Mobile Plus Social Equals Opportunity," November, available at http://www.emarketer.com/Article.aspx?R=1007373 (accessed November 16, 2009).
--- (2009b), "Facebook Keeps Its Cool Among College Students," December, available at http://www.emarketer.com/Article.aspx?R=1007418 (accessed December 15, 2009).
Fallows, Deborah (2005), How Women and Men Use the Internet, Pew Internet & American Life Project, December, available at http://www.pewinternet.org/~/media/Files/Reports/2005/PIP_Women_and_Men_online.pdf.pdf, (accessed September 8, 2008).
FTC (2007), "FTC Staff Proposes Online Behavioral Advertising Privacy Principles," http://www.ftc.gov/opa/2007/12/principles.shtm (accessed December 20, 2007).
--- (2009), FTC Staff Report: Self-Regulatory Principles for Online Behavioral Advertising, February, available at http://www2.ftc.gov/os/2009/02/P085400behavadreport.pdf (accessed February 12, 2009).
Felt, Adrienne and David Evans (2008), "Privacy Protection for Social Networking APIs," presented at the Web 2.0 Security and Privacy Workshop, in conjunction with the 2008 IEEE Symposium on Security and Privacy, May 22, available athttp://w2spconf.com/2008/papers/s3p1.pdf (accessed May 5, 2009).
Fogel, Joshua and Elham Nehmad (2009), "Internet Social Network Communities: Risk Taking, Trust and Privacy Concerns," Computers in Human Behavior, 25, 153-60.
Future of Privacy Forum (2010), "Future of Privacy Forum Releases Behavioral Notices Study," available at http://www.futureofprivacy.org/2010/01/27/future-of-privacy-forum-releases-behavioral-notices-study/ (accessed January 29, 2010).
Garbarino, Ellen and Michal Strahilevitz (2004), "Gender Differences in the Perceived Risk of Buying Online and the Effects of Receiving a Site Recommendation," Journal of Business Research, 57, 768-75.
Gross, Ralph and Alessandro Acquisti (2005), "Information Revelation and Privacy in Online Social Networks (The Facebook Case)," Proceedings of the 2005 ACM Workshop on Privacy in the Electronic Society, 71-80, available at http://portal.acm.org/citation.cfm?doid=1102199.1102214 (accessed September 6, 2008).
Hargittai, Eszter (2008), "Whose Space? Difference Among Users and Non-Users of Social Network Sites," Journal of Computer-Mediated Communication, 13, 276-97.
Hoy, Mariea Grubbs and May O. Lwin (2007), "Disclosures Exposed: Banner Ad Disclosure Adherence to FTC Guidance in the Top 100 U.S. Websites," Journal of Consumer Affairs, 41 (2), 285-325.
Interactive Advertising Bureau (IAB) (2009a), "Social Advertising's Best Practices," May, available at http://www.iab.net/media/file/Social-Advertising-Best-Practices-0509.pdf (accessed November 16, 2009).
--- (2009b), "Self-Regulatory Principles for Online Behavioral Advertising," July, available at http://www.iab.net/media/file/ven-principles-07-01-09.pdf (accessed November 16, 2009).
Irwin, Tanya (2009), "Study: Women Look to Online Peers for Product Reviews," Mediapost, April 29, available at http://www.mediapost.com/publications/index.cfm?fa=Articles.showArticle&art_aid=105123 (accessed April 30, 2009).
Jackson, Linda A., Kelly S. Ervin, Philip D. Gardner, and Neal Schmitt (2001), "Gender and the Internet: Women Communicating and Men Searching," Sex Roles, 44 (5/6), 363-79.
Javelin Strategy and Research (2009), "Latest Javelin Research Shows Identity Fraud Increased 22 Percent, Affecting Nearly Ten Million Americans: But Consumer Costs Fell Sharply by 31 Percent," available athttp://www.javelinstrategy.com/2009/02/09/latest-javelin-research-shows-identity-fraud-increased-22-percent-affecting-nearly-ten-million-americans-but-consumer-costs-fell-sharply-by-31-percent/#more-1498 (accessed May 1, 2009).
Kolek, Ethan A. and Daniel Saunders (2008), "Online Disclosure: An Empirical Examination of Undergraduate Facebook Profiles," NASPA Journal, 45 (1), 1-25, available at http://publications.naspa.org/cgi/viewcontent.cgi?article=1905&context=naspajournal (accessed September 10, 2008).
MacMillan, Douglas (2009), "Facebook Connect: Your 8,000 Hidden Friends; Do Facebook Connect Users Understand How Extensively Their Profiles and Activities Are Used by Third Parties?" BusinessWeek, April 2, available at http://www.businessweek.com/technology/content/apr2009/tc2009041_649562.htm?site=cbs&campaign_id=djm (accessed April 2, 2009).
Madden, Mary, Susannah Fox, Aaron Smith, and Jessica Vitak (2007), Digital Footprints: Online Identity Management and Search in the Age of Transparency, Pew Internet & American Life Project, December, available at http://www.pewinternet.org/pdfs/PIP_Digital_Footprints.pdf (accessed December 12, 2007).
Marsden, Rhodri (2009), "Is Facebook Just a Passing Fad?" The Independent, February 5, available at http://www.independent.ie/business/technology/is-facebook-just-a-passing-fad-1629006.html (accessed May 5, 2009).
Milne, George R. and Mary J. Culnan (2004), "Strategies for Reducing Online Privacy Risks: Why Consumers Read (or Don't Read) Online Privacy Notices," Journal of Interactive Marketing, 18 (3), 15-29.
Perez, Juan Carlos (2009), "Facebook Will Shut Down Beacon to Settle Lawsuit," PC World, September 19, available at http://www.pcworld.com/article/172272/facebook_will_shut_down_beacon_to_settle_lawsuit.html(accessed September 28, 2009).
Pew Internet & American Life Project (2008), "Who's Online? Demographics of Internet Users," available at http://www.pewinternet.org/Data-Tools/Download-Data/Trend-Data.aspx (accessed May 5, 2009).
Ragan, Steve (2009), "Privacy Issues Plague Facebook Users-Yet Again," The Tech Herald, September 14, available at http://www.thetechherald.com/article.php/200938/4431/Privacy-issues-plague-Facebook-users-%E2%80%93-yet-again (accessed November 16, 2009).
Rainie, Lee (2006), "Americans and Their Cell Phones," April, available at http://www.pewinternet.org/Reports/2006/Americans-and-their-cell-phones/1-Data-Memo-Findings/02-Other-overall-findings.aspx?r=1 (accessed November 16, 2009).
Rapleaf.com (2008), "Rapleaf Study Reveals Gender and Age Data of Social Network Users," available at http://business.rapleaf.com/company_press_2008_07_29.html (accessed October 10, 2008).
Sheehan, Kim Bartel (1999), "An Investigation of Gender Differences in On-Line Privacy Concerns and Resultant Behavior," Journal of Interactive Marketing, 13 (4), 24-38.
ShesConnected.com (2009), "The Power of Social Networking for Women Research Study," available at http://shesconnectedmultimedia.com/pdf/report.pdf (accessed November 16, 2009).
Smith, Justin (2009), "Facebook Ads Now More Powerful with ‘Friends of Connections' Targeting," available at http://www.insidefacebook.com/2009/11/11/facebook-ads-now-more-powerful-with-friends-of-connections-targeting/ (accessed January 18, 2010).
Stone, Brad (2008), "New Tool from Facebook Extends Its Web Presence," The New York Times July 24, available at http://www.nytimes.com/2008/07/24/technology/24facebook.html?ex=1374638400&en=005f9a606dcd42f2&ei=5124&partner=permalink&exprod=permalink (accessed July 24, 2008).
Story, Louise (2007), "Facebook Is Marketing Your Brand Preferences (with Your Permission)," The New York Times, November 7, available at http://www.nytimes.com (accessed November 7, 2008).
--- and Brad Stone (2007), "Facebook Retreats on Online Tracking," The New York Times, November 30, available at http://www.nytimes.com (accessed November 30, 2008).
Stutzman, Frederic (2006), "An Evaluation of Identity-Sharing Behavior in Social Network Communities," Journal of the International Digital Media and Arts Association, 3 (1), 10-18.
Tufekci, Zeynep (2008), "Can You See Me Now? Audience and Disclosure Regulation in Online Social Network Sites," Bulletin of Science, Technology & Society, 28 (1), 20-36.
Turow, Joseph, Jennifer King, Chris Jay Hoofnagle, Amy Bleakley, and Michael Hennesey (2009), "Americans Reject Tailored Advertising and Three Activities that Enable It," technical report, Annenberg School for Communication, University of Pennsylvania.
Vascellaro, Jessica E. (2008), "WSJ Update: Facebook Takes Big Step to Linking with Media Sites," The Wall Street Journal, December 4.
Viegas, Fernanda B. (2005), "Bloggers' Expectations of Privacy and Accountability: An Initial Survey," Journal of Computer-Mediated Communication, 10 (3), available at http://jcmc.indiana.edu/vol10/issue3/viegas.html(accessed September 10, 2008).
WHOA (2008), "Comparison Statistics 2000-2009," available at http://haltabuse.org/resources/stats/Cumulative2000-2007.pdf (accessed April 30, 2009).
About the Authors
Mariea Grubbs Hoy is Professor of Advertising, School of Advertising and Public Relations, University of Tennessee. Her research interests include advertising disclosures and online consumer privacy. She has published in a variety of journals, including Journal of Public Policy & Marketing, Journal of Advertising, Journal of Consumer Affairs, and Journal of Current Issues and Research in Advertising. She holds her Ph.D. in Business Administration, Marketing, from Oklahoma State University. E-Mail: mhoy@utk.edu.
George R. Milne has been on the marketing faculty of the University of Massachusetts Amherst since 1992. Most of his research efforts have focused on issues pertaining to consumer information privacy, database marketing, and interactive marketing. E-Mail: milne@mktg.umass.edu.
