What we did

Social media can increase social capital, provide entertainment, and enable meaningful discourse. However, threats to safety experienced on social media platforms can inhibit users’ ability to gain these benefits. Threats to safety – whether real or perceived – keep people from enjoying their online interactions and damage the quality of online social spaces.

While prior work has individually explored specific threats to safety – privacy, security, harassment – in this work we more broadly capture and characterize the full breadth of day-to-day experiences that may influence users’ overall perceptions of safety on social media. We contribute a novel, multidimensional taxonomy of how social media users define the term “safety” centered around security, privacy, and community.

This work is described in depth in our paper “I just want to feel safe”: A Diary Study of Safety Perceptions on Social Media, which was published at the International Conference on Web and Social Media (ICWSM) this week.

How we did it

We used a diary study to collect users’ experiences of safety and threats on Facebook over three weeks. Diary studies, in contrast to standard lab studies, allow participants more freedom of expression, which often results in more natural responses.

To avoid imposing a specific or more narrow definition of “safety,” such as “security” or “privacy,” we made sure to use the term “safety” itself to allow users to define for themselves the concepts and affordances that contribute to their sense of feeling safe (or unsafe) on a platform.

Finally, beyond Facebook, we compared memorable experiences of safety and threat across all platforms (including banking platforms, social media platforms, and online dating applications) and the intensity of those experiences.

Results

Overall, we find that participants felt safe on Facebook when discovering new privacy controls and when they received a notification about their safety or were asked to fulfill additional login requirements such as two-factor authentication (2FA). This gave them the sense that the platform was looking out for them.

As one participant of the study noted, “I felt really safe today when I received a notification from Facebook in my email that there had been somebody signing into my account off a browser that wasn’t recognized and a device that wasn’t recognized … and then it was asking for this two-step security process to prove who I was and everything. I just liked the fact that Facebook was vigilant; that it pays attention to its clients, so to speak.”

Participants also felt safer after completing a Privacy Checkup or a Security Checkup to audit their settings. One participant in the study said, “Another thing that makes me feel secure on Facebook is my ability to tailor my friend list as well as my ability to share my posts just with specific groups of people.” Participants in the study reported feeling more safe and secure knowing that options to control their privacy and security were available to them.

Perhaps surprisingly, the second most frequent experience of safety revolved around neither digital security nor privacy features, but around experiences on Facebook that helped people feel supported by or connected to their communities. For example, 76 experiences (34 percent of the 222 experiences we collected) involved feeling safe because of a community-building experience, such as connecting with offline community (e.g., neighbors) during a flood or being able to report harassment.

In addition, participants expressed feelings of safety from the mere existence of a reporting button, which is perhaps to them another indication of the platform having their safety in mind. It seems that security is more multifaceted (related to more controls) than security, offline experiences, and community standards are.

When comparing experiences across platforms, we found that experiences on Facebook look similar to those on other social media platforms. In contrast, experiences on platforms such as banking are more one-dimensional, in the sense that they are mostly about authentication and not privacy.

What’s next

We present the first framework of safety as a multidimensional online experience, called A Framework for Thinking About Safety.

Theoretical framework for understanding perception of safety: Account is a house containing many “possessions”; safety experiences lie on one or more dimensions related to these possessions.

We find that safety does not fit cleanly into the existing buckets of “security” and “privacy.” Rather, safety is a complex set of micro-experiences that involve not only digital privacy, security, and harassment, but also offline safety and well-being and the upholding of community values.

Our results suggest that we should consider the concept of community on social media as a two-sided coin: When a community is violated or ill-protected, people may withdraw from social media due to the risk of unsafe experiences. When community is strengthened, however, people feel safer and maintain the opportunity to build and capitalize from their social relationships. Thus, hyperlocalized, community-centric features (like location-based and custom group stories on Instagram and Snapchat) may help make people feel safer.

Our paper concludes with a discussion of how safety perceptions can be used as a metric for social media quality, and it details the potential for enhancing safety perception through community-enhancing affordances and algorithmic transparency. Click here to read more.