For all the benefits it bestows, the internet has a dark side. Recently, much attention has centered on online harassment. It is a phenomenon that can take a variety of forms: name-calling, trolling, doxing, open and escalating threats, vicious sexist, racist, and homophobic rants, attempts to shame others, and direct efforts to embarrass or humiliate people. While some accept online harassment as a nuisance, others face situations that prompt them to take serious action and precautions.

At a basic level, there is no clear legal definition of what constitutes “online harassment.” Traditional notions of libel, slander, and threatening speech are sometimes hard to apply to the online environment. In addition, the anonymous and pseudonymous nature of the internet can make it easy for people to attack others without repercussions.

Further, there is little consensus as to who should be responsible for monitoring bad behavior online. The Telecommunications Act of 1996 does not hold website administrators liable for content posted by users. Websites and other technology companies claim monitoring bad behavior is economically challenging. Many sites utilize “community standards” or self-reporting mechanisms to bring inappropriate behavior to their attention. However, these efforts have often been seen as unsatisfactory or ineffective. This was evident when Zelda Williams, the daughter of the recently deceased comedian Robin Williams, shut down her Twitter account after a barrage of insensitive comments and abuse related to her father’s suicide became overwhelming to her.

Zelda Williams ultimately reinstated her account following the intervention of Twitter executives, but this issue has recently become the focus of female journalists and commentators who have shared similar stories of harassment, bigotry, and threats in various online spaces. Journalist Amanda Marcotte recently shut off “mentions” on her Twitter feed—the ability of other users to tag each other in their tweets—after feeling defeated from years of harassment. Soraya Chemaly, a media critic and activist, recently outlined why online harassment is uniquely harsh and cruel to women. Jill Filipovic detailed her experience with how easily online harassment becomes offline harassment. Game developer Zoe Quinn and video game critic Anita Sarkeesian were forced to leave their homes after harassment surrounding what’s called #Gamergate, where the tension between the traditional “boys club” mentality of gaming and a growing call for gender parity in the community came to a head. As the controversy escalated, Ms. Sarkeesian was forced to cancel an appearance at Utah State University when she did not feel security under Utah’s gun laws would be adequate after receiving a “school shooting” threat. Amanda Hess, a freelance writer, argued that online harassment creates a “chilling effect” whereby women are disinclined to participate professionally, socially, or economically online.

The sexual nature of many of these incidents is a running thread through these journalists’ stories. The most recent release of private, nude photographs of celebrities like Jennifer Lawrence, Kate Upton, and Rihanna sparked a debate over expectations of security and responsibility online. Many commentators likened the release to revenge porn, in which nude or Photoshopped photos and videos of women are published on the internet, typically by way of a disgruntled ex-partner or spouse. Efforts by activists like Charlotte Laws, whose daughter was a victim of revenge porn, highlight the difficulty in regaining control of one’s image from the digital Pandora’s Box in an area with little legal precedent and resources.

At the same time, a number of prominent websites have begun to review their practices surrounding harassment. Twitter vowed to improve its policies after the Zelda Williams fallout, reassessing how to provide support for those harassed beyond their “Report Abuse” button. In the meantime, third-party software developers at the Electronic Frontier Foundation released “Block Together,” an app that allows Twitter users to block accounts and share lists of troublesome users. Last year, Facebook set up a new review process for pages and groups to determine if their content was too offensive for advertisements. Gaming company Riot Games assessed the abuses woven throughout their community and instituted new participation standards and mechanisms for players to police one another.

While these stories provide insights into online harassment, there has been little data about the prevalence of harassment online, when, how, and to whom it occurs, and its impact on victims. Much of the existing research focuses on teens, a group considered particularly vulnerable to bullying and whose social lives increasingly incorporate digital means. The Pew Research Center has examined issues such as kindness and cruelty and privacy on social media within the teen population. Some have argued that teens may have more resources than adults when it comes to online harassment, such as parents, teachers, coaches, school administrators, and others who can provide guidance and aid. When it comes to adults, if the steps one can take personally do not go far enough to end the harassment, local law enforcement is typically the only other option. But some research has argued that law enforcement agencies are either incapable or unwilling to take on such a complex world.

This is the context for the following study by the Pew Research Center. It is based on survey data from the Pew Research Center’s American Trends Panel. The first part of this report explores the incidence and demographics of online harassment. Part 2 describes perceptions of online environments and where online harassment takes place. Part 3 outlines how people responded to their online harassment. In Part 4, the emotional aftermath of online harassment is explored. Part 5 concludes the report by examining the incidence and demographics of witnessing harassment online.