Abundance of trolls makes social media not so sociable for Canadians, says new poll

[More than a quarter of all respondents to a new Angus Reid poll reported being harassed on social media. AP Photo/Alastair Grant]

One in three Canadian social media users have experienced harassment or abuse online and more than half of Canadians believe that social media companies aren’t doing enough to curb this anti-social behaviour, says a new poll.

Men and women surveyed experienced harassment at roughly equal rates but women were more likely to experience certain types of abuse and were more likely to report it, says the survey by the Angus Reid Institute.

“We are a country of social media users and the propensity of harassment, I would say, is significant,” Shachi Kurl, executive director at Angus Reid, told Yahoo Canada News.

Pollsters put five real-life examples to respondents. They found a “significant gender gap” between what constitutes abuse for women and men. Younger men, in particular, were more likely to find the examples acceptable and by huge margins compared to women.

“Age plus gender is a huge driver,” Kurl said. “If you look at what 18- to 34-year-old men are thinking compared to what women their mom’s age may be thinking, some huge differences.”

Similar gaps have appeared in previous surveys on gender issues and sexual harassment, she said.

The most frequent users of social media, 18-34 year olds, reported the highest proportion of abuse. Members of the LGBTQ community also reported higher rates, as did members of visible minority groups.

The most prevalent form of harassment was being called offensive names, which 17 per cent of harassed users reported. Four per cent reported sexual harassment and four per cent being physically threatened.

“Women are twice as likely as men to say they’ve been stalked or sexually harassed, and one-and-a-half times more likely to say they’ve been subjected to unwelcome comments about their appearance,” says the report.

Somewhat surprisingly, users who reported being harassed were also more likely to find controversial comments acceptable. Kurl says this may be a “numbing effect.”

A Justice Canada “Handbook for Police and Crown Prosecutors on Criminal Harassment” notes that online harassment often turns into offline harassment.

“Technology also enables stalkers to cause a great deal of distress without leaving their home, which emboldens those who would not engage in offline harassment to stalk online,” it says.

“Moreover, the ability of the perpetrator to hide behind the mask of anonymity or to take on a false identity can make it very difficult, if not impossible, to tell the perpetrator to stop the harassment.”

Twitter, in particular, has come under fire for its failure to address abuse.

Headline-making stories such as the campaign against Ghostbusters actress Leslie Jones over the summer prompted the company to institute a quality filter that limits the responses users see in their feed and, earlier this year, Twitter made improvements to abuse reporting tools. The company has also put in place an advisory council of outside organizations and experts on the issue.

“With hundreds of millions of tweets sent per day, the volume of content on Twitter is massive, which makes it extraordinarily complex to strike the right balance between fighting abuse and speaking truth to power,” Twitter said in a statement.

But 53 per cent of social media users surveyed by Angus Reid said social media companies are not doing enough to curb offensive behaviour online and 61 per cent said they censor themselves out of fear of the responses they might get.

“The social media companies themselves have the power in their hands to look at their terms of reference, look at what they can do,” Kurl said.

“Canadians are really calling for those social media companies to take a more proactive and a harder line on offensive, harassing or inappropriate comment.”

The Angus Reid Institute surveyed a randomized representative sample of 1,530 adults in the Angus Reid forum online. Online polls do not allow for a calculable margin of error. For comparison purposes, a probability sample of this size would carry a margin of error of plus or minus 2.5 percentage points, 19 times out of 20.