Skip to content
Join our Newsletter

Truth lies behind closed browser windows

Part one of two In public settings, only the rudest or most ignorant individuals know not to spit out racial slurs. In private conversations, however, racism is still alive and well.
edit.20170616_6152017.jpg

Part one of two

In public settings, only the rudest or most ignorant individuals know not to spit out racial slurs.

In private conversations, however, racism is still alive and well. Few people would admit to that, of course, but they have no problem fostering their racism online, as Seth Stephens-Davidowitz writes in his book Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are.

Racist jokes are a common search item in Google across the United States, the data scientist found, but it's not a North-South divide as would be expected but an East-West divide.

Broadly speaking, residents of New York, Pennsylvania and Ohio are as likely to search for racist jokes as residents of Georgia, Tennessee and Florida. Residents in those six states will ask Google to find that material far more frequently than residents of California, Nevada or Arizona.

In general terms, Google Trends can be used to answer a question like are Americans more racist than Canadians.

Based on the frequency of their Google searches for racist jokes over the last five years, they are. For every 100 times an American will type in "racist jokes," that only happens 77 times in Canada. To make it a fair comparison, those numbers are a proportional rate against all search queries in the two countries.

One country's residents can't be declared more racist than another based on that single piece of information. More data is needed, starting by trying to drill down into the age, gender and other demographic characteristics of the people searching for racist jokes.

Are they rich white high school boys or politically incorrect old ladies?

Breaking down the Google Trends results by province, Manitoba residents search for racist jokes the most, closely followed by residents in Nova Scotia, Saskatchewan and Alberta. Residents in B.C. and Newfoundland are noticeably less likely to search for racist jokes. Quebec is by far the lowest, but that may have far more to do with a large French population that is far less likely to type in the words "racist jokes" into the search bar.

Again, that does not mean Manitoba is the most racist province in Canada.

Perhaps the results mean Manitobans like to live on the edge when it comes to politically incorrect humour. The Google Trends finding is simply a piece of data that points to a certain result but needs further data and study to confirm or reject the finding.

Stephens-Davidowitz and a growing body of data scientists are looking at online activity to answer these kinds of questions.

He called his book Everybody Lies because people intentionally or unintentionally exaggerate or minimize things they do.

For example, 80 per cent of drivers say they are better than average at driving, which is statistically impossible. Sometimes they don't even know they're lying. Online data researchers have found that people will tell pollsters they're undecided on who they will vote for in an upcoming election but they will tip their hand based on how they search Google for information about the candidates.

Stephens-Davidowitz notes that people searching for "Trump Clinton" were leaning towards voting for Trump while people searching for "Clinton Trump" were leaning towards voting for Clinton, based on the frequency of those searches in the final weeks of the U.S. presidential election and how the states where those searches came from actually voted.

Online data is also valuable to question commonly-held beliefs.

As Stephens-Davidowitz points out, the data doesn't support the existence of the echo chamber stereotype, where people only go online for news and information that confirms and reinforces their view of the world. He points to detailed discussions on the websites of far-right and racist organizations about articles in the New York Times, meaning that the members of those groups are reading and sharing stories published in a liberal mainstream media outlet.

That example is seen all the time on The Citizen's website and on its Facebook page, where people will insist that nobody reads or likes The Citizen as part of their complaint about The Citizen, its coverage and/or a particular article, seemingly unaware of the inherent contradiction of such a claim.

The simple reality is that, whether it's The Citizen, the New York Times or any other mainstream news outlet, people will read or watch and then go online just to argue about it.

Whether it's our views of news media coverage, political candidates or people of other races, what we say in public and what we practice online are often two different things.

The increasing use of online data crunching is telling us more all the time.

Yet sometimes we don't need to be experienced researchers to get a bird's-eye view of who we are. Sometimes all we need to do is start typing into the Google search bar.

More on that tomorrow.

-- Managing editor Neil Godbout