Another look at Newsweek’s harassment data

Newsweek article has recently posted an article in regards to Gamergate “Is GamerGate About Media Ethics or Harassing Women? Harassment, the Data Shows.”

I thought the article would analyze their data to prove their point, a commendable move that others should follow. However, while reading the article it is made pretty clear that the author has a strong bias on the issue. This wouldn’t normally be a problem except that I believe that some important points in the data were either toned down or ignored completely. And while the data itself seems accurate, the analyses applied to it were shallow at best. As a site fighting against misconceptions, I thought it would be important to address some of the issues.

In the article, they bring out statistics in regards to the use of #gamergate. Through Brandwatch, they analyzed roughly 500,000 tweets that contained #gamergate (“25 percent of the more than 2 million tweets about GamerGate since September 1”). From those tweets they determined how many included certain twitter tags and whether they were positive, negative or neutral. *Note* They took the stats from the 500,000 tweets and multiplied by 4 to get their data. If you’re not sure about this information then look at the graphs provided on their article which proves this. Also, every number they provided is divisible by 4, which is more than coincidence.

Out of the tweets, they found that people were mentioned in #gamergate tweets this many times

Brianna Wu= 38,952          Sarkeesian= 35,188          Kotaku= 23,500

Alexander= 13,296          Quinn = 10,400          Totilo= 1,708          Grayson= 732

Out of those tweets, the percentage of negative tweets were highest for Brianna Wu(8.5%), Stephen Totilo(7%) and Nathan Grayson(7%). *note* the percentages were estimates made from the provided graph since the actual numbers were not provided. However, because of the sheer amount of tweets received, this still meant that Wu, Kotaku, and Sarkeesian each received highest number of negative tweets overall. Newsweek combined Wu’s, Quinn’s and Sarkeesian’s negative tweets which equaled to more than Kotaku, Totilo and Grayson’s overall negative tweets. They took this as proof that the people of Gamergate were more interested in the harassment of the women than journalists.

My view is that people are innocent until proven guilty. Therefore, I would like to take this data and look at it in other directions.

Through Topsy, I decided to look at how often these people used #gamergate in their own statuses. Because of these people’s stances, I believe it is fair to assume that their statuses regarding GG are generally made in a negative/neutral manner. Therefore, it is likely that those people received extra attention with each usage. My hypothesis is that the number of #gamergate statuses from those accounts will positively correlate with the number of gamergate statuses they were tagged in overall.

*Note* Topsy only shows data for the last month (9/26-10/26). This is what I found. (This only includes tags of #gamergate and does not include gamergate as a keyword)
Wu=70      Quinn=32          Sarkeesian=20
Alexander=9          Grayson=3           Totilo=0          Kotaku= 0
As we can see, the people who have been mentioned in the most #gamergate tweets are the ones who personally tweet about it the most, with the notable exceptions of Quinn(will be explained below) and Kotaku.

I would also like to make another point here. I will take Wu’s example since she was mentioned in the most #gamergate tweets. They say that she was mentioned in 38,952 tweets out of 2mill. This is slightly less than 2%(1.95%) of #gamergate tweets that mentions her. If we add in Sarkeesian (1.76%) and Quinn (.52%) this equals to 4.23% (84,540). If Gamergate is based about harassing these women, than what are the other 1,915,400 tweets about?

And finally, the analysis above does not exclude positive/neutral tweets. If we were to focus on only the negative tweets the number is even smaller. Wu, with the largest negative percentage, revived less than 1 negative tweet out of every 10. If we take their overall average, I will add in a margin for error by rounding up, of 7%, the number is made even smaller. It shows that a little less than .3% of the #gamergate tweets are aimed towards harassing them.

On a final note from the Newsweek data, the women in general received many more positive tweets than the males. Grayson, in particular, received virtually no positive tweets in regards to Gamergate.

The data discrepancies can be further explained by taking the media into account.

1. The ones receiving the most #gamergate tweets are also the ones that are the most prominent in the media.
Wu has appeared on tv in regards to Gamergate and has been mentioned in many news articles.
Sarkeesian, already well-known in the feminist circles, was also involved in the USU gun threats where she canceled an upcoming talk.
Kotaku is one of the most well known gaming journals in the world where articles regarding Gamergate are still being published periodically.
Alexander still publishes gaming articles and her work is reviewed for any possible signs of corruption.
Quinn, Totilo and Grayson have all remained relatively obscure in the media.
– This could be a main factor in why Quinn is tagged in a relatively low amount of #gamergate tweets compared to the prevalence of her own #gamergate statuses.

2. Further clarification on Grayson’s extremely low Twitter count- When controversy was first brought up over Quinn’s 5 Guys Scandal, Grayson’s involvement had been acknowledged, researched and then released in an official statement for his innocence. After that, he has remained relatively quiet about Gamergate. During this time period, Quinn and her supporters had been very active on Twitter which kept her in public’s view. Though Quinn remains vocal, her lack of appearances in any media coverage has caused Gamergate supports to move onto more public figures, Wu and Sarkeesian. Their media appearances and inflammatory tweets have kept them in the spotlight.

In summary, I am not claiming that Newsweek’s data is wrong or misleading. I am promoting the idea that people should always keep an open mind to whatever information is being shown to them. There are always two sides to every story. The same data being used to prove the harassment agenda can also be used as a defense against those same claims. Even with statistics, the facts can be hard to find.

Here is an informational blog piece on the Newsweek data with further [real] statistics:

If you have any objections or wish to add on extra information then please feel free to share.

To clarify, this site is NOT pro-Gamergate. We are merely pro-Truth.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s