Published on April 5th, 2013 | by Andy Astruc2
Games industry wage survey details reveal less than people thought
Earlier this week the results of a Game Developer Magazine survey were released and made the press rounds in our section of the internet. It appeared, to some, that the figures showed a tremendous gap between the wages of male and female employees involved in game development, and many outlets were keen to jump on the results as proving long-standing assumptions about sexism in the industry. The magazine’s editor has now spoken up about the survey, post-kerfuffle, and it seems like that might have been a wee bit premature. Or completely wrong.
Posting on Gamasutra, GDM Editor Patrick Miller provided some much needed perspective on the statistics. He expressed appreciation that people were interested in the issue, noting it was a common trend in previous surveys. Miller went on to expand on the chief concern held over the figures: the relationship between experience and wage as it relates to gender.
Unsurprisingly, Miller explains that “[a]cross all disciplines, the men… surveyed are more likely to have more experience”. The largest portion of male developers had more than six years experience, whereas the largest group of females had six or less. He concedes that this longer time in the industry likely leads to higher wages, meaning men will have higher wages than women overall. This is at odds with the more alarmist interpretation about ongoing sexist hiring policies, although Miller comments it does show the industry has been inhospitable for women in the past. However, it’s just as likely these number mean the industry is on the way to having more long-term female developers. Miller acknowledges much the same thing in the comments of his post, noting data is not sufficient to tell which interpretation is correct.
Which brings us to Miller’s other point: the sample size is very small. They surveyed a group of salaried developers — notably only in the US — and kept the female sample size at 11%, the same ratio as the overall pool. Miller notes this splicing of the sample (starting with only game developers, then moving down to salaried women and further splitting them into disciplines and experience level) inevitably leaves you with a very small set of numbers. In one comment, in response to a query about what the numbers would be if adjustments were made for experience level, Miller concedes it would be “way too small for an adjusted comparison”.
So it seems that, while interesting, these numbers aren’t really useful when drawing conclusions about pay equity in the games industry. Of course these details were present in the original data, and could have easily been inferred by anyone who took an unbiased and critical look at the figures. That’s probably why Fraser did exactly that two days ago when the survey results first appeared. Statistics can indeed be pushed and pulled to “reveal” whatever someone wishes, and in this case it seems the numbers were co-opted by a collective desire for evidence to support conclusions, rather than the other way around. This is, perhaps, a wake-up call for the gaming press that a little research and critical thinking is better than a lot of hysteria.