Monday 21 October 2013

Gender & Statistics

In recent weeks I've been teaching a number of my students about the issue of gender in games design, making use of the research I did a year ago, covering the ideas of Sheri Graner Ray & Jesse Schell.  (Games design is a small world, so it was little surprise to discover that Graner Ray now works for Schell's games company.)

By coincidence I've also been re-visiting older blog pages to fix broken links, due to an increase in traffic* over the last month.  And one page with lots of broken links was Gender & Games Design.

* Hopefully it's other people finding my research useful, but I'm becoming slightly alarmed by the fact that 10% of total hits are for the Showaddywaddy page, and that 80% of my total traffic comes from the USA.  The data suggests that somewhere in Texas there is a mock-1950s glam rock revival waiting to happen.

During the process I stumbled upon some interesting info, and thought it was worth posting as an update to my original blog entry.

Statistics!


It's very easy to quote numbers for gender in gaming, but it usually boils down to the same two statistics: women account for only 20% of console gamers, yet they represent 70% of the casual games market.  (Sheri Graner Ray quotes these figures herself, in this video interview.)

However, it's difficult to find the origin of these statistics.  My own notes suggest that the 70% figure comes from 2004, so is it still valid?

Therefore I was very pleased to discover Nick Berry's DataGenetics' analysis of gaming based on Facebook "affinity".  This 2010 webpage contains detailed charts, which give a graphical breakdown by age & gender in the USA market, allowing comparisons to be made.


The chart for casual game Bejeweled Blitz shows a very clear bias to female players (78%), with middle-aged women accounting for a major chunk.

But is the Facebook userbase predominantly female anyway?  Social networking is certainly something which appeals strongly to women.  Cross-referencing with 2012 data from Google's AdClick service shows that Facebook is fairly evenly distributed for gender (60% female) but biased toward ages 25-44.  So Bejeweled definitely exhibits a primary female pull; the age thing might just be a reflection of the survey pool.



Compare and contrast Farmville with Call of Duty 4.  It says an awful lot about the gender split.


This split is not unusual, though.  In the movie world, Star Wars has a predominantly male following, whereas the Twilight series is synonymous with teenage girls.  (Berry notes the "double hump" in Star Wars age ranges, based on original and prequel movies; I'd also suggest there's an element of parents passing down their passion to their own kids)

 

TV is the same.  Consider Oprah Winfrey's 89% female audience:


Certainly, in the wider creative media production industries, these skewed demographics are not seen as an imbalance: instead, they represent an opportunity to target a specific audience more efficiently.

Danger, Will Robinson!


However, in my own personal opinion, these statistics are something we should be wary of.  Data such as this allows creative media producers an excuse to pander to the majority at the expense of the minority, creating a feedback loop of self-fulfilling prophecy.

For example, when Activision considered the cover design for Call of Duty 4 they undoubtedly knew that their primary audience was young & male (from previous games in the series) and chose imagery to appeal to that demographic.


But what if they'd tried to appeal to a female audience?  What could they focus on instead?  Maybe emphasise co-op gameplay?  Is there the risk that trying to widen the gender appeal could damage potential sales to their main market?

This is certainly a problem which has affected the book publishing industry, which has attempted to maximise gender or age sales using targeted cover imagery.  Writers like Jaqueline Wilson have spoken out about this gender profiling and mainstream media has commented on the rise of "pink plague".

I believe that this over-reliance on demographic profiling represents a form of social engineering, reducing consumer choice.  The "Rural Purge" in 1970s US TV programming is an example of this.  If we're not careful we may end up in a Gattaca-style dystopia where people are condemned to be allowed only those things which fit their profiling. And that's a scary thought.

Conclusion


There's a lot to think about here.  The key thing these statistics prove is that boys & girls can each like different stuff, and that's not a bad thing.  As a game designer it's important to reflect on things like this, and ensure that we're not stuck in a fixed mindset of our own likes & interests.  There may be a danger in taking these statistics too seriously, but there's also an injustice if we fail to consider them at all.