Close
Are you sure? Are you sure you want to report this comment? I understand, report it. Cancel

Women increasing representation among US gamers - ESA

Women increasing representation among US gamers - ESA

Thu 24 Apr 2014 7:36pm GMT / 3:36pm EDT / 12:36pm PDT
People

Trade group finds 48 percent of players are female, largest share in over a decade

Men still outnumber women among the US gaming audience, but the gap is getting smaller, according to the Entertainment Software Association's latest Essential Facts pamphlet. According to the trade group's latest annual survey, 48 percent of US gamers in are female, while 52 percent are male.

That's the closest the two have been in at least a decade, and a far cry from 2006 and 2007, when the ESA found men made up 62 percent of the US gamer population, compared to women's 38 percent. The trade group also noted that women over the age of 18 make up 36 percent of the gamer population, more than double the 17 percent of the market made up of boys under the age of 18.

As expected, gamers are also getting older. The number of female gamers 50 and over jumped by 32 percent year-over-year, while the overall average age of gamers in the US crept up from 30 to 31.

The ESA's survey was conducted by Ipsos MediaCT, and includes responses from more than 2,200 US households. More of the group's demographic findings are available in the full Essential Facts 2014 release.

5 Comments

Andrew Watson
Programmer

92 200 2.2
I'd personally like to see what these statistics look like when you separate them by platform. I bet a large portion of the female increase is because of mobile/social gaming, whereas console gaming still has a much higher male population.

(Not that any of that's a bad thing!)

Posted:3 months ago

#1

Shehzaan Abdulla
Translator

81 174 2.1
Where's the raw data? The report consistently puts apples and oranges together, comparing the ratio of male gamers in one age-group against the ratio of female gamers in another unrelated group. And this kind of selective data presentation happens consistently making it very hard to get an overall picture of what the results were.

The ESA's report clearly has a slant on it to represent gaming positively as a highly social medium that is already embraced by all including all genders, age groups, life circumstances (parenthood) and so on. Am I the only one that got that impression from reading this?

I'd rather they presented the raw facts as they are instead of selecting the ones that make the points they want made. If the situation is the way they are presenting it the conclusions from the data should speak for itself.

Edited 3 times. Last edit by Shehzaan Abdulla on 25th April 2014 4:03pm

Posted:3 months ago

#2
Could the relative number of women playing games outstrip men in a few years? I think it is possible. We should not assume that 50:50 will be the equilibrium.

Much more interesting would be the proportion of women making games in the industry and how this compares with previous years. But you cannot collect this data asking 2200 households in a survey.

Posted:3 months ago

#3

David Serrano
Freelancer

299 270 0.9
@Shehzaan Abdulla

The Essential Facts report is a promotion tool with a history of playing run and gun with statistics and facts. Because the purpose of a promotional tool is to supply asymmetric information... not objective, truthful or accurate information. It should actually be called the "What The ESA Would Like You To Believe Are The Essential Facts About The Computer And Video Game Industry... But Really Aren't."

And I agree, it did seem like the goal this year was to promote the social and on-line aspects of gaming. And how on-line appeals to a wider demographic. It doesn't seem to align with reality, but it's what the ESA would like you to believe.

Edited 1 times. Last edit by David Serrano on 26th April 2014 3:48pm

Posted:3 months ago

#4

Yvonne Neuland
Studying Game Development

33 59 1.8
I would also be interested in viewing the raw data for this study. It would be very useful to be able to perform my own statistical analysis on the data.

I am curious about whether or not the data displays heteroscedasticity. Establishing whether or not the sample population demonstrated homogeneity, both amongst populations and between them, or heterogeneity would also be extremely revealing.

What kind of testing was used to analyze the data? Simple correlation testing? If that is the case, the results are nearly useless. To obtain meaningful results across a population with so many potential confounding variables requires ANOVA, MANOVA, MANCOVA and Chi-Squares testing in order to meaningfully examine the interrelations between various sub-groups.

Knowing the specific methods they used to gather their data is also required to evaluate the chances of sampling bias as a confounding variable.

What type of significance testing was used to determine the significance level of various correlational relationships is crucial to accurately analyzing whether or not the data is actually meaningful. I don't see any significance test values on the slides, and wonder if any significance testing was performed at all.

There is a 100% correlation between humans who get cancer and humans who live on a planet with a blue sky. That doesn't make it a relevant statistic. Changing the color of the sky to neon-magenta is unlikely to lower cancer rates.

Correlation strength is absolutely useless as a measure of meaningfulness. To evaluate this information in a way that allows for application, T-testing and Z-testing significance analysis should be performed. Is there a 1 tailed or a 2 tailed p-result for any of the correlations?

I would bet that there are dozens of meaningfully significant relationships between the various data points, but without the relevant data this cannot be established.

There is a vital need for studies on this topic, but data cannot be put to good use when it is not evaluated or distributed properly. There are plenty of people who will put it to bad uses, however, with propaganda that draws emotional support to negative reactions and strengthens the influence of confounding variables.

Edited 1 times. Last edit by Yvonne Neuland on 2nd May 2014 8:41am

Posted:3 months ago

#5

Login or register to post

Take part in the GamesIndustry community

Register now