≡ Menu

2/3 of Facebook users aren't (necessarily) women.

Facebook has given marketers an powerful new tool for segmenting audience.  It's called Facebook Ads.  Once the "Create Ad" button has been pushed, potential advertisers can slice and dice audience any way they like.  As an advertisers tool it's fabulous.  As a market research tool, it's a gold mine. 

Paul Francis, author of Travels With My Overnight Bag, segmented Facebook users by male / female and country a couple of days ago.  TechCrunch picked this up, proclaiming that 2/3 of Facebook users are female.  However, a number of individuals noted that some Facebook users leave gender unspecified on the profile.  Francis redid his table to show the data more accurately.

I had noted something similar in the process of buying advertising for our 8G iPod promotion (still on, if you'd like enter) over the last week.  In fact, I thought I'd seen a pronounced change as age increases.  As you can see from the table below, it turns out that this is not only true, but a substantial number of older Facebook users decline to specify gender.


Overall, 13% of American and 20% of Canadian Facebook users don't provide gender data.  However, among users aged 25 and up, that number rises substantially.  And of course, there's no clue as to the reason why these older Facebook users might decline to provide gender data. 

Do men decline to give gender information more often than women?  How many of these individuals who have declined to give gender information are playing a role – men posing as women or vice versa?  In Francis' revised table, 50% of Turks decline to give gender information.  Is that because Facebook represents a way for men and women to meet virtually outside the confines of strict Islam?  Who really knows?  All that is clear is that when such a substantial number of individuals decline to give gender information, there's reasonable doubt associated with any claim that says that 2/3 of individuals are one sex or another. 

{ 6 comments… add one }

  • Aswath November 23, 2007, 9:16 am

    As an extension shouldn't we question FB's ability to segment its user base itself because the raw data doesn't seem to be reliable?

  • tim November 23, 2007, 12:07 pm

    though for what it's worth, i suppress my gender and also lied about my year of birth. (1925? not so much — late 80's. but who says mark z. needs to know?) maybe the cagier you are about your gender, the cagier you are about your birth year.

  • Alec November 23, 2007, 12:20 pm

    You both make good points. I suspect that the young don't lie about age and gender. They're (a) more naive and (b) likely more interested in meeting people of their own age.

  • MGU November 24, 2007, 11:34 am

    I wonder if some people have entered themselves several times in the hope of finding different segments of people whose interests, hobbies, living places or whatever interest them.

  • Alec November 24, 2007, 12:52 pm

    There wouldn't be a point, because you can list all of that stuff on your profile. Some people do have multiple profiles, however, for the different roles they have in life — professional, or personal, for instance.

  • Dave Muzzi November 27, 2007, 12:49 am

    A comparison with national sex ratios with the Facebook data often show correlations, except India and China, which are inverse.
    I made a map of this info: http://www.geointel.com/images/facebook_user_map.gif

Leave a Comment