That ratio ignores all the however many (possibly way more) users that do not use steam while logged into their google accounts. So it doesn't tell us much except the stats for this very specific group (steam users who use steam while logged to google).
Not saying there isn't a gender skew, mind. Just that this doesn't tell us much outside these particular circumstances. Do we have any idea how many people fit into that group, and how many use steam in general?
Well we know that ~48% of Steam users were not logged into a google account. 2% were women logged into google, 50% were men logged into google. I think it's pretty convincing unless we can show that there would be some particular reason why women are more likely to not be logged into their gmail/youtube while browsing steam than men are.
I'm not saying "YOU'RE WRONG AND I REFUSE TO BELIEVE YOU" but I'd like to consider additional factors that could be affecting this:
Is it not the case that you cannot even log into Google in Steam? If I'm not mistaken about that, you'd have to have browsed one of Steam's sites in an actual browser. Assuming that whoever did the checking
1 checked the google analytics for all of them, anyone that detected would have had to have been logged into google in that browser, not been blocking google analytics, and also have told google what your gender was (the choices google gives are (male | female | decline to state) and I'm going to wager the default is 'decline to state' rather than male). So you could be missed because you don't have a google account
2, or haven't told google that you're a woman
3, or because you have google analytics blocked
4, or because you just never visited the steam homepage in a real browser.
1: Article said "Google Display Planner" but when I searched for that, I was redirected to AdWords, which apparently wanted me to buy ads
2: I know my mother didn't have one until she got an android phone, and doesn't even use the email associated with it. She has facebook and yahoo email. Of course, she doesn't use steam or even play games, so that doesn't mean anything. Obviously this isn't any kind of evidence, being just an anecdote, rather than data - but I don't think data should necessarily be trusted unconditionally either just because it's data - it may not actually reflect reality.
3: Maybe you don't want people bothering you on google+ or youtube or w/e; maybe you only have a google account for android and never set your gender in your account. Or you just haven't set it.
4: One of the default lists in umatrix blocks it, for example. My guess would be 'hpHosts’s Ad and tracking servers'. I don't know if this one is less likely than the others or not. I guess it would depend on the percentage of each gender which uses things like that.