My online questionnaire has been live for three weeks and I'm still getting responses dribbling in - but only 15% of the responses are from men. I'd really like to have a better gender balance, but I'm not sure what I should do.

It's partly my own fault since some of the places I publicised the survey are very female-heavy in terms of population (fandom blogs, weight loss sites etc) - but others aren't (Facebook, Twitter etc).

I'm not sure whether I should aim for more responses from men (by posting on a more male-focused site, maybe) or whether I should just shrug my shoulders and focus on women in my interviews.

(I'd ask my supervisor, but she's not around (as usual), and I'm not 100% convinced she'd give me a useful answer anyway - but that's a whole other post...)

Unless you're asking a female-focussed question then such a massive gender imbalance in your responses will skew your results. I would make a concerted effort to redress this. Or change your research focus to just women. You may not succeed, but you should definitely try or else you're opening yourself up to too much methodological criticism.

I have a gender skew in my research. There's nothing I could really do about it, so just added it in as a limitation in my qual stuff and controlled for it in my quant stuff.

Hello Twomules, I think it depends on your dependent variable, if it is count, continuous, etc...
If it is a count dependent variable, you can use a zero-inflated poisson or negative binomial to balance the data, as in reduce too many females in the sample (reduce too many zero's.)
If it is a continuous variable, I would go with Sneaks, control for the gender.

Or I would analyze the data as it is (a lot of females, too few males), run a second analysis but this time, I'd reduce the female sub-sample by 30% and keep the male sample. Then I would compare the two analyses to see if they are different. If they are not different, then the original sample would be ok.

That's just my guess....maybe somebody can confirm?

There are indeed ways to handle skews, and legitimate reasons for them occurring. But if, as you say, it's arisen because of your approach to data collection, and if all it would require you to do to rectify this is a bit of targeted marketing of your survey, then I can't really see a reason not to make the effort, unless you are very short of time. Otherwise you may be laying yourself open to avoidable criticism at a later point.

I'd agree with HazyJane, my sample was skewed because it was already a restricted population and I'd bugged the male half of that population to the point where they were probably going to take a restraining order out on me lol.

Why not target some online male communities like football forums or car forums (stereotypical - me!? ;-) ).

One advantage (I think) is that my quant questionnaire results are primarily for the purposes of selecting interview participants (mixed methods - Participant Selection Model), so it is less of an issue as far as the analysis is concerned.

But it's still an issue, and Hazyjane is right - I can take a bit of time to try to get a more balanced gender demographic.

Thanks for your opinions!