The situation described at the tradeshow in this article has also happened to me, repeatedly and consistently. The difference? I AM technical. I just also happen to be a woman. I love the tech industry. I've been in it professionally since 1997, and I've run tech startups as a CEO for the past 13 years. And if there's one thing I can count on, it's the consistent, pervasive assumption that I'm not technical. I hate g…
Is it because I'm sexist? No! I'm not a very extroverted or social person and I find it personally very difficult to make initial contact with anyone I don't already know. Given that I'm already talking to your fiance, we can infer that something or someone has broken the ice and hence we are engaged in a conversation.
If you were the person with whom the ice was broken with first, it would be your fiance who would be seemingly "snubbed" by me, unless introduced explicitly by yourself. Maybe non-extroverts, shy, or just socially awkward individuals need to wear a public service announcement or something to make this clear to everyone.
"Disclaimer: I'm not sexist; I'm just socially awkward. Please break the ice and say hello to me!"
I'd also like to add, for your consideration, that some people are better at one-on-ones than navigating the social jungle that is multi-person conversations. I have no idea how to engage multiple persons at once outside of the context of a formal or informal presentation of some sort. This could be that the topics of conversation that I usually engage in with others are not usually of the anecdotal variety, which I imagine are amenable to group conversation, where others can more easily "participate" passively.
Finally, I'd like to point out that I think your social party experiment as you framed it will always generate results that are biased toward your assumption. Just because a phenomena has been observed to exist (in this case the phenomena, according to your testimony, is "men in a technical environment are more inclined to strike up technical conversations with men to the exclusion of the women also present") does not actually tell us anything about -why- this is so.
You can come up with any number of anecdotally reasonable hypotheses, but until you actually test these hypotheses in a well defined and scientific manner with an experiment designed to eliminate all of the potential biases, your experiment is no better (rhetorically speaking) at proving anything than a sexist diatribe along the lines of "why my ex-{girlfriend or wife}'s {anecdotally negative conduct} proves women are {some universal claim about all women}".