Categories
Programming Psychology

Conference Circuit Sexism

Following on from my Journey Through Sexism and after yet another technology-industry-gender-incident, I formed a theory and decided to put it to the test.

Here is my theory.

My theory is that if I trawl Twitter for feedback on talks and conferences I will find that male speakers would be praised for their skill and/or knowledge whereas female speakers would be praised for a mixture of their skills and/or knowledge but also for other unrelated elements (such as looks) that male speakers are not praised for.

I restricted my study to “feedback Tweets” that were hash-tagged to a conference, written by an individual about a talk by another named individual. For practical reasons I only included English-language Tweets. I excluded Tweets written about individuals on non-individual accounts (such as magazines, organisations etc).

I filtered a preliminary list of several thousand Tweets from 15 conference hash-tags using the techniques above.

I will not disclose the names of any individuals or organisations named in the collected data and will not reproduce direct quotes. Instead, I will create examples that are indicative of trends to illustrate the findings.

After several hours of research the original theory has not survived the test. The research does have some interesting data though, which I summarise below.

  • 76% of Tweets in the sample were written by males
  • 74% of Tweets in the sample were written about male speakers

What I actually discovered from this data is that in almost all cases, the object of praise is actually the talk, rather than the speaker. For example “Super intelligent talk by [speaker]” or “Interesting talk by [speaker]” or “Great talk on [topic] by [speaker]”.

  • 91% of Tweets primarily praised the talk, rather than the speaker
  • 6% of Tweets primarily praised the speaker
  • 3% of Tweets primarily praised the topic

Of the Tweets that primarily praised the speaker, only two Tweets praised characteristics that could not be categorised as relevant to the topic. In both cases, the Tweet was written by a male about a female speaker. In one case, the speaker’s looks were complimented and in the other, the comments were related to a photograph included in the Tweet of the speaker preparing for their talk. All other cases where the speaker was praised, they were praised for presentation ability, story-telling ability or deep knowledge of their topic.

So what does this tell us?

The research suggests that the normal way to compliment a talk at a conference is to talk about how good the talk itself was, how dynamic or interesting or most commonly, how awesome or how great the talk was. This is the most common type of compliment no matter what the gender of either party is.

It is less common to directly praise the ability of the speaker without directing that praise via the talk, although I notice that some speakers have a style that encourages this (a particularly good presentation style or a note-worthy mastery of their subject matter).

It is uncommon to compliment the looks of a speaker and there were no incidents of male-to-male or female-to-male Tweets in this respect at all.

Is this surprising?

Not really. It is 2013 after all. The good news is that most people have made it past the “gender-barrier” and can rate a talk without referring to their attraction to the speaker. If you are still complimenting speakers on their looks rather than on their talk, topic or skill you probably need to realise that this isn’t a common way to praise a speaker and if they have worked particularly hard to present well, or have researched a topic in great detail I’m sure they would rather be complimented on these areas.

A note on my methods. In an ideal world I would have spent more time designing the methods for collecting data, categorising the content and having the categorisation verified. This isn’t my day-job though, so I haven’t spent the time on this I would like to. However, I believe you could repeat this test and get a similar result.