Men Are Entertaining and Women Are Awkward: An Analysis of Speaker Perceptions

 

This week, Annie Pettit follows up on her recent post, Promise Me You'll Submit to a Conference This Year, by sharing some of her own research exploring how men and women perceive themselves as speakers. Our takeaway from the findings is simple: the more women can see their own strengths and display them on stage, the more they and their male peers will realize they belong there.

 

Why are women underrepresented as speakers?

Why are women underrepresented as speakers, particularly at the conferences I go to where half of the audience members are women? Does fear chase them off the stage in disproportionate numbers?

I’ve pondered this question for years but I never knew if my hypothesis was grounded in fact or in stereotype. Fortunately, or unfortunately as the case may be, the opportunity presented itself and here we are pondering real data from a survey I did of 297 male and 252 female computer or data scientists, and market researchers aged 25 to 49 — people who ought to be on their way to securing spots on the conference circuit.*

One of the questions in the survey asked people to imagine speaking at an event and to choose any attributes that would describe themselves as a conference speaker. I was careful to include an equal number of both positive and negative attributes so as to avoid leading people to choose a greater percentage of positive (or negative) items.

Not surprisingly, the most popular choice among the 549 participants was ‘nervous’. I think we can all agree that being a conference speaker is nerve-wracking and I always assure new speakers that they are not alone in this regard. The second and third most popular attributes were the notion of seeming ‘intelligent’ and ‘entertaining’. And that also makes a lot of sense as anyone who gets on stage as a speaker ought to have sufficient expertise and presentation skills to be there. On the other hand, the least commonly chosen attribute was ‘Fake or Fraud’ which tells me that once on stage, people believe they deserve to be there. The good news is that the positive attributes were selected 60% more often.

Annie Pettit chart

Men and women differ in their perceptions of how they see themselves as conference speakers.

However, we know where this is going. Do men and women differ in their perceptions of how they see themselves as conference speakers? 100%, yes. Oui, si, ja, hai, and no ifs, ands, or buts.

In the next chart, I have ordered the attributes based on the difference between the percent of women and men who chose it (attributes with differences less than five points have been left out). Thus, on the left side are attributes that were chosen more often by women. On the right, are attributes that were chosen more often by men. Now look at the order of the attributes.

Without exception, men were more likely to choose positive attributes to describe themselves (e.g., intelligent, comfortable, entertaining) and women were more likely to choose negative attributes (e.g., nervous, awkward, terrified). In fact, men were much more likely to choose positive attributes — eight attributes generated differences of 10 points or greater.

Now, I may pride myself on loving statistics (I am @LoveStats on Twitter after all), but even someone who purports to hate math doesn’t need a statistical test to know that these differences are meaningful and important.

Annie Pettit chart

But let’s break things down even further. Let’s focus solely on people who have actually gotten up the nerve to speak at a conference at least once before. Surely the differences won’t hold up among people who’ve overcome the nerves and made it on stage?

Sadly, the trend continued. Though the overall rates of choosing negative attributes were much lower among speakers, female speakers continued to generate higher rates than did men. For example, though 59% of all women claimed to be nervous, only 39% of female speakers claimed to be nervous.  Compared to only 22% of male speakers.

On the other hand, levels of comfort were higher among speakers. Whereas 33% of all women said they would be entertaining speakers, 55% of female speakers said they are entertaining. But this still does not compare to 73% of male speakers who said they are entertaining.

Simply put, even among experienced speakers, women are still more likely to claim negative attributes (e.g., nervous, awkward) while men are more likely to claim positive attributes (e.g., comfortable, memorable, entertaining).

Annie Pettit chart

I still don’t know why women are so much more likely to claim these negative feelings. I’ve attended many conferences and I can assure you that male speakers don’t demonstrate that they are more amazing, more intelligent, more entertaining, or more comfortable than women. And women certainly don’t demonstrate that they are more nervous or awkward than male speakers. They may claim it but they don’t display it.

So what can we do about this?

It’s simple.

ENCOURAGE each other.

You’ve seen your female colleagues speak in staff meetings, team meetings, client meetings, volunteer meetings, soccer club meetings, book club meetings, and more. You’ve seen first hand that they are good speakers even when they think they are terrible. Help them cast that negative self-talk aside, or at least, act in spite it.

Just a few weeks ago, I asked you to promise me you’ll submit to speak at a conference this year. I’d like to add another request to that. Promise me you’ll encourage a woman to submit to speak at a conference this year.


* This non-probability survey was conducted in August 2017 with 297 male and 252 female residents of the USA whose ages ranged from 25 to 49, and whose careers included data science, market research, computer science, and university teaching. This sample included 30 women and 60 men with conference speaking experience. The data were weighted such that men and women mirrored each other on age, tenure, and household size. All sample surveys and polls are subject to multiple sources of error, including, but not limited to sampling error, coverage error, and measurement error. Many thanks to madai for programming and sample.

 
Annie+Pettit.jpeg

Annie Pettit is a market research methodologist who specializes in social media research, survey design and analysis, and data quality. She blogs at the LoveStats Blog and tweets @LoveStats.