Monday morning at the office. I grab my third cup of coffee for the day. Open my email (which I’ve honestly already opened twice but been too un-awake to comprehend), and see a few replies from scientists I’ve contacted in the days before. Good! Time to schedule interviews! Then I just happen to notice the replies.
From the male scientist: “Dear Ms. Brookshire.”
From the female scientist: “Dear Dr. Brookshire.”
I have a Ph.D. in science, and while I use the knowledge of it every day, getting called “Dr.” is a rare event, reserved for airline tickets and people who want to suck up to me. But as I thought back, I remembered the times that scientists had called me “Dr. Brookshire.” And I remembered that most of them were women.
I sent a tweet. It was, if I may say so, cleverly crafted. It made a clear observation. It fed into the current zeitgeist. The spacing made it punchy.
Monday morning observation:
I have “PhD” in my email signature. I sign my emails with just my name, no “Dr.” I email a lot of PhDs.
Men: “Dear Bethany.” “Hi Ms. Brookshire.”
Women: “Hi Dr. Brookshire.”
It’s not 100%, but it’s a VERY clear division.
— Sci Curious (@scicurious) January 22, 2018
The tweet, as I had half suspected, took off. As of this writing, it has more than 2,000 likes and 550 RTs. This thing had gone viral.*
The replies put their fingers on what I was feeling. I can’t help but feel that men get automatic respect more often than women. That an email from a woman is often read, by men, as an email from a little girl. Other women felt the same way.
But as I feinted and parried with the replies, I began to squirm. I gave an anecdote. But I have a Ph.D. in science and one of the first things we learn is that our anecdotes aren’t data.
And I HAD the data. It was sitting right there in my inbox. I decided I had to know.
Spoiler. I was wrong.
I went through my work email account and my blog email account (through which I send requests for interviews for my podcast, Science for the People), and hunted down all the successful interview requests for 2016 and 2017. These are requests where I reached out to someone who I did not know for the first time, and they replied. I carefully cleared out all the people who I had talked to before (meaning we had an established relationship), or anyone who might be a friend or colleague. I also removed the replies from the students I email when I cover kids at science fairs, figuring they may not be up on email etiquette (these days they mostly text me anyway).
I send a lot of emails requesting an interview with a scientist or an outside comment from a scientist not on a study (to make sure I’m not telling you lies). These emails follow a very similar form. It’s not technically a template, but it might as well be.
Dear Dr. (or Mr., Ms., Mx., on the occasions when I know the person I’m emailing doesn’t have a terminal degree. I look them up beforehand to determine this),
My name is Bethany Brookshire and I’m a writer at Science News(Science News for Students). I came across your study on X, appearing on Y date in scientific journal Z. I am writing about the study.
Would you be willing to chat briefly about the study and its implications? Please do let me know. My deadline is FarTooSoon PM on AVeryCloseDate. I apologize for the short notice.
Thank you very much for your time,
Immediately beneath this appears my automatic signature, which looks like this:
Bethany Brookshire, Ph.D.
And a link to my site.
Of the emails I sent from January, 2016, to December, 2017, I received 268 total replies from scientists I had never contacted before. 142 of those replies were from men, 126 from women, a 53/47% gender split. (For the past three years I’ve made it my business to interview at least one woman for every single article or podcast. A couple of times I’ve struck out. I also try for people of color, but I need to focus more on it in the future. It is not tracked here.)
I went through all of the replies and sorted them into four salutation types.
- “Hi Bethany”
- “Hi Ms. Brookshire”
- “Hi Dr. Brookshire”
- “Hi” (or just launching into the email with no salutation at all)
I separated those salutation types out by gender (it’s generally easy to assign, and when I couldn’t, I went back to my published articles and found out easily enough.)
Overwhelmingly, people reply with my first name. “Hi Bethany” makes up 78% of all replies (209 people) and is split evenly 50/50 between men and women.
2.6% of replies (only 7 people) used “Hi Ms. Brookshire.” The gender split, however, was not 50/50. Instead, 57% of those salutations were from women, 42% from men.
Then we get to “Hi Dr. Brookshire.” This salutation was used by 7% of scientists (a total of 19). And more of those people were men than women. 63% of people who called me “Dr.” were men, and 37% were women.
Edited to add: Reader @sTeamTraen pointed out that with an odd number of replies (19), a 50/50 split on this would in fact be impossible. It’s a solid point.
In the “no salutation” category, the percentages showed a similar split. 11% of replies offered no salutation (a total of 31 people). Of those, 68% of men launched right into the meat of the matter, as opposed to 32% of women.
The total breakdown of replies by gender is included below.
The results here are clear. My original tweet, stating that I’d observed women calling me “Dr.” more than men, was wrong. There aren’t really stats I can do here (as the data has no variability and the numbers are low), and because of that, my email records can’t show any real difference in how men or women reply to me.
Why might my data be so different from what I felt in my tweet? One obvious reason came to my mind: confirmation bias. I had a pre-existing belief that men wouldn’t respect me as much in email as women, and I most remembered the observations that confirmed this belief, forgetting entirely the evidence that showed me my belief was wrong.
Another type of bias was also probably at play: recency bias. I was giving more weight to things that I had observed recently, forgetting things I had observed in the past.
There could be other things that skew the data. For example, it could be that people without Ph.Ds respond differently than people with Ph.Ds. Most of my contacts hold a terminal degree. But some don’t. They are sometimes graduate students, or sometimes scientists working with an organization who simply don’t have the degree. It would be interesting to see if people without Ph.Ds refer to me as “Dr.” more often.
The numbers here are also small. There were only 19 TOTAL people who called me “Dr.” in email in the past two YEARS. That’s not a sample size you can make any conclusion on.
Do my emails mean women don’t face sexism? Obviously not. Does it mean the many anecdotes I’ve received from other women on this topic are invalid? Absolutely not. My experiences only apply to me, and can’t be generalized to other women, other women Ph.Ds, or even other woman Ph.Ds who are science writers at my outlet.
There is plenty of evidence that women do not receive as much respect for their work or their accomplishments as men. In academia, men get more favorable grant reviews. In science writing, men get more long feature bylines. Female doctors (MDs) are granted the accolade in introductions from their colleagues much less often than men.
Additionally, when presented with a one-page CV equal in all other respects, male and female scientists alike prefer “John” to “Jennifer.” It could be that the low numbers of “Dr.” emails could reflect this. There could be a bias exhibited by male and female scientists alike. Unfortunately, the total numbers are simply too small to tell.
Edited to add: Reader @LJZigerell pointed out that I forgot the work of Williams and Ceci (and he’s right I did), which shows that, in faculty hiring specifically in Science, Technology, Engineering and Math, women get hired at higher rates than men (except in economics). These would all be candidates who had Ph.Ds and long publishing records. And that particular study has come under a lot of fire for issues of…bias. So findings of bias are not universal across all domains. This does not, of course, mean that bias doesn’t exist.
To find out who calls who “Dr.” in my field, we’d need a real experiment. We’d probably want to compare men and women with Ph.Ds in different areas of science. We’d want to vary if they had Ph.D. in their signature or not. We’d want to look at staff writers vs. freelancers and beginners vs experts.
My data is merely a retrospective. It’s not even very good data (though it helps that my original emails and signatures are extremely consistent in content and tone). It does not encompass many other aspects of my life, such as how people treat me on the phone, when people introduce me in person, etc. I only have the data from a specific set of emails over the last two years. There is no reason whatsoever to think that my overall experience is different in some way from the studies showing that women get less consideration and respect than men.
But my observation was technically wrong. And it was viral.*
I was careless. My friends and colleagues RT’d it. They identified with it. Some friends defended me against people questioning me about my experience, which may be the worst thing of all. I feel like I betrayed them in some small way. We all identified with something because it sounded like our reality. In many cases, it may well be reality.
But my observation about my emails was a mistake. And I worry that social media, right now, doesn’t allow for mistakes. People want black and white, biased or unbiased, fair or unfair, not the shades of grey that make up life experiences. I worry these days that any little mistake opens someone up to false equivalency. This tweet was incorrect, therefore women are liars. This tweet was incorrect, therefore there’s no bias against women. Women don’t lie at rates higher than men, and there is bias against women (and men are biased against believing it!). But we’re all people. We all mess up. We all talk out our butts sometimes and find out, to our shame, that we’re wrong. That the data don’t show what we thought, or, in my case, that the data don’t show much of anything at all.
I want to do something to make it right. Because I was a scientist. Scientists make corrections when they are wrong. They acknowledge when their hypotheses don’t bear out, and when the data simply isn’t strong enough to say. I am a journalist. Journalists make corrections when they make mistakes. They are accountable for what they put out into the world. I need to tell you what is true. Even if I’m embarrassed, what is true is more important than what I feel.
Please, if you’ve made it this far, share this. Something black and white went viral. It needs to be defeated by the reality – shades of gray.
*Virality is subjective. In my case, an average tweet generates about 2,000-4,000 impressions (the number of streams it appeared on). This one is as 211,000 impressions and counting. For me? That’s pretty viral.
Edited on 1/30 to add a note about 50/50 splits in odd n’s, and a note about the work of Williams and Ceci.