I'm giving this its own top-level post as I suspect it would otherwise escape most people's attention:

Remember our previous discussion of the 'ambient belonging' study? The one about how environmental cues alone (Star Trek poster, comics, video-game boxes, pizza boxes, etc.) can influence how much people (don't) want to pursue computer science? In a stunning display of generosity, the contact author for that study (Prof. Sapna Cheryan) took the time to answer some of our questions on it.

My mail to her:

Professor Cheryan--

I am sure that you are thoroughly sick of dealing with questions regarding your paper, Ambient Belonging: How Stereotypical Cues Impact Gender Participation in Computer Science. However, I can promise you that I am not part of the popular press. Instead, I am merely a Computer Science ex-Professor, discussing the issues you raise with other Computer Science professors.

We have two technical questions regarding your results that you may be able to answer offhand. Well, okay, probably not, but we cling to the hope that you might find a moment to dig up the answers for total strangers at some point over the summer:

1) We note that Figure 1 shows z-scores. For our purposes, however, we'd like to know how large the effect was on the original 1-to-7 scale. Is there any chance you might recall the means and deviations of Figure 1's data, before they were standardized?

2) We found it worrisome that some of the 'stereotypical' objects might be strongly associated with males-- particularly the technical books, technical magazines, and electronics. Star Trek posters and video games, sure. But if the technical documents and components of our field are strongly associated with males, we have a much tougher job ahead of us. (How can we change the environment if our textbooks are associated with males?) Fortunately, we note that you asked your subjects of Study 3 to rate the masculinity/femininity of each item. Prithee, do you recall if the technical magazines were as strongly gendered as the other stereotypical objects?

Again, we are well aware of how busy Assistant Professors are, and that we are total strangers asking you for a favor. But if you do happen to have a chance to find the answers, we would very much appreciate knowing exactly how badly we are serving our female students.

Thanks again. (And by the way-- I very much enjoyed the paper. Thank you for writing it.)

Her reply (posted with permission):

Hello Dr. Herzog,

I am sure you have lost all hope of getting a response to this message now that it's been several months, but I kept your message in my inbox and knew I would respond someday. That day is finally here!

First of all, thanks so much for your interest in my work and for reaching out to me. It's been great for me to see people outside of social psychology respond to my work, and I'm happy to facilitate that any way I can (even if it takes me a while).

To answer your questions-

  1. For Study 1, I have the means/SDs for the question of how much they considered majoring in CS. Women went from M=1.17, SD=.39 in the stereotypical room to M=3.40, SD=1.96 in the non-stereotypical room. Men ratings did not differ across room type (stereotypical: M=3.00, SD=2.07, non-stereotypical: M=2.33, SD=1.94). I have started included more easily interpretable effect sizes (Cohen's d) in my more recent publications. I'm attaching one of them.

  2. Regarding the masculinity of these objects, I would argue that CS objects are not inherently masculine but have come to be constructed that way by society and could therefore come to be constructed as less masculine. Second, many fields (e.g., biology, chemistry, math) currently have masculine objects associated with them too (e.g., microscope, calculator), yet women have entered these fields to the point where they are the majority of graduates in biology and nearly 50% in chem and math. Research has shown that women are willing to engage with some degree and some types of masculinity. Diekman et al. (2010, Psych Sci) has an interesting perspective on this: Masculine fields that women have entered (e.g., medicine) are ones that are perceived as serving humanity and affording interpersonal interaction. Thus, changing students' perceptions of CS so that more women feel there is a fit between what they are looking for in a career and what the field affords them may be key to drawing more women (and some men!) into the field.

I discuss this exact issue, and other related topics, in a public televised lecture I gave last month. If you are interested, you can find the lecture here:

Stereotypes of Computer Scientists and Their Consequences for Women’s Participation," Sapna Cheryan. http://www.uwtv.org/programs/displayevent.aspx?rID=33000

One final thing to note is that you mentioned relating this paper to your female students. Although it's possible women who are already in the field would respond similarity to these stereotypes, our work has only tested the recruiting question (i.e., how to draw women in who have not already expressed an interest). Women in the field, because of self-selection or a change over time, will respond differently to these stereotypes. We are investigating that question now.

Hope that answers your questions. Please let me know if you have further questions!

If anything, it looks like the published paper understated how much of an effect these environmental cues had...