Dolls Who Code: American Girl’s “Courtney” is the game developer the ’80s needed.

A doll wearing a denim skirt, crop top, Walkman, and side ponytail in front of a Pac-Man arcade game.
Courtney at the arcadeAmerican Girl Dolls

The mid-’80s was a notable time for women in computer science—because that was when they started disappearing. From 1970 into the start of the 1980s, the percentage of computer science degrees conferred to women rose, peaking at 37.1 percent in 1984. But this number then dipped drastically, and we’ve never recovered. The most recent report from the Computing Research Association shows the number of women graduating with computer science bachelor’s degrees, in its sample of U.S. institutions, in 2019 at 21 percent.

Chart of percentage of women with computer science bachelor's degrees
Source: IESNational Center for Education StatisticsNatalie Matthews-Ramo/Slate

So, after decades of women representing both pioneers of computing and a large percentage of the day-to-day programming workforce, what happened in the ’80s? One explanation is the rise of the personal computer. Those computers were marketed in part as toys (you can play Pong!)—and they were also marketed at boys.

These marketing decisions have had profound, lasting impacts. As explained in Jane Margolis and Allan Fisher’s 2001 book, Unlocking the Clubhouse: Women in Computing, the problem had to do with exposure. By the mid-’90s, students arriving at college to study computer science had already been playing with computers for much of their lives. They were also all almost all male—and boys were twice as likely as girls to have been given computers for gifts as children. Together, these differences resulted in the proliferation of stereotypes that suggested who could be a computer scientist as well as in women arriving in college with less experience with computers. Many subsequently felt not only out of place but also behind from the start.

Stereotypes are hard to unwind, especially when they are constantly reinforced by media portrayals and marketing. And sometimes, when we take a step forward, we take another step back. For example, in 2010 (almost 20 years after Barbie told the world that “Math is tough!”) Mattel released Computer Engineer Barbie, representing one of Barbie’s first careers in STEM. However, a few years later, the book that had accompanied the doll, Barbie: I Can Be a Computer Engineersparked controversy when a blogger brought it to the world’s attention. The plot: Barbie needed boys to help her write the code for a game she was designing (“I’m only creating the design ideas, I’ll need Steven and Brian’s help to turn it into a real game!”) and to fix her and Skipper’s computers after she infected them both with a virus.

I became part of this controversy when, in a fit of dissertation-fueled procrastination and righteous indignation, I created a feminist remix of the entire book, rewriting the narrative so that Barbie was a competent computer engineer (and, sure, Steven and Brian could help too).
Following this controversy, Mattel publicly apologized, and in 2016 it introduced Game Developer Barbie. Unlike Computer Engineer Barbie, whose computer screen literally had 1s and 0s on it, Game Developer Barbie could actually code! Subsequently, I did some consulting with Mattel on a computational thinking book that accompanied the release of 2018’s Robotics Engineer Barbie.

But another Mattel product that seemed like a fix-it attempt presented other problems. With the 2017 Barbie STEM Kit, in which Barbie wears a lab coat and stilettos, children can help Barbie with such engineering projects as creating a washing machine and a jewelry rack. Similar to IBM’s #HackAHairdryer campaign, which encouraged women to “blast through the bias” by sharing engineering projects related to hair dryers, there seems to be some idea that women will only be interested in STEM if you can make it “girly” enough. (One rocket scientist responded, “That’s ok … I’d rather build satellites instead.”)

Perhaps other toy companies can learn from Mattel’s missteps, however, and even bring us back to an era when the number of women in computer science was still on the rise. This week, American Girl announced its newest doll: Courtney. “Real to the max” in 1986, Courtney listens to her Walkman, plays Pac-Man at the arcade, and codes her own video game. American Doll has also partnered with Girls Who Code to raise money for girls pursuing further education in computer science.

The description for the accompanying book, Courtney Changes the Game, informs us that Courtney is the best gamer at the arcade but “she can’t understand why there aren’t more girl characters in the games she plays.” So she decides to make her own game with a girl hero. (This kind of motivation—that someone might learn to code in order to increase their own representation or amplify underrepresented voices—is actually a focus of my own research on broadening participation in computing.) One difference between Courtney and the STEM Barbies (or even more direct learn-to-code dolls) is that her gaming and coding skills are part of the story, but not the entire story. As the trailer for a Courtney movie reveals, she has a lot going on: Her mom is running for mayor, her sister doesn’t like her guinea pig, she’s really into Care Bears, and … the Challenger explodes. Including coding as one aspect of a character helps to normalize computer science—just one more thing a girl can do as part of a full life.

Maybe Courtney can follow in the footsteps of game designers like Roberta Williams—who helped establish video games as a storytelling medium with graphical RPGs like King’s Quest and was recently featured in the Netflix documentary series High Score—along with other women in computing who “changed the game” in the ’80s. And maybe if we continue to think carefully about media portrayals, marketing, and ways to normalize coding for everyone, we can all go back to the totally rad era when the number of women studying computer science was on the rise. 


Leave a Reply