It’s a twist even more surprising than the end of Frozen: Disney princesses may not be so bad for children after all. For years books like Peggy Orenstein’s Cinderella Ate My Daughter and studies on the short-term impact of watching, playing with and dressing up as Disney princesses suggested that those fairy tales starring waifish women could have a negative impact on girls’ body image, and the more retrograde princesses—the damsels like Sleeping Beauty and Snow White who do little but wait around to be saved by princes—might reinforce antiquated ideas about women’s helplessness for girls and toxic masculinity for boys.
Sarah Coyne, a professor and researcher at Brigham Young University, played a major role in the debate over princess culture. Inspired by her own anxieties over her daughter’s obsession with princesses, she conducted a study with 307 4- and 5-year-old children (about half of whom identified as boys and half of whom identified as girls) in 2012 and 2013 in which she asked children and parents how often they watched Disney princess films and TV shows, how often they played with princess toys and asked them to rank their favorite toys from a box of play things that are typically considered “feminine” (tea sets and dolls), “masculine” (action figure and tool set) and “neutral” (puzzles and paint sets). She found that high engagement with princess culture was associated with more female-stereotypical behavior one year later for both boys and girls—a reassuring result for parents of boys who love princesses and grow to be more in touch with their emotions, but a scary one for parents trying to find ways to teach their daughters strength and independence. The results launched 1000 headlines about how gendered play reinforced negative stereotypes for girls and rankled many princess-positive parents.
But now Coyne has published a follow-up study in which she interviewed about half of the children in that same group (as many as were willing to participate five years later), now 10 and 11 years old. She asked them to indicate how much they related to stereotypical gendered statements like “Swearing is worse for a girl than a boy” or “I like babies and small children.” She was shocked to find that girls who were obsessed with princess culture at 5 were actually more likely to hold progressive views about gender roles—to advocate for both female empowerment and for men to express their emotions—at age 10. And there was no discernible difference in body image for children who played more or less with princesses. In fact, princess culture seemed to have a positive effect on body image for children from lower-income families, upending Coyne’s expectations that princess culture would lead to obsession with weight and image.
In fairness to Disney, their princess movies do center on women’s stories, perhaps tacitly communicating that women are worthy of their own narrative. But it’s also worth noting that the first study was conducted before the release of Frozen, a massive hit in which a princess’ “true love” is not her suitor but her sister, which launched the most successful animated franchise ever and changed the trajectory of the genre. The princess movies that followed, like Moana and Raya and the Last Dragon, explicitly aimed to empower young girls with stories full of adventure and devoid of romance. More subtly, they’ve also tried to model empathetic and egalitarian behavior in the men who play supporting roles in those movies.
Coyne spoke to TIME about the many surprising results from her study and what Disney could still improve about princess culture.
What were your expectations going into the study?
This is a follow-up study from what we did in 2016 when kids were 4 to 5. That study found that girls, especially, who were really into princess culture tended to be more gender stereotyped a year later. So I expected that to continue onward. But we didn’t find that at all.
And in the early study, we didn’t find any impacts of princesses on body image. And that’s actually why I followed up again when they were 10 and 11. I thought at age 4 most kids have pretty good body image, but I thought that early princess exposure would be related to worse body image in early adolescence. Again, nope. It was the other way around.
O.K., so when it comes to views of gender, what changed for those girls who were more gender stereotyped in the first study and developed more egalitarian views over five years?
From ages 4 to 5, kids’ perceptions of gender are fairly rigid. And when we look at gender development as a whole, things become a little bit more flexible over time.
The kids’ favorite princesses also changed dramatically from when they were 4 to when they were 11. Early on, they liked the more gender stereotyped princesses like Cinderella, Sleeping Beauty, Little Mermaid. And when they were older, we asked them again, and they liked less gender stereotyped princesses like Mulan, or Elsa, Moana, the princesses who are more independent, whose stories are about adventure rather than romance.
I wonder if as you get older and understand the story more it becomes less about the signifiers of a princess—the dress, the sparkles—and more about what she’s actually doing. And those early princesses from the ‘50s don’t have much to do but be saved, whereas the modern princesses get to go on these big adventures that usually have nothing to do with falling in love.
I think so. We also looked at who the children most closely identified with at 4 years old. And we assumed that a kid who identified with Cinderella would be very different from a kid who identified with Merida [from Brave] a few years later. But they weren’t.
I wonder if it’s just that princess culture as a whole does a great job of putting princesses at the forefront of the story. And there are lots and lots of different princesses. So I suspect that a lot of the kids in our sample liked more than one princess. And they were exposed to all of these women, many of them strong and independent women in leading roles. So I think that has an impact over time.
You got a lot of hate mail when you released that first study saying princesses had a negative impact on gender roles. What has the reaction to this study been?
I’ve published a lot of things over the course of my career, like over 100 things, right? And I have never received such a crazy response as to that study. People called for me to be fired. People sent me physical hate mail telling me how useless I was. They called me personal bad names and said I was a bad mother. It was wild.
This one has just been out for a hot second here, but BYU posted it on their Twitter account. And again it’s just hatred, calling for me to be fired again, saying I have a leftist agenda, calling for the president of BYU to be fired. I learned that people really have strong feelings about princesses.
I personally was surprised by the results of the study.
Just because there have been so many shorter-term studies suggesting that ideation of these unnaturally thin women negatively impacted girls’ body image. Barbie launched a curvy doll exactly because people had been complaining for 50 years she was bad for girls’ body image. There’s been a whole movement to obliterate the pink aisle in toy stores. And, yes, Disney princesses are getting more diverse, but they’re all still thin with big eyes and big hair and the dolls are usually sold in dresses, even if they never wear those dresses in the movies.
Yes. The marketing and merchandise seem to always be a step behind the actual movies. In the early study, when we asked the kids who their favorite princess was and why, “I like Rapunzel because she’s blonde” was our No. 1 answer, which was really disturbing to me. I think we had one kid that said, “Mulan because she saved China.” And that stuck with me all these years that only one kid said that. And I just thought as parents, we have a responsibility to talk about media with our kids, to help them make good choices, to help them become critical viewers.
Body-image issues are still clearly a problem among young girls. So if it’s not Disney princesses’ fault, whose fault is it?
Barbie. I think you can blame Barbie. [Laughs.]
This was one that I was like, “Oh my gosh, I can’t believe this had no impact.” In fact, for low-income kids, [Disney princesses] had a positive impact on body image. I’m trying to wrack my brain about why that would be because study after study shows if you’re exposed to the same body ideal over and over again, that has a negative impact on body image.
My only theory is when you think about the stories themselves, a lot of them focus on finding the beauty within. I think of Belle from Beauty and the Beast. And although Belle is very thin and beautiful, the whole message of that story is look beyond appearances and focus on the person within. Maybe that gets through somehow beyond just the beautiful yellow ball gown and the tiny waist. But I don’t know. I was flabbergasted when I didn’t find that effect.
Whereas Barbie doesn’t have a backstory, really, for children to latch onto. She’s literally just a body.
You alluded to a surprising socio-economic difference in the study.
Basically we found that princess culture was more positive for kids coming from lower-income families than those coming from middle or higher. And this was fascinating because princesses themselves come from high socio-economic status usually. They’re royalty. They live in castles. They have expensive clothing. Psychological theory would suggest that we model those that we perceive as most similar to ourselves.
But maybe it’s because there’s also a lot of these rags-to-riches kind of stories. Tiana is one that comes to mind who works really, really hard for her job and she has all these ambitions and at the end gets her own restaurant.
I know you pulled kids from Utah and kids from Oregon from your study. Did you ask the parents about their politics?
We did not. But we are collecting data right now on princesses in the more diverse sample where we do ask about politics. And so watch this space. I suspect it will have an impact.
There needs to be more research in this area because our study was 87% white kids. I don’t know how generalizable it is to all sorts of different populations.
Is it possible that parents are just sitting down with their kids while they’re watching these movies and talking through some of the more problematic tropes with them, offsetting possible negative effects?
I don’t have an answer to that. In the early study we asked how often they watch media with their kids and how much they talked about it. And what we found is talking about the movie resulted in the kids being even more gender stereotyped. I wonder if the way that parents talk to their kids about princesses changes over those five years. Early on I think that they focus really on a lot of the appearance-related things, kind of the glitz and glamour of princesses. That’s not the case by 10 or 11.
We asked parents themselves why they liked princesses and why they don’t. And we had a wide variety of responses. Some love ’em, some hate’ em. I remember a mom saying, “I’m a super feminist. I vowed to never have my child watch princesses, and I was powerless to stop it. Like all she wants to do is princesses, and I’m freaking out.” I think a lot of mothers are like that. Maybe the study will just allow people to relax a little bit and find the magic in the princesses while still talking about them with their kids and not just buying your kids a pretty dress.
Do you think there’s anything Disney can do better in terms of their princess content?
They’re doing better in terms of representation of race and ethnicity but representation of body size and shape—we have this unrealistic body still. Moana is the one princess I can think of that is at least average size. She is very muscular. But we don’t see any princesses with a large body type. And in terms of representation and helping kids with their own body esteem, I think that would be a delightful move, to be able to show princesses of all shapes and sizes, with all colors as well.
But what happens when Disney reads you saying this and thinks, “But your study just said it didn’t matter so we can make as many skinny princesses as we want”?
Damn it. Look, probably for the majority of people, it doesn’t have an impact. But there’s probably one or two kids out there who maybe are a little bit overweight, who are thinking, “Where am I? Who am I? I don’t see myself. Does that mean I’m not pretty enough? Does that mean I’m not good enough?” So it’s probably like a smaller minority that maybe got washed away in the larger data set. We didn’t examine how much each child weighed and then moderate by body type or size—or for that matter, many other factors. So that would be something interesting to look at in the future to see if certain groups of kids are higher risk.
How do you feel as a parent about the results of your study?
I feel reassured. I really do. I’ve done a lot of research on all sorts of topics, but this one has also really changed me as a parent. We let our kids watch princesses, have dolls, all of that, which you might find interesting given our early study.
My daughter Hannah inspired the study when she was 3 years old. She loved princesses. Loved to dress up. Everything about her room was princesses. And I read Cinderella Ate My Daughter by Peggy Orenstein, and I thought, “Oh my gosh, I’m destroying my child. That’s it. She’s gonna have an eating disorder when she’s 13.” And Peggy said there hasn’t been a lot of research on this topic. And I thought, “I’ll do the study. Let’s find out how it actually impacts kids.” And we found that early study had kind of those negative results for girls.
But instead of banning princesses, which I think I could have done—she was 5 or 6 at the time when the results came out—I decided to really shift the way that I talked about princesses and focus on the personal qualities. When the study came out, Brave had just come out in theaters. And I’m like, “This is not gender-stereotyped princess at all. This is a fantastically brave, strong, independent woman.” And so I really used princesses as an opportunity to discuss what I consider to be an example of strong womanhood. And we would point out things—like when she had a Merida doll, it didn’t look like Merida from the Brave movie; it was more feminized and slimmer, and they took away her bow and gave her a sparkly sash. And we were able to have that conversation about the differences.
And then the second thing it changed was how I parent my son. So I have one daughter and four sons. One of my sons Liam adores princesses. He was obsessed with them as a young child, loved to dress up like Elsa, has all the dolls. Still loves princesses at age 8. And the study allowed me to just be really relaxed about it. I live in a really conservative area. I live in Utah where a lot of parents kind of look funny at their son if they want to do that kind of thing. But I think because I did the earlier research that found that it was great for boys, I was able to let him explore that side of himself. And this kid’s fantastic. He’s this budding little feminist who will call out sexism every time he sees it. There’s all sorts of things that can contribute to that, but I’d like to think maybe with our new results, there’s a small part of princess culture that helps them see these amazing strong women in film, as opposed to superheroes, which are predominantly the hyper masculinization that we often feed our little boys.
Right, in addition to finding that Disney princesses were actually good for girls’ perception of gender, the study found that boys benefit too. Why do you think that is?
I think it’s seeing strong women onscreen. But I also think the more recent films have more positive portrayals of masculinity. One of my favorite characters is Christoph in Frozen. He’s not like your muscular, ideal manly man. He’s a little softer, and he sings about his emotions. He works together with Ana. He asks for consent when he kisses her for the first time. It’s this beautiful egalitarian relationship. The male characters are changing at a far slower rate, I would say, on the whole, but we’re getting somewhere. We’re making progress.