How racist and sexist was Survivor? A new study has answers

Over its nearly 23 years on the air, we’ve certainly had plenty of evidence of sexism and racism on Survivor: from conversations on camera to conversations kept from us, abusive behavior in the game and racism from fans, sexism from its players and repeatedly from its host.

Players have discussed it and written about it; I’ve asked CBS executives about it; others have analyzed data.

Now, researchers at the University of California Merced and University of Dayton—Dr. Erin M. O’Mara Kunz, Dr. Jennifer L. Howell, and Nicole Beasley—have just released a comprehensive study of the first 40 seasons of Survivor, which will be published in the journal Psychological Science.

In looking at the iconic CBS show, the researchers asked “whether there is evidence for racial and gender bias in the voting patterns of contestants on Survivor.”

What the ‘Surviving Racism and Sexism’ study reveals

The researchers did, in fact, “find evidence of racial and gender bias at multiple stages,” in particular “a systemic bias in favor of White men, and against women of color.”

That’s according to the pre-print—a draft published before “copyediting, typesetting, and review of resulting proof”—of the full study, which is titled Surviving Racism and Sexism: What Votes in the Television Program Survivor Reveal About Discrimination.

The scholars also published their research materials, including Excel files with data about all players in seasons 1 through 40.

Those Excel files are remarkable documents that have more than 130 data points for each of the 731 players from Survivor: Borneo to Survivor: Winners at War.

(Since season 40, CBS expanded its casting to be more diverse, and two women of color and one white man have won Survivor in the three seasons that have aired since.)

Instead of just looking at when players are voted out, like earlier studies have, the researchers set out to “examine bias at multiple critical points of the game,” including tribes’ very first votes, who was able to make the merge, who ended up in the final two or three, and who the jury ultimately chose to win $1 million (or $2 million for WaW).

There is ‘evidence of race & gender discrimination’

Dr. Erin O’Mara Kunz tweeted a thread about the study she and her colleagues conducted, noting that there is “evidence of race & gender discrimination in who is the first person voted out of their tribe and who makes the merge.”

Interestingly, she wrote that “We found no evidence of race & gender discrimination regarding who becomes a finalist, BUT we did find evidence of gender discrimination regarding who wins the game.”

Challenges are not a factor, she wrote: “The bias in winning holds whether you take into account the # of individual immunity challenges won or not.”

You can read the full study below. To summarize, though, looking at the first 40 seasons, they researchers came to this conclusion:

Compared to men, women have greater odds of being voted out of their tribe first, lower odds of making it to the individual-competition stage of the game, and lower odds of winning. Compared to White contestants, BIPOC contestants have greater odds of being voted out of their tribe first and have lower odds of making it to the individual-competition stage of the game.

That’s illustrated in this graphic:

The researchers noted that Survivor is particularly useful for study because of those 40 seasons’ similarity to the overall population of the United States: “the context demographically mirrors contexts where racial and gender disparities occur: The contestants on Survivor are largely White and most seasons have a ratio of White-to-BIPOC contestants greater than 2:1—a number mirroring U.S. demographics.”

There are some caveats, or as the scholars put it, “several limitations” of the study, including that “small sample sizes across the different racial groups preclude us from conducting analyses testing for discrimination toward specific racial groups.”

They also point out that “there are likely selection effects,” meaning contestants may not be representative of the general population, because producers have selected people “who will make for entertaining television.”

Also, they recognize that they cannot know why Survivor players made certain decisions, and their “racial and gender attitudes nor how they evaluated the physical strength and intelligence of their fellow contestants” are unknowable. They point out that “a more nuanced, qualitative analysis would be required to better understand contestant-level thoughts about their co-contestants.”

Given the complexity of the game, never mind the time that’s elapsed since most seasons were filmed, I wonder if that’d even be possible. For now, this study provides a very comprehensive look at Survivor’s game through the lens of race and sex, and provides data that might be useful to others studying the show in the future.