Code Review: A Blog

Big Brother’s Diversity Problem by the Numbers

A deep dive into casting data might explain why the reality show is constantly accused of bias

T

he image of three glum Big Brother contestants wearing obnoxious scouting uniforms while sitting in front of host Julie Chen-Moonves will go down as one of the reality competition’s most embarrassing moments. Since the show aired on CBS almost twenty years ago, diversity, or the lack thereof, has been a major source of criticism from fans. Network executives and producers have played down accusations of racial bias in casting decisions and among players for years. But this season’s twist, which forced the first three eliminated players (all of whom were people of color) to remain on the show until they could be ousted as a group, only accentuated the show’s perpetual issues with race, representation and diversity.

I was in my preteens when the American version of Big Brother premiered in 2000. I’ve followed every season since, including its spinoffs Big Brother Over The Top and Celebrity Big Brother. Over 24 cycles, I’ve watched the show evolve from fun social experiment—much like its European prototype— into one of the most cut-throat game shows on American television. For those who don’t know, Big Brother USA is a strategy game in which a group of (mostly) strangers are locked in a house together all summer long. During the lockdown, the contestants fashion a combination of physical, mental and social skills to survive weekly democratic eliminations. The last one standing wins $500,000. The game is generally one of the most-watched network shows during the summer months, with various platforms for audience engagement, including the CBS broadcast that airs three nights a week, a 24/7 online livestream of the house, a nightly three-hour livestream TV show on CBS’s Pop TV channel, and dozens of blogs, podcasts and amateur analysts giving daily updates and commentary about the going-ons of the house. That's not to mention the super active social media communities on Reddit, Twitter and other online forums.

What makes the show so compelling to watch is how the house inevitably becomes a real-life chessboard, where players orchestrate a number of calculated moves to stay in the game. They can try to earn safety by competing in weekly challenges, form strategic alliances, make deals with one another, lie, cheat, blackstab, throw one another under the bus, or even fake romances if it means surviving one more week. Most of this gameplay is done in secret, carried out through whispers in quiet corners and empty rooms. It’s secret, that is, to everyone but the millions of viewers listening in on everything via the 24/7 livestreams.

What makes the show painful to watch, however, is how year-after-year its casting, twists and structure give way to racism and implicit bias. In a game where the lines between strategic relationships and authentic ones are blurred, prejudices, both deliberate and unintended, run rampant. Every year, fans watch as a group of predominantly white, orthodoxly attractive, athletic “cool kids” form super-cliques and start picking off the “outcasts” one by one— at least until the outsiders prove they are just as trustworthy or competitive as their whiter, younger, fitter counterparts (if they survive long enough to make the case). Older, heavier, nerdier, darker contestants are often left out of these initial super alliances and, unprotected and outnumbered, end up being the first targets. Black women in particular are almost always described by other houseguests as being “shady” or “untrustworthy” just days into the game. In particularly sadistic seasons where surprise twists force the house to immediately eliminate a player before the group has had time to get to know one another, the older and minority players (being the elephants in the room) are almost always targeted. In Season 14, the first-ever Day One elimination in the series’s history saw contestant Jodi Rollins, the only black person and the oldest contestant in the game, infamously booted before she could unpack her bags. This year, a twist forced contestants to pick a leader within hours of first meeting one another. Despite a diverse pool of nominees, the house voted for an athletic, straight, white male housemate as their first leader. He ultimately used his power to force three ethnic minorities and the oldest contestant to compete in a challenge where the loser was immediately eliminated from the game. Ultimately, three of those contestants ended up being among the first four people eliminated (with one winning his way back into the game).

Whether the contestants are aware of their bias, or if there is any bias at all and not just some strange coincidence that occurs practically every year, is debatable. What’s clear is that CBS and producers are aware of the problem. Big Brother’s "race issue" has been well-publicized. The 2013 season made national news when several contestants constantly made racially insenstive and homophobic remarks about fellow players, both in private and to their targets' faces. Fans blamed the show for its lack of cast diversity, arguing that placing one or two people of color or LGBTQ people in a house full of straight white people only opened the door for that amount of segregation and bullying. There is also the argument that from a sheer numbers perspective a lack of diversity decreases the odds of a person of color winning the game, despite any racism or bias. Fans also accuse the show of editing, ignoring or downplaying controversial incidents during broadcasts, or even going so far as to warn problematic contestants to change their behavior, in an apparent effort to protect the players and/or the show’s image. With all this criticism and the country’s tense socio-political climate, one might wonder if the show is starting to improve in this area.

Not much. I created a database of the 300+ contestants who have been on the show. It includes their age, race, placement and sexual orientation, among other things. I looked at the data to see how diverse the show really is and by how much, if anything, things are changing. Here is what I found:

Less than 30 Percent of Big Brother Contestants are People of Color

Of the 307 contestants who have played the game so far, a little more than 80 are people of color. Black contestants account for 12 percent, while Asian contestants make up 4 percent. Latinx and Hispanic players account for roughly 7 percent of contestants.

Some would argue that this breakdown is similar to the racial breakdown of American society. They use that argument to dismiss calls for increasing diversity on the show. But that wrongfully assumes that the composition of reality show casts must reflect society’s demographics. The overwhelming majority of Big Brother cast members, for instance, are 20-somethings, even though 20-somethings account for only less than 15 percent of the population. Despite that, the show overrepresents the white population by 20 percent and underrepresents the Latinx/Hispanic population by 61 percent. Asian-Americans are also slightly underrepresented. So reflecting the country’s demographics doesn’t seem to be the reason behind Big Brother’s diversity problem. Something else is going on.

Big Brother is not (Really) Getting more Diverse

One would hope that Big Brother producers are listening to audience feedback and making good-faith efforts to diversify, but if changes have been made they have been weak. Sure, the last three or four seasons of Big Brother have been some of the franchise’s most diverse, but there is no steady increase, with earlier seasons being just as diverse, if not more, as recent ones. Even with slight improvements the show is nowhere near a level of equality or true diversity (a balanced representation of broad groups). To research this I used the “multi-group entropy index,” which measures the actual distribution of groups against a perfectly even distribution (an equal representation of black, white, Asian, Latinx/Hispanic, Middle Eastern & Arab, and other contestants). What I found was that the most diverse season of the series was Season 18, which was the only season to have more contestants of color than white contestants, and had the highest entropy index score. Seasons 11 and 20 were the second and third most diverse seasons. Few other seasons have come close to being as mixed.

Big Brother’s entropy scores are not the best, but using entropy to measure diversity can be somewhat flawed in that the categories it measures are arbitrarily chosen, especially with race and ethnicity being relative social constructs. Looking at the ratio of white contestants to brown contestants, though, Big Brother’s “diversity problem” is still painfully obvious. A ratio of 1:1 would be the most balanced result, but throughout Big Brother’s run, this ratio is rarely less than 2:1, meaning there is almost always at least twice as many white contestants as players of color in any given season. Season 18 was the only season to cast more people of color than white players and had the closest ratio to 1:1 (it was 0.8:1), other than Celebrity Big Brother 2, which was the only cycle to have a 1:1 ratio. The unfortunate takeaway, though, is that the ratio of white players to brown players is better than older seasons, but remains very unbalanced.

In other words, any efforts to diversify Big Brother have been weak, if not only very recent.

LGBT Representation Stable, but Skews gay

Things look brighter when it comes to LGBTQ representation. Adult members of the LGBT community make up 4 percent of the population, but account for 12 percent of Big Brother contestants. Only two seasons failed to have at least one LGBT cast member. It’s a fitting overrepresentation of a group that has historically been underrepresented in media. It also suggests that more can be done to overrepresent people of color, too.

But there are still a few issues. Among the major critiques of Big Brother’s queer casting is how it erases queer romance. Producers typically recruit an equal number of men and women then market the show by teasing viewers with the inevitable trysts, flings and romances that arise from the group. These “showmances,” are almost always cis, straight romances. It’s rare that Big Brother casts more than two members of the same queer identity, making it virtually impossible for the show to depict queer people’s capacity to fall in love or build requitted romantic relationships. Yet heterosexual romancing is a seasonal selling point.

Fans also say the show tends to cast LGBT members as exaggerated tokens whose sexualities are watered down by caricature. In line with this critique is that most LGBT players are gay men. There has only been one transgender player and five lesbian players. Contestants who casually identitfy as bisexual or questioning are rarely depicted on the show discussing the intracasies of their sexual identity. That is usually limited to the online livestream, while gay and lesbian cast members have their sexuality more front and center, including having it featured in production packages for the broadcast show.

Big Brother Casts are Overwhelmingly Young

Aside from being white, straight and extremely good looking, most Big Brother contestants are relatively young. Almost 90 percent of cast members were in their 20s and 30s when they were on the show, with most falling between 22 and 28. The only time there were more 40+ year-olds than 20-and-30-somethings was during both seasons of Celebrity Big Brother. Other than the second season of Celebrity Big Brother, when there were no 20-somethings in the cast at all, 20-somethings have dominated Big Brother casts.

Whether this is intentional is unclear. Julie Chen-Moonves told Parade Magazine that the show casts 20-something because they are more interesting to watch. But in another interview casting director Robyn Kass argued that sometimes fewer older people apply for the show because they think they won’t get cast, or can’t commit the time. That means she has a smaller pool to choose from, she said, adding that if she could cast more older people, she would.

Some casting directors seem to love using this “we just can’t find them,” line. The “pipeline” excuse is not only used in entertainment, but echoed across industries, from tech to design to gaming. But it has been debunked as a sham that uses a false perception of “quality” to hide old-fashioned bias. Reality shows on other networks like MTV, VH1 and Bravo appear to have no problem finding diverse pools of candidates. If the research is to be believed, the lack of diversity in reality TV is not a matter of quality or pipeline, but that the people in charge of making casting decisions simply don’t want to select (too many) people of color or have some implicit bias, even invisible to themselves, that prevents them from seeing some people of color (or older, or LGBT people) as viable candidates, when they are.

Do Demographics Impact Success?

When it comes to performance, black contestants face a mysterious challenge. For white and Latinx/Hispanic contestants an equal number of players have finished in the top 50 percent of players in their seasons as those who finished in the bottom 50 percent. That makes sense, assuming everyone had an equal and random chance at finishing in any place, 50 percent should have finished in the top 50 percent and 50 percent should have finished in the bottom 50 percent, more or less. However this isn’t the case for black contestants. Most black players, 64 percent, finished in the lower 50th percentile of their seasons.

It’s unclear why this might be, though some fans have theories. Taran Armstrong, who co-hosts the daily Big Brother recap podcast Rob Has A Podcast, has touted a theory that minority players, specifically black players, did not start losing so early into the game until Season 15 when Big Brother began introducing complicated game twists that all but forced the houseguests to clique up immediately.

“It's these big massive format twists that I keep saying benefits the big alliances and makes people stick together,” Armstrong said in a recent podcast. “Ever since the introduction of those, the African American contestants have done considerably more poorly than prior to that.”

Turns out there may be some truth to this theory. The data shows that after Season 13, no black person has finished in the top 50 percent of players in normal editions of the show (not including spinoffs Big Brother Over The Top and Celebrity Big Brother, which had a higher-than-usual number of minority contestants than normal seasons). But before Season 14, the number of black players who finished in the bottom 50 percent was more or less equivalent to the number of black players who finished in the top 50 percent— interesting, considering Season 14 introduced the Day One eviction twist that sent Jodi Rollins, a middle-aged black woman, home six hours into the game.

Another theory blames implicit bias and low odds from lack of representation. After all, players’ longevity in the game depends on their fellow players’ votes. The problem with this theory is that other underrepresented groups have had better success. More than 60 percent of Asian-American players made it to the top 50 percent of players, with two of them going on to win the top prize. Meanwhile four of the five lesbian contestants to play the game have made it to the top five finalists, with one (also Asian-American) winning the game. The only black woman to win the game was Celebrity Big Brother contestant Tamar Braxton. Braxton’s season included the second-highest number of people of color in one season in Big Brother’s run.

If that image of two black contestants and the series’s first Bangladeshi-American player sitting in front of Julie Chen as the first three people eliminated from the game didn’t suggest that maybe Big Brother has a “diversity problem,” the data does. Big Brother is a social experiment that showcases (and exploits) human nature. The pressures of the game, its twists, and the isolation extract the most fascinating behavior from human beings. This includes racism and bias, as the game forces its players to make snap judgements about one another, form packs with those they immediately think they can trust, and destroy those they think they can’t. Representation might help combat the inequality that results from these human impulses. But if Big Brother is truly an examination of human behavior, one must wonder if the biases and blindspots fans see in the house also exist outside the house, including in the casting process and production decisions.

Read more about the methodology and process behind this project here. It also contains a link to the full dataset.


Updates and Corrections

A previous version of this post incorrectly stated the ratio of white contestants to contestants of color for Season 18 as .09. It is actually .08. The diversity bar charts also incorrectly calculated entropy and ratios. The charts and story have been updated to reflect the accurate data. This did not affect the section’s conclusions.

The first version of this post incorrectly classified Celebrity Big Brother 2's Ryan Lochte as white (Anglo-American). Being half-Cuban he should have been grouped with Latinx/hispanic. The post and data have been updated.


ABOUT THE BLOG

Code Review is a blog by Vince Dixon exploring the relationship between data-driven visual journalism and American society. Vince is a digital media journalist and communications specialist.