We analyzed over 1,000 games in Woke Content Detector's list to see if wokeness and user review ratings are related.
"Woke" might be the most popular word in recent years. Not only do people frequently use it in discussion, but they also blame it for destroying culture and fun in many sectors, including games. But not many of those accusations are backed by the data when most discussions around woke topics are so subjective. So here we are.
Yeah, video games are *way* too woke. It kills the vibe.
— Elon Musk (@elonmusk)March 14, 2024
There are many discussions about the definition of woke. Merriam-Webster describes "woke" as
Our main data source, Woke Content Detector (a group with over 2,000 members that aim to label computer games with wokeness label), defines "Woke content" in their methodology as
In the Woke Content Detector's list, there are 0 games with 200 or more reviews on Steam.* It was grouped into 3 levels of wokeness:
1. Not Woke - 0 games
No Woke themes present (Recommended)
2. Subtle Woke - 0 games
Woke themes present with subtle/no specific messaging (Informational)
3. Overty Woke - 0 games
Woke themes present with overt messaging (Not Recommended)
Good and bad games are subjective and opinionated. We use Steam's game reviews as a collective good/bad indicator. Steam is a video game digital distribution service. People who buy games on Steam can give them a "Positive" or "Negative" review. When a game has enough reviews, Steam will show the percentage of positive reviews.
If a woke game is a bad game, now can you guess the positive review percentage that best separates "Not Woke" and "Overtly Woke" games? (Let's ignore subtle woke games for now)
Mostly
Overty Woke
Mostly
Not Woke
When divided into wokeness levels, "Not Woke" games seem to have the highest average positive rating at 88%, followed by "Subtle Woke" at 87% and "Overtly Woke" at 85%.
Look like the more woke the game is, the worse review it tends to be by an average of 3% between the "Not Woke" and "Overtly Woke" groups. The question is, does this different matter? In statistics, we can use "Cohen's d" to measure how large is the difference between the average of two groups relative to their sparsity.
We calculated Cohen's d value of the two most extreme groups, "Not Woke" and "Overtly Woke" and here is the result (let's focus only on 80% positive reviews and above to see the details).
M1-M2 = NaN
SDpooled = NaN
According to the Cohen's general guideline, Cohen's d of 0.28 can be interpreted as "The difference between two groups is small and difficult to see with the naked eye".
NO.
Using data from the Woke Content Detector, we cannot separate the game's wokeness with positive reviews percentage, and the difference in averages between the wokeness groups is too small to be significant.