Hacker News Reader

Comment by CryptoPunk | Hacker News Reader

Counter-evidence from a controlled experiment:

https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=2674...

tl;dr police were 25 times less likely to shoot an unarmed suspect in a simulation, if that suspect was black instead of white.

A little bit of common sense explains why. It would simply be inconceivable that the officer who shot Daniel Shaver to death would have been acquitted had Shaver been black instead of white. The kind of social backlash that an acquittal in such a case would elicit would ensure the political and judicial system punish the officer.

View on HN

First of all, you and the person you responded to are giving examples that covers wildly different types of interactions, so it's not "counter-evidence". It does provide additional evidence for a certain type of interactions, however.

With that said, there's a massive caveat here: This is based on simulations

For it to be generally applicable you also need to then demonstrate that this matched real-world observed behaviour. On its own it is not at all obvious whether or not this accurately reflects anything at all.

There are many possible confounding issues here. E.g. for starters a typical excuse in shootings is that the officers claim to have feared for their safety. You then need to account for whether or not the simulation accurate trigger the same fear, otherwise someone with a racial bias in what makes them scared in "real life" could be expected to act differently in a simulation where they know they're safe.

I don't know whether that is the case or not, but there are many enough possible issues like this that a simulation only allow us to draw very limited conclusions here without a far more complicated experiment. Without understanding why.

Yes, data from real cases has significant advantages over a controlled experiment using simulations, but also significant disadvantages.

Still, a 25 times lower likelihood of shooting a suspect if they are black in the simulation-based experiment..

That is entirely corroborated by what common sense would predict given the backlash a typical police officer would expect in the event that they shoot and kill a black suspect versus a white one.

I assign zero value to "common sense". "Common sense" to me would also suggest that people would reveal less of their real biases and more of their concern of what those observing them might think when in an artificial setting where they know their actions will be scrutinised. We can come up with all kinds of "common sense" ideas about what might happen to explain this, and they're all nothing more than wild guesses.

Common sense informs what hypothesis to test and what axioms to accept.

Putting common sense aside, a 25X lower likelihood of shooting unarmed black suspects in the experiment suggests massive bias in favor of black suspects.

Assuming it suggests bias in favor of black suspects is assuming a hell of a lot about the accuracy of the simulation relative to real life and whether or not it captures bias or peoples beliefs about how they are expected to act or changes to e.g. fear levels, that are not supported (it's also not contradicted - there's as far as I can tell no data either way) by the experiment in question.

All it tells us in isolation is that this is how a set of people in current society acts in a simulation of that kind. It does not tell us why, and trying to draw conclusions about why without further experiments is irresponsible.

The experiment would have to be flawed to an absurd degree for the findings to not be an indication of bias in favor of black suspects.

Even multiple minor/medium sized flaws in the experimental design that produce black favoritism not reflected in real world behavior wouldn't in the aggregate account for producing a 25 times exaggeration of bias.

The bias is undoubtedly there.

> It would simply be inconceivable that the officer who shot Daniel Shaver to death would have been acquitted had Shaver been black instead of white

Why is that inconceivable? The officers in the Breonna Taylor shooting weren't even charged. At least this officer went to trial and was found not guilty by a jury of his peers.

Because there is a video of Shaver crying for 2 minutes, while displaying zero defiance and total willingness to cooperate, before the officers shot him.

I cannot even imagine the kind of nation-wide riots that scenario would generate as the video of the shooting, which would be attributed to racial bias, circulated on social media.

That doesn't at all square with the actual data we have though. Why look at a "controlled experiment" instead of a wealth of real-life data?

The real life data we have is not obviously indicative of racism. Men are more than 10X more likely to be shot by police than women. Asians are less likely to be shot by police than whites. This doesn't prove police are sexist against men, or racially biased in favor of Asians.

Reply on news.ycombinator.com