What Does Bias Look Like? | Hacker News Reader

What Does Bias Look Like?

theatlantic.com

24 pointscwan10 comments

Someone should point back to this article every time there is a discussion about the lack of women/blacks/latinos in tech. I don't generally go into a situation thinking people are looking to exclude me, but I've experienced instances of many of the institutional biases throughout my life.

I've always tried to be open to all sorts of people and situations, but it can be very hard to be surrounded by people who simply assume their way of being is the norm. I've been in situations where everyone except me is talking about the concert they went to. Why? Because no one thought to invite me since it wasn't a rap/r&b concert.

It makes me nostalgic for my high school years where there were so many kids of so many different ethnicities, incomes, and interests that we always had to take special care to understand each other.

> They're all right, of course: you can't simply infer bias from statistical underrepresentation

Technically speaking, if something is statistically under- or over-represented, then there is a source of bias ("an extraneous latent influence on, unrecognized conflated variable in, or selectivity in a sample which influences its distribution and so renders it unable to reflect the desired population parameters"). Whether that bias reflects overt or covert discrimination on the part of those in academia is another question, of course.

One meta-analysis of studies, for example, found that "discrimination against blacks was more likely to occur when potential helpers had more opportunities to rationalize decisions not to help" by invoking "justifiable explanations having nothing to do with race."

Isn't this saying that when people had reasons to make a decision not based on race, they actually did make a decision based on race. How can you actually produce that conclusion?

No, it's suggesting that, for example, when interviewing a black and a white candidate, each of whom has a reference letter and a set of test scores, the interviewer would favour test scores if the white candidate had higher test scores, but would favour reference letters when the black candidate had higher test scores. If only test scores or only reference letters were available, they'd hire based on merit.

In other words, people engaged in discriminatory behaviour when there was a way to rationalize/cover it up, and pretend they weren't doing it -- possibly telling themselves that, too.

Power.

That is the answer pure and simple. You can cloak it in deep discussions about statistics and attitudes and backgrounds. But there it is, naked for all to see. Liberals OWN acadamia and will not give it any of their power over the curriculum, philosophies, budgets and directions to any group who holds other ideas, much less ideas the may change the balance of power they hold.

Sadly, the most important part of her article, "We are never the best interrogators of our ideas. It requires motivated critics to lay bare our hidden assumptions, our misreading of the data, our factual inaccuracies" will go unoticed by the people who need to hear this information the most. It is ironic that the very essence of universities: scholarship and research, are the things most damaged by conservative discrimination. The ideals that should be held highest are the ones most damaged, and also the things LEAST cared about, behind money and power.

People will be people won't they?

Those people offered their own alternate theories [snip]

Another theory not in his list: Reality itself has a well-known liberal bias.

yeah, i hear that one all the time. based on some arguments i've gotten into with liberals recently (i'm a centrist), i can state somewhat definitively that it's not true.

You're using anecdotal evidence to refute anecdotal evidence. A bit surreal, at least to my mind.

heh. fair enough. i suppose i could go through the trouble of pulling out the various facebook threads in question, and redacting the names etc., so that i can back my argument up, but frankly, i'm not interested in doing that, and i doubt you're interested in it either.

if i'm wrong and somebody really really wants to know about the fights i've been getting into about the history of the second amendment, i'll do the leg work, but i figured for a quick comment here it wasn't necessary.

To be fair, he's using anecdotal evidence to refute an overly broad, unsupported generalization. Some evidence is better than none at all, but only barely.

Reply on news.ycombinator.com