Hacker News Reader

Comment by andreasvc | Hacker News Reader

No you cannot. If being a Facebook user is highly correlated with being depressed, the figure could be considerably lower. See Simpson's paradox. If you take a subset of a dataset, you can find the opposite trend or a completely different mean compared to the whole population. This is because the subset can have particular features causing it to be not representative of the whole population.

View on HN

Can you elaborate? It seems like simple logic, if 300 million people out of the total population of humans have depression and Facebook has 1200 million users. Then at least 75% of its user base must not be depressed. Because if any more then 75% were depressed then there would be more then 300 million people with depression.

I doubt this is what the parent comment was suggesting, but...

~10% profiles are fakes/bots

x% are duplicates

x% are created as business profiles

FB has more fake profile than 186 out of 196 countries actually have people, in fact all but the top 10 countries by population. Therefore, more than 75% of FB users (distinct from profiles) could suffer from depression.

Oh, my mistake. I didn't realize that was what the 75% figure was based on.

By WHO statistics 350 million people suffer from depression. According to the article, 350 million = 1/4 of Facebook's user base. So at most 1/4 of Facebook users suffer from depression, so at least 3/4 do not.

Reply on news.ycombinator.com