4 ms·
To quote from the article, > We polled our users and asked them what they would be willing do donate and the average was twenty seven dollars, and of course we
by sqrt17 7y ago
To quote from the article,
> We polled our users and asked them what they would be willing do donate and the average was twenty seven dollars, and of course we have thousands of users
Which is based on the illusion that you get an unbiased sample when you ask users to respond to a poll. You don't - the people most likely to sacrifice a moment of their time are also those more likely to come forward with money, conversely those who didn't fill out the questionnaire probably have a majority of people who won't give any money.
- amelius 7y agoHow would you fix this kind of bias?
- sqrt17 7y agoThe only way I've seen this addressed is using the idea that late responders are more similar to non-responders than early responders - so in a study where you suspect that non-responders have a different bias to responders you'd use late responses to get an idea about what non-responding people would have answered.