5 ms·
This sounds like self selection bias. Only 12k of 28k respondents included their income. If you decrease a sample artificially then the resulting statistics are
by ianai 9y ago
This sounds like self selection bias. Only 12k of 28k respondents included their income. If you decrease a sample artificially then the resulting statistics are all suspect. (I stopped reading once I saw the attrition rate)