Possible explanations for the different results in case-control and cohort studies of the association between induced abortion and breast cancer include ________.
1) Differences in study design
2) Selection bias
3) Confounding factors
4) Random chance