Attrition bias is defined as which of the following?

• A. Systematic differences between groups in withdrawals from a study
• B. Random sampling errors
• C. Observer bias
• D. Selection bias

Answer: (A)

Attrition bias happens when people drop out of a study in a way that differs across groups, so the final groups being analyzed aren’t comparable. If more participants with certain outcomes leave one group than another, the remaining data can mislead about the true effect of the intervention—either exaggerating or understating it. This undermines internal validity because the comparison at follow-up reflects a biased subset rather than the original randomized groups.

This is different from random sampling error, which is random fluctuation due to sample size; from observer bias, where the person measuring outcomes is influenced by expectations; and from selection bias related to how participants are chosen or assigned, which can create differences before follow-up even starts.