A researcher studying memory retrieval e-mails a one-time online survey to 1,000 adults asking how often they experience the tip-of-the-tongue phenomenon, in which a known word temporarily cannot be retrieved. Only 200 people complete it, and those who do report far more frequent tip-of-the-tongue lapses than is typical. Which type of bias is MOST illustrated here?
- AAttrition bias, because a large share of participants dropped out over the course of a longitudinal memory study
- BSocial desirability bias, because respondents wanted to appear to have better memories than they actually do
- CObserver bias, because the researcher's expectations about memory lapses influenced which participants chose to respond
- DSelf-selection bias, because the people who chose to respond likely differ in how often they have memory lapsesCorrect
Why D is right
This is self-selection (volunteer) bias: when participants decide for themselves whether to take part, those who volunteer can differ systematically from those who do not. People who often have tip-of-the-tongue experiences, a retrieval failure studied in the memory topics of Unit 2 (Cognition), are likely more interested in a survey about them, so the 200 respondents overrepresent frequent sufferers. The results therefore cannot be generalized to the whole population of 1,000 adults, much less to adults in general.
Why the other 3 choices miss
- A
- Incorrect because attrition bias involves participants dropping out partway through a study that follows them over time; this was a single, one-time survey, and the problem is who chose to respond at all, not who quit later.
- B
- Incorrect because social desirability bias would push respondents to under-report memory lapses to look capable; here respondents reported unusually many lapses, and the issue is who responded, not how they slanted their answers.
- C
- Incorrect because observer bias occurs when a researcher's expectations distort how data are recorded or interpreted; the researcher did not choose the respondents, the respondents chose themselves.