Type III error

Show Summary Details

Quick Reference

A term occasionally used to refer to an error arising from a misinterpretation of the nature of the scores being compared in a significance test, as when a significant difference in an interrupted time-series design is misinterpreted as relating to a comparison between the mean of the baseline scores and the mean of the post-intervention scores, whereas the true comparison is between weighted means of the raw scores, the weights depending on the components of the statistical model (such as autoregressions and/or moving averages; see ARIMA). Compare Type I error, Type II error. [Named by the US psychologist Bradley E. Huitema (born 1938) by analogy with Type I and Type II errors]

Subjects: Psychology.

Reference entries

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.