Judged: two independent readers; copy check clear.

Lord’s paradox

Lord’s paradox Lord’s paradox refers to a misleading result that can appear when researchers study change from pretest to posttest while statistically adjusting for the pretest score.

Lord’s paradox refers to a misleading result that can appear when researchers study change from pretest to posttest while statistically adjusting for the pretest score. If the predictor is related to initial scores and those scores contain random error, this adjustment can create a regression-to-the-mean artifact that makes a variable seem to predict improvement when it may not.

Also see

Sources

Drafting source: Eriksson Kimmo; Häggström Olle. Lord’s Paradox in a Continuous Setting and a Regression Artifact in Numerical Cognition Research. 2014; DOI 10.1371/journal.pone.0095949; article XML SHA-256 95f0d4df6ca7f62304780c4d4913d7955933d516c80790889a140ad64c83f008. “Lord’s Paradox in a Continuous Setting and a Regression Artifact in Numerical Cognition Research” (10.1371/journal.pone.0095949).
CC-BY-4.0 · Definition uses new wording; original source retains its stated licence. · Source snapshot SHA-256: 95f0d4df6ca7f62304780c4d4913d7955933d516c80790889a140ad64c83f008

These citations identify the drafting sources. 2/2 independent reader passes against the pinned source packet. The drafter does not count as a reader. Source institutions do not endorse this dictionary.

Cite this entry

PsychTerms (psychterms.org), CC BY 4.0, version 1. “Lord’s paradox.” Development preview.

Version 1 · SHA-256: f67f6b7eeae9db96632e62dc571fe7760f135e2bff4f46414bc287c9a004d91e

Edit this entry on GitHub