Answers are grounded in this paper and cite their sources.
This article addresses the characterization and circumvention of Simpson's paradox in designs involving two ordered variables and illustrates the recommended procedure using a two-sample application involving the estimation of correlation and a single subject application involving the estimation of lag(1) autocorrelation.
No takes yet. Share an insight, caveat, or question.
Paul R. Yarnold (1996) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: