The Regression Effect

"The 'regression effect', also known as 'regression to the centre', is initially difficult to understand. This effect can be described in abbreviated form as follows: Measured values of people tend to be closer to the mean value when measured again. How can this happen? Are statistical ghosts at play here?

It is obvious that such an effect has great significance for the interpretation of effects If, for example, in clinical efficacy studies, the mean values of the people examined change between two measurement points, one would like to attribute these effects to the treatment and not to incomprehensible statistical oddities. In the context of developmental psychology, the regression effect is described as ubiquitous (Furby, 1973, p. 172), and Lord sees this as one of the main reasons why "studies of growth may become confusing or confused" (Lord, 1963, p. 24).

Much is written about the regression effect, but it is rarely made explicit. Many descriptions of the regression effect, even in relevant textbooks, are abbreviated and do not provide a comprehensive understanding of the phenomenon. Sometimes regression effects are described as statistical artefacts caused by a lack of reliability of the measuring instrument ß (Bortz & Döring, 1995, p. 517). In other textbooks, the regression effect appears primarily as an effect of the use of regression lines (e.g. Mortensen, 1995, p.109). The matter is therefore difficult to understand for non-experts. The aim of this article is to provide an understandable and clarifying introduction to this topic."

Source: Nachtigall C., Suhl U. (2002). Der Regressionseffekt Mythos und Wirklichkeit. methevalreport. Psychological methodology and evaluation research at the Institute of Psychology at Friedrich Schiller University Jena. retrieved on 12 October 2022 from https://www.metheval.uni-jena.de/materialien/reports/report_2002_02.pdf