regression toward the mean
- noun
- /rɪˈɡrɛʃən tɔrd ðə miːn/
- Specialized
:
regresión hacia la media
- Teachers noticed a regression toward the mean in student performance after the initial test results, as scores became more consistent with class averages.
Examples
-
We believe these sources of error were greater at follow-up testing because of influences such as regression toward the mean, practice effects, and the absence of task novelty affecting speed of performance.
Academic text (2006) -
Although a complete discussion of regression artifacts is not practical here, it should be mentioned that whenever extreme scores are used on an instrument for classification purposes, there should be large amounts of regression toward the mean when a second test is given (Campbell & Kenny, 1999).
Academic text (2006) -
All this will happen because, as we statisticians like to say, we're experiencing a great regression toward the mean.
Blog text (10) -
As noted previously, from a psychometric perspective, part score variation can be explained by regression toward the mean.
Academic text (2013) -
There may also be some regression toward the mean, although over 10-year periods that effect is likely to be muted.
Academic text (2016) -
In other words, regression toward the mean can make it appear that students in special education programs are improving when the increase in their scores is due to a statistical artifact.
Academic text (2015) -
Because of regression toward the mean, we would expect, for individuals with ID, that some part scores would be far closer to the mean than the others and that those part scores demonstrating the lowest correlations with others in the profile and the lowest reliability would be affected most by this phenomenon.
Academic text (2013) -
For a particularly able group of students (as indicated by high reading and mathematics scores in the third grade), expected growth would be adjusted slightly upward because more growth would be expected of a more skilled group, and then adjusted downward to account for regression toward the mean.
Academic text (2001) -
The phenomenon of regression toward the mean explains why exceptionally high test scores are often followed by lower scores in subsequent evaluations.
-
In statistics, regression toward the mean occurs when extreme observations are followed by values that are closer to the average in later measurements.
Surface Forms
Morphology
regression + toward + the + mean
The components 'regression' (a return/movement in a statistical context), 'toward' and 'mean' (average) give a clear general sense of 'movement toward the average,' so learners who know the words can grasp a basic idea. However, the precise probabilistic nuance — that extreme values tend to be followed by values closer to the average on subsequent measurement — is a technical statistical inference not fully predictable from everyday word meanings, making the expression only partially transparent.
Etymology
Regression toward the mean paints a simple image: after a very high or very low result, things often move back toward the centre like a ball rolling to the middle. Here regression means 'moving back' and mean means the 'average', so the term explains why extreme results are usually followed by values closer to the 'average'.
