Range Restriction — When Logic Wears a Disguise
Range restriction occurs when the variability in one or more variables is artificially reduced, typically through sample selection, truncation, or censoring. When a variable's range is restricted, correlations with other variables are attenuated — they appear weaker than they truly are in the full population. This can lead to incorrect conclusions about the strength or even existence of relationships between variables.
Also known as: Restriction of range, Truncation effect, Selection artifact
How It Works
Selection processes that filter out extreme values are common and often unnoticed. Researchers may not realize their sample has been pre-selected on the variable of interest, and the resulting weak correlations are taken at face value rather than recognized as artifacts of restriction.
A Classic Example
A company studies whether SAT scores predict job performance but only examines current employees, all of whom had high SAT scores sufficient to be hired. The restricted range of SAT scores makes the correlation with performance appear near zero, leading the company to conclude SAT scores are useless — when in the full applicant pool, the relationship is substantial.
More Examples
A sports psychologist studies whether mental toughness scores predict athletic performance among Olympic sprinters. Because all athletes at that level have already been filtered for exceptional mental toughness, the scores cluster tightly together, and the correlation with race times appears negligible — even though mental toughness strongly differentiates performance across the broader population of athletes.
A dating app analyzes whether profile attractiveness ratings predict match success, but only examines profiles that received at least 50 swipes — effectively excluding the least attractive profiles. The restricted sample shows almost no correlation between attractiveness and matches, because the variability in attractiveness within the included group is too narrow to detect the true relationship.
Where You See This in the Wild
Common in personnel selection research, university admissions studies, and clinical populations where only patients above a diagnostic threshold are included.
How to Spot and Counter It
Identify whether sample selection has reduced the range of key variables. Apply statistical corrections for range restriction (e.g., Thorndike's correction formulas). When possible, obtain data from the full, unrestricted population. Report the range of observed values and compare it to the expected population range.
The Takeaway
The Range Restriction is one of those reasoning errors that sounds perfectly logical at first glance. That's what makes it dangerous — it wears the costume of valid reasoning while smuggling in a broken conclusion. The best defense? Slow down and ask: does this conclusion actually follow from these premises, or am I just connecting dots that happen to be near each other?
Next time someone presents you with an argument that "just makes sense," check the structure. The feeling of logic is not the same as logic itself.