controlling for background variables
A statistical method of adjusting comparisons for confounding factors—such as family wealth, parental education, and home assets—so that observed differences can be attributed to the schools rather than to pre-existing pupil characteristics.
As defined in James Tooley on Private Schools for the Poor and the Beautiful Tree
Where it comes up · 1
Most important claims · 2 of 2Search all claims for “controlling for background variables” →
- Testing roughly 35,000 children in math, English and another subject, and controlling scientifically for background variables like family socioeconomic status, maternal education and wealth proxies (TV or radio ownership), shows children in private schools categorically outperform government-school children—so the gap is not merely brighter or wealthier pupils self-selecting.James Tooley on Private Schools for the Poor and the Beautiful Tree
- Even after controlling for measurable background variables, unobservable traits such as parental drive, passion, commitment and love cannot easily be captured on a form, so parents who choose private schools may differ in ways that confound the comparison.James Tooley on Private Schools for the Poor and the Beautiful Tree