Simple rules of thumb fail to determine required sample sizes for regression
In short: Researchers evaluating common guidelines for conducting multiple regression analyses found that using simple constants or fixed ratios of subjects to predictors is unreliable. The study suggests that determining an adequate sample size requires accounting for both the number of predictors and the expected effect size.
The study compared various rules-of-thumb against power analyses for tests of multiple and partial correlations. While some formulas involving the number of predictors showed promise, they were limited because they assumed all studies have a medium-sized relationship and can yield impractical values as the number of predictors increases.
The researchers introduced a more complex rule-of-thumb that estimates minimum sample size as a function of both effect size and the number of predictors.
Why it matters: Researchers should use sample size determination methods that incorporate effect size rather than relying on fixed numbers.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Research: Green (1991). How Many Subjects Does It Take To Do A Regression Analysis. Multivariate Behavioral Research.
To cite this card, link to https://intellectplus.co.uk/app/science-research-simple-rules-of-thumb-fail-to-determine-required-sample-size.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Science · Latest · A–Z · All topics