How different methods for removing data outliers affect statistical accuracy
In short: A simulation study comparing ten different ways to exclude response time outliers found that the choice of method significantly impacts bias. Researchers tested how well various exclusion techniques could recover an original, uncontaminated distribution after introducing outliers either at the tails or overlapping with the genuine data.
The study used simulations to measure bias, defined as the deviation in the proportion of significant differences after exclusion compared to the original samples. While some methods produced high rates of Type-I errors and should be avoided, using z-scores or standard deviations introduced only small biases. However, the study was limited to a simulated environment and did not test real-world experimental data.
Why it matters: When cleaning response time data, using z-scores helps maintain statistical accuracy, while ignoring outliers or using certain specific methods can introduce significant bias.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
To cite this card, link to https://intellectplus.co.uk/app/science-research-how-different-methods-for-removing-data-outliers-affect-stat.
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