The bias and precision of reporting the average age of human participants
Year: 2024
Type: article
Abstract: Most research papers in psychology study the behaviour of a sample of participants. To characterise this sample, authors report various characteristics, frequently including the mean age and the associated standard deviation. However, based on reports from authors who publish in Acta Psychologica and from respondents on X/Twitter, the present paper shows that some authors use rounded-down ages whereas others don't, which lead to an uncertainty of 0.5 year in the average age. The results furthermore show that the authors tend to report the average age with two decimals precision, irrespective of the uncertainty of this average. I recommend that publications should explicitly mention how the average age is determined and report its value using a number of decimals that reflects its uncertainty. • Authors report mean age without disclosing whether it is based on rounded-down age. • This causes a potential bias of half a year in the reported value. • The number of decimals generally suggests more certainty than the data provide. • Two recommendations are made to ensure reliable age-reporting.
Source: Acta Psychologica
Author: Jeroen B.J. Smeets
Institution: Vrije Universiteit Amsterdam
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,030
OpenAlex ID: W4401485926