Effect sizeBMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6980.672 (Published 11 March 1995) Cite this as: BMJ 1995;310:672
EDITOR,—In their article on the development of a symptom based outcome measure for asthma Nick Steen and colleagues erroneously define effect size as the difference between baseline and follow up scores divided by the standard error of baseline scores.1 This is incorrect: the effect size is the difference between baseline and follow up scores divided by the standard deviation of baseline scores.2 This mistake has been made by others.3
The purpose of calculating effect size is to determine whether a change in health status is clinically relevant and to permit comparison with other instruments, interventions, or studies. If the standard error of baseline scores is used as the denominator the effect size will increase roughly in proportion to the square root of the number of subjects in the study, negating these advantages. Guyatt et al recommended using twice the average within subject standard error derived from two or more baseline observations as the denominator to derive an index of an instrument's responsiveness.4 This index would increase roughly in proportion to the square root of the number of replicate baseline observations.
Log in using your username and password
Log in through your institution
Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial