Hi, I reviewed a paper recently, and the authors had used standard deviation (SD), and I queried the use of this as opposed to the standard error of the mean (SEM). The reply was that the SEM "should NEVER be used". I was under the impression that the SD is used to describe the data, whilst the SEM is used to assess how precisely the sample mean matched the population mean. Certainly one should specify which measure has been used when writing up the data, but I was unaware of a viewpoint that the SEM should not be used. The link I was given is 

 http://www.sportsci.org/resource/stats/meansd.html

Your thoughts please, as this site was written in 2003, yet there are many later references on the web to the use of SEM and SD in the context I have described, and is how I use SD and SEM since my statistical training in 2010-2012.

More Richard Michael Payne's questions See All
Similar questions and discussions