High incorrect use of the standard error of the mean (SEM) in original articles in three cardiovascular journals evaluated for 2012.

Abstract

RationaleIn biomedical journals authors sometimes use the standard error of the mean (SEM) for data description, which has been called inappropriate or incorrect.ObjectiveTo assess the frequency of incorrect use of SEM in articles in three selected cardiovascular journals.Methods and resultsAll original journal articles published in 2012 in Cardiovascular Research, Circulation: Heart Failure and Circulation Research were assessed by two assessors for inappropriate use of SEM when providing descriptive information of empirical data. We also assessed whether the authors state in the methods section that the SEM will be used for data description. Of 441 articles included in this survey, 64% (282 articles) contained at least one instance of incorrect use of the SEM, with two journals having a prevalence above 70% and "Circulation: Heart Failure" having the lowest value (27%). In 81% of articles with incorrect use of SEM, the authors had explicitly stated that they use the SEM for data description and in 89% SEM bars were also used instead of 95% confidence intervals. Basic science studies had a 7.4-fold higher level of inappropriate SEM use (74%) than clinical studies (10%).LimitationsThe selection of the three cardiovascular journals was based on a subjective initial impression of observing inappropriate SEM use. The observed results are not representative for all cardiovascular journals.ConclusionIn three selected cardiovascular journals we found a high level of inappropriate SEM use and explicit methods statements to use it for data description, especially in basic science studies. To improve on this situation, these and other journals should provide clear instructions to authors on how to report descriptive information of empirical data.