The goal this month is to relate the importance of sample size as told by investigators and statisticians themselves. Sample size can cause considerable difficulty for researchers and is very likely the most common reason for consulting a statistician. Unfortunately, the dilemma increases for those who do not consult a statistician before collecting the first datum for a study, as the study’s hypotheses, design, and even the data already collected cannot be changed mid-stream.
“One could argue that it is as wasteful and inappropriate to conduct a study with inadequate power as it is to obtain a diagnostic test of insufficient sensitivity to rule out a disease.” {Eng J. Sample size estimation: How many individuals should be studied? Radiology 2003;207(2):309-13}
A sample size should be large enough to detect a difference (if there truly is a difference), but not so large that small clinically unimportant differences are detected. Appropriate sample size even has application for negative results,
“If a trial with negative results has a sufficient sample size to detect a clinically important effect, then the negative results are interpretable – the treatment did not have an effect at least as large as the effect considered to be clinically relevant. If a trial with negative results has insufficient power, a clinically important but statistically nonsignificant effect is usually ignored or, worse, is taken to mean that the treatment under study made no difference. Thus, there are important scientific reasons to report sample size and/or power calculations.” {Moher D, Dulberg CS, Wells GA. Statistical power, sample size, and their reporting in randomized controlled trials. JAMA 1994;272(2):122-4.}
And even has an ethical base,
“A study with an overlarge sample may be deemed unethical through the unnecessary involvement of extra subjects and the correspondingly increased costs. On the other hand, a study with a sample that is too small will be unable to detect clinically important effects. Such a study may thus be scientifically useless, and hence unethical in its use of subjects and other resources.” {Altman DG. Statistics and ethics in medical research. Br Med J 1980;281(6251):1336-8.}
Next month, statistics in small doses will touch upon those factors that can influence sample size – might their appropriate use help your sample size?