Misuse of statistical methods: critical assessment of articles in BMJ from January to March 1976.

Abstract

Sixty-two reports that appeared as Papers and Originals (excluding short reports) in 13 consecutive issues of the British Medical journal included statistical analysis. Thirty-two had statistical errors of one kind or another; in 18 fairly serious faults were discovered. The summaries of five reports made some claim that was unsupportable on re-examination of the data. Medical investigators should consult with people who have a real understanding of statistical methods throughout their projects.

O papel do comitê de ética em pesquisa na avaliação de testes estatísticos

“…Nesse último caso, por exemplo, é comum que a aplicação de um teste qui-quadrado produza na mesma tela o teste de Pearson, a correção de Yates e o teste de Fisher 2 . Cada um desses resultados se enquadra em uma situação especial 35 . Disso se depreende que a exploração de dados à procura de p significativos que se adaptem a hipóteses é um caminho que não deve ser aceito.…”

Advice on Statistical Analysis for Circulation Research

“…[1][2][3][4][5][6][7] Among the main errors identified were incorrect use of multiple comparisons of several independent groups, low statistical power, and inadequate analysis of repeatedmeasurement studies. The problem in the multiple comparisons of means from more than two populations has been well recognized since then, and now it is rare that misuse of repeated t tests is published in peer-reviewed journals.…”

Can Statistics Cause Brain Damage?

“…On a more positive note, in the fol lowing section an attempt is made to outline areas in which improvements could be made in the exe cution and presentation of data analysis. It was no surprise that most of the problems discussed below were also identified by Glantz (1980) and by Gore et al (1977) in Circulation and the British Medical Journal, respectively. In addition to studying the articles mentioned above, the interested reader would do well to read the excellent follow-up pa pers in Circulation Research (Wallenstein et al, 1980) and the British Medical Journal (Gore,198 1).…”

“…Recent surveys of the use and abuse of statistics have clearly docu mented this fact (see for example, Glantz, 1980 andGore et al, 1977). These surveys indicate that proper statistical analyses would severely weaken and often completely invalidate the conclusions in many of the articles studied.…”