To evaluate the quantity and quality of the use of statistics in Austrian medical journals, all “original research” papers in No. 116/1-12 of Wiener Klinische Wochenschrift (WKW) and 153/1-24, 154/1-24 of Wiener Medizinische Wochenschrift(WMW) were screened for their statistical content. Types, frequencies and complexity of statistical methods applied were<br />systematically recorded. A 46-item checklist was used to evaluate statistical quality for a subgroup of papers. 74.3% of WKW papers contained inferential methods beyond descriptive statistics. Only 43.7% of WMW papers employed methods of inferential statistics. There was a statistical significant difference regarding the use of statistical methods between the two journals (p = 0:009). In addition, complexity and sophistication of statistical analysis was considerable higher for WKW papers (p = 0:02). Statistical errors<br />and deficiencies were identified in a large proportion of papers. Although inferential statistics were frequently identified in papers from WKW, only a minority of WMW research had analytical character. Types and frequencies of statistical errors identified, did not vary meaningful from findings of similar studies for a wide range of medical journals. There is reason to assume, that the journal impact-factor does not seem to be a powerful predictor for the statistical quality of published research.
In a classic significance test, based on a random sample with size , a value will be calculated at size aiming to reject the null hypothesis. The sample size , however, can retrospectively be divided into partial samples and a test of significance can be calculated for each partial sample. As a result, several partial samples will provide significant values whereas others will not show significant values. In this paper, we propose a significance test that takes into account the additional information from the values of the partial samples of a random sample. We show that the values can greatly modify the results of a classic significance test.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.