## Why are so many research findings false?

Douglas Altman wrote in the BMJ editorial “The Scandal of Poor Medical Research” in 1994 that “We need less research, better research, and research done… Read More »Why are so many research findings false?

Douglas Altman wrote in the BMJ editorial “The Scandal of Poor Medical Research” in 1994 that “We need less research, better research, and research done… Read More »Why are so many research findings false?

Many methods sections include the sentence “Normally distributed variables were compared with the independent groups t-test”. The sentence is problematic for a number of reasons.… Read More »A typical statistics section

Medical research publications are no longer restricted to case reports and presentation of hypotheses. Until the second half of the 20th century, expert opinions were… Read More »Medical research is based on statistical inference

Both p-values and confidence intervals are uncertainty measures, but while the confidence interval describes estimation uncertainty, the p-value reflects the probability of the data given… Read More »Why confidence intervals are more useful than p-values

Is the sample size of 3 really based on a sample size calculation with acceptable risks of false positive and false negative outcomes? Or are… Read More »Lab analyses with a sample size of n=3

Multiplicity issues exist when multiple null hypotheses are tested. Testing more than one null hypothesis increases the risk of getting at least one false-positive test… Read More »Multiplicity and Bonferroni corrections

First, a null hypothesis may or may not include assumptions about a parameter, and a non-parametric null hypothesis can often be tested using a distribution-free… Read More »Non-parametric data?

P-values are often misunderstood, incorrectly believed to be descriptive measures. A finding in a sample is often interpreted as practically important when p < 0.05,… Read More »Why p-values are controversial

Why? Some statistical methods, such as Student’s t-test, are based on an underlying assumption of a Gaussian distribution, but why should all continuous variables in… Read More »Always start a statistical analysis by checking that all continuous variables have Normal distribution.

The odds ratio is in some cases (e.g. in case-control studies) a relevant measure in itself, but in other cases (e.g. cohort studies) it is… Read More »Why odds ratios are controversial