The fragility of the EmPress
Fragility index. Fragility index describes the minimum number of participants whose status has to change for p to go from being statistically significant to not being so.
Science without sense...double nonsense
Evidence Based Medicine pills
Science without sense...double nonsense
Evidence Based Medicine pills
Category Statistics
All roads lead to Rome
Comparison of proportions. We review the methods for comparison of proportions: chi-square, Fisher’s exact test and approximation to a normal. This expression has its origin in the crazy habit that came to Romans for making roads connecting the capital of the…
Brown or blond, all bald
Comparisson of variables. We describe the statistical tests that should be used for comparisson of variables of different kinds.
Cheating Gauss
Transforming data. Sometimes our data set will not follow a normal distribution. We can try transforming data to change this situation.
When the zeroes of p are important
Bonferroni’s correction. By making multiple comparisons, the risk of Type I error increases proportionally. This can be avoided with the Bonferroni’s correction.