When to use statistical tests or regression

Exposure is numerical, binary (two groups) and parametric

Statistical test like z test, t test

Exposure numerical, binary non parametric

Man Whitney u test or wilcoxen rank sum

Exposure is numerical, binary paired parametric

Paired z test if large sample paired t test if small

Exposure is numerical, binary, paired non parametric

Wilcoxen rank sum

Exposure is numerical, unordered categorical variable parametric

One way ANOVA

Exposure is numerical, unordered categorical variable non parametric

Kruksal-Wallis test

Binary outcome variables difference between proportions

A z test

Binary outcome variable 2×2 table of ORs large sample

Chi squared

Binary outcome variable 2×2 table of ORs small sample

Fishers exact test

Unordered categorical r x 2

Chi squared test

Unordered categorical paired/matched 2×2 table

Mc Nemars chi squared

Ordered categorical rx2

Chi squared test for trend

What is a t-test used for?

To compare the means of two groups to determine if they are statistically different from each other.

What does ANOVA stand for?

Analysis of Variance, used to compare the means of three or more groups.

What is the purpose of a Chi-squared test?

To determine if there is a significant association between two categorical variables.

When would you use a paired t-test?

When comparing means from the same group at two different times.

What is the Wilcoxon signed-rank test used for?

To assess whether the mean ranks of two related samples differ. Either unpaired data small sample or paired data small sample. Numerical data.

What type of data does the Kruskal-Wallis test evaluate?

It is a non-parametric method for testing whether samples originate from the same distribution. Unordered categorical data.

What is Fisher’s exact test used for?

To determine if there are nonrandom associations between two categorical variables in small sample sizes.

When is a z-test appropriate to use?

When comparing sample means from two groups with large samples or known population variance.

What is the significance of effect size in statistical tests?

Effect size quantifies the size of a difference and helps to understand the practical significance.

What does a p-value indicate in hypothesis testing?

It indicates the probability of observing the test results under the null hypothesis.