Statistical method selection tool

Step 1. What are you interested in?

 If the mean of a variable is different from a certain fixed value A confidence interval for the mean If the mean of a variable in one group is different from the mean in another group A confidence interval for the difference between two means If the mean of a variable is different between several groups (three or more different groups) If the mean of a variable is different between several groups (two or more different groups), controlling for other variables If the proportion of 'successes' (e.g., proportion of survivors or proportion of people buying a product) is different from a certain fixed value A confidence interval for the proportion of successes If the proportion of successes in one group is different from the proportion in another group A confidence interval for the difference between two proportions If the proportion of successes is different between several groups (three or more different groups) If the scores on a variable differ between two groups (more general question than several questions above: we will narrow it down in subsequent steps) If the scores on a variable differ between three or more groups (more general question than several questions above: we will narrow it down in subsequent steps) If there is a relationship between two variables If one variable can predict another variable If one variable can predict another variable, controlling for other variables If several variables can predict another variable If several variables can predict another variable, controlling for other variables