# Comparing means

## Learning outcomes

On watching this video, students should be able to:

- Identify settings that involve independent samples and paired samples and know that each requires a different statistical method.
- Distinguish between the t-distribution and normal distribution and explain when and why a t-distribution should be used.
- Describe the effect the t-distribution will has on a confidence interval and a p-value from a hypothesis test compared to the normal distribution and appreciate why this is sensible for smaller sample sizes.
- Confirm that when comparing groups we should choose a population parameter that quantifies this comparison and provides a measure of effect, such as the mean difference. Hence, estimating population parameters such as a mean in each group does not allow a direct comparison or provide a measure of effect.
- State the basic form of a confidence interval and test statistic, and appreciate that these can be applied to many different settings.
- Interpret the results of these analyses in the context of the original research question.
- Confirm that these methods rely on assumptions about the data that must be checked and that other methods should be investigated when these assumptions are not met.