Chapter 6 t-tests

t-tests are usually one of the first families of statistical tests that students learn when they take a research methods subject. I (Dan) pretty much learnt only t-tests until my third year of my psychology major (I learnt chi-squares and other tests through taking separate statistics subjects through my uni’s Department of Statistics before I learnt them in psychology). It’s not hard to see why this is the case - t-tests are really intuitive and simple to conduct, and so are an accessible way into learning statistical tests (even though chi-squares are even easier).

The family of t-tests come into play when we have one categorical IV with two levels, and one continuous DV. As you can imagine, there are many instances where this kind of design comes into play, and you will see as much in the datasets and examples this week. There are nine datasets for you to play around this week (3 for each kind of test) - so hopefully that will give you plenty of practice!

By the end of this module you should be able to:

• Describe how a t-test works in principle
• Conduct three forms of chi-square tests: one-sample, independent-samples and paired-samples
• Calculate and interpret an appropriate effect size for the above tests

Figure 6.1: xkcd: Control Group