The T-test

Statistics for Psychologists – Analysis of Variance (ANOVA)

In this course, Dr Andrew Bell (King’s College London) explores the one-way analysis of variance, or one-way ANOVA. In the first lecture, we think about t-tests in their three primary forms of one-sample, two-samples and paired samples. In the second lecture, we think about how to compare more than two groups of data and the inherent multiple comparisons problem. In the third lecture, we think about the analysis of variance (ANOVA) as a way of comparing more than two groups of data while avoiding the multiple comparisons problem. Next, we look at how to perform a one-way ANOVA. In the fifth and final lecture, we think about some key terminology and points of understanding when using an ANOVA. Photo by Chris Liverani on Unsplash

The T-test

In this lecture, we think about t-tests, focusing in particular on: (i) the three primary types of t-test being the one sample (comparing a dataset to a known distribution), the two-samples (comparing two unrelated datasets) and the paired samples (comparing two related datasets) t-test [for more information, please see ‘Statistics for Psychologists – Data and the T-test’]; (ii) using comedy and drama films as an example of comparing two unrelated datasets; (iii) the two-samples t-test being the appropriate comparative tool to use in the case of comparing these two independent samples; (iv) an example of a two-samples t-test using values from the film scores example; (v) using a p-value table to establish meaningful results from a two-samples t-test; (vi) understanding the relationship between a t-statistic and a p-value, in relation to the probability of a significant difference.