statistical test · BEGINNER LESSON
t-test
A t-test is a parametric statistical test used to compare the mean values of two groups. It uses each group’s mean and standard deviation to judge whether the groups differ more than would be expected from random variation.
What it is and what it measures
A t-test is a statistical method for comparing two groups by looking at their means, which are measures of central tendency. It also uses the standard deviation, a measure of how spread out the values are in each group. The goal is to judge whether the groups differ in their central values.
[1] recommendations for reporting statistical methods - PMC - NIH
How it works
The test compares the group means while taking account of how much variation each group has. Because it is a parametric test, it depends on the data meeting a normality assumption, meaning the values should follow an approximately normal distribution. If the data are not normally distributed, the t-test is considered inappropriate.
[1] recommendations for reporting statistical methods - PMC - NIH
When it should and should not be used
The source states that the t-test is used for two-group comparisons. If the data do not follow a normal distribution, other tests are preferred: the Mann–Whitney U test for two independent groups and the Wilcoxon signed-rank test for two dependent, or paired, groups. For three or more non-normal groups, the Kruskal–Wallis test or Friedman test may be used depending on whether the groups are independent or paired.
[1] recommendations for reporting statistical methods - PMC - NIH
Limitations and reporting
The t-test does not replace checking the distribution of the data. Normality should be assessed before choosing the test, and when data are not normal, the mean and standard deviation may not describe the data well. In that case, the median and quartiles are more appropriate summaries. For comparisons involving more than two groups, homogeneity of variance should also be checked after normality.
[1] recommendations for reporting statistical methods - PMC - NIH
SEE THE IDEA
How a t-test is used
Collect data from the two sets being compared.
Summarize the central value and spread for each group.
Confirm the data are approximately normally distributed.
Use the t-test only if the normality assumption is suitable.
The test compares group means, but only after checking whether the data fit the required distributional assumption.
A t-test is chosen after checking normality; if normality is not met, different tests are recommended.
TRY IT YOURSELF
Test your understanding
Choose an answer, then check the explanation.
PUT IT TOGETHER
Match each concept to its meaning
Drag a label onto its description, or select a label and then select a description. This also works with a keyboard.
4 pairs to match
Sources and further reading
This general lesson was generated from public source material. It explains a concept; it does not interpret a private paper or provide medical advice.