Use: Comparing the mean difference of three or more measurements of the same continuous variable. You should have measurements in different columns with each row representing one subject.
Dependent (outcome) variable: Continuous
Independent (predictor) variable: 3+ categories e.g. weight measured at three time points or three types of biscuit
Example: testing for a change in weight on a diet at three or more time points or comparing taste scores for 3+ different biscuits
Summary statistics/graphs: Summarise the means at each time point and a mean bar or line chart
Use: Testing for a difference between 3+ measurements of the same variable when the variable of interest is ordinal or the dependent variable is very skewed.
Dependent (Outcome): Ordinal or skewed repeated measures
Independent (predictor): 3+ time points or conditions
Example: Each person tries three different biscuits and gives a taste rating from disgusting to delicious or weight on a diet is very skewed
The Friedman test can also be used when you have asked people to rank a list of options and wish to see if there are general preferences
Summary statistics/graphs: Use medians, Interquartile range and boxplot for ordinal or skewed data.
Note: If you have taken the mean or sum of several ordered questions (scale mean), use parametric tests such as t-tests. Some disciplines use parametric tests for individual ordinal questions but a wider range of responses (ideally 7+) is needed
We don't currently have resources on this topic in Jamovi so book an appointment to see a stats tutor
The resources guide you through the r code and interpretation of the relevant summary statistics and test for comparing two different groups of subjects. The script files with all the code can be adapted to run each technique on your own data.
We don't currently have resources for this topic in SAS but we do have paired t-test and it's non-parametric equivalent the Wilcoxon signed rank for paired measurements
These resources show the calculations for the specified techniques. We don't currently have resources for this topic but we do have paired t-test resource for comparing two time points
Need help choosing the right test?