Hi,

Repeated measures are taken over the same individuals or the same variables over different conditions or various time points. Since the units are the same as the baseline they act as their own controls. The purpose is to see whether there is a change in the measure which is significant statistically. Also, fewer subjects are required when the sample size is calculated.

So, the measuring tools will be the same for intervention 1,2 and 3 for single group.

Hi,

Yes, and standardized tools need to be used. for ex. If the Blood Pressure apparatus is used then the instrumental fluctuations have to be minimized to a large extent, in order to observe any actual changes over time or conditions.

Practical implications regarding the analysis: Definitely as my colleague said, you have to use the same measurements and the device should be reliable. If you want to know whether the 3 or 4 repeated measurements differ from each other or not, you have to test for the normality of data first. If the data are normally distributed, in this case you have to use 'Repeated measures ANOVA'. If the data are not normally distributed, you have to use Friedman repeated measures (two way ANOVA) but in fact it is not a real two way ANOVA. There is a post-hoc test for each that shows the differences between each two readings. If you have two groups of people, you have to calculate the difference, eg between reading 1 and 3 and then compare the difference between the two groups using the Unpaired t test or Mann-Whitney test depending on normality of data.