## Talking about statistics

**What does a statistical test do?**

Having collected numerical data from a number of participants it will be useful to find out whether or not the data shows changes in activity. A statistical test is used to establish whether these changes are attributable to chance or are statistically significant. The Wilcoxon Matched Pairs Signed Ranks Test is used.

**Why Wilcoxon?**

The Wilcoxon Matched Pairs Signed Ranks Test assumes that the data are at an ordinal-metric level, i.e., that the original data can be validly ordered, that the data after the intervention can be ordered, and that the difference between the two sets of data can be validly ordered. This assumption is slightly less critical than the interval level assumption necessary for a t-test. The assumption of there being a normal distribution does not have to be met, and the Wilcoxon test is very powerful. If all the assumptions are met the Wilcoxon has about 95% of the power of the much more common statistical test that is called the ‘t-test’.

**What significance level is given?**

It is customary to use the 5% significance level as an adequate indication of correlation, and this is what has been done here. If the results are even better than this, then the spreadsheet will report the relevant significance level.

**How many Inclusion Webs can be analysed?**

The Wilcoxon test is effective with data from a minimum of 10 people, although some results can be drawn from an even smaller group. If there are data from between 30 and 50 people, then the spreadsheet provides a different statistical test. Groups above 50 are too large for the spreadsheet and you will need to either split the data into smaller bundles or seek independent statistical advice.