After conducting research we must test whether the results we have found are ‘significant’. To do this we use several different tests, dependant on the type of data and experimental design. The four tests are Chi², Spearman rank coefficient, Mann-Whitney ‘U’ test or the Wilcoxon ‘T’ test.
How to select which test to use
There are 3 questions you need to ask:
1) Is my data nominal (is the data in categories)? If so use Chi². If not continue to 2.
2) Am I looking for a relationship or difference in the data? If relationship use Spearman rho, if difference continue to 3.
3) Am I using independent groups or repeated measures? If independent groups use Mann-Whitney ‘U’ test, if repeated measures use Wilcoxon ‘T’ test
How do we know if our results are significant?
Using various methods each test will give you a certain number. You then trace across on the table of critical values, which will change depending on the test, and if your results are equal to or lower than the corresponding number you know you have achieved a level of significance.
Symbols for each test
Chi² = x²
Spearman Rho = Rho
Mann-Whitney ‘U’ = ‘U’
Wilcoxon ‘T’ test = ‘T’