On completion of the test you will have what is called an observed or calculated value. In order to see whether your findings are statistically significant you need to consider several factors:
- Was your hypothesis directional or non directional? If it was directional you will choose a one tailed test if it was non-directional you should choose a two tailed test.
- How many Pps did you have?
- If conducting a Chi Squared, what are the degrees of freedom (the instructions to calculate this are provided with the rest of the equation)
- What is your chosen level of significance? In Psychology we use the 5% level, (p<0.05) which means that there is less than a 1 in 20 probability that the results were due to chance.
Comparing critical and observed values: Once we have the answers to all of these questions we can look up the critical value. The next job is to compare the observed value and the critical value.
- For a Chi Squared or a Spearman’s Rho, we can only accept the hypothesis if the observed value is greater than the critical value.
- For a Mann Whitney U test we can only accept the hypothesis if the observed value is less than the critical value.
Type 1 errors: A type 1 error occurs if an experimental/alternate hypothesis is accepted in error, i.e. the null was in fact true. This can result from choosing a significance value which is too high. A type 1 error is a false positive, e.g. saying the results were significant when they weren’t.
Type 2 errors: A type 2 error occurs if an experimental/alternate hypothesis is rejected in error, i.e. the null was in fact false. This can result from choosing a significance value which is too low. A type 2 error is a false negative, e.g. saying the results were not significant when they were.