Inferential statistics

Why do we have to complete statistical testing?

• Researchers use statistical tests to determine the likelihood that the effect/difference/relationship they have found has occured due to chance

• So in other words, we complete statistical testing to make sure any difference between conditions or relationship between variables are truly significant as opposed to a fluke (due to chance factors)

How do you choose a statistical test?

1) Do I need a test of difference (for an experiment) or a test of relationship (for a correlation or association)?
2) If a test of difference is required what is the experimental design? (Independent, repeated or matched pairs)

3) What is the level of measurement/data (nominal, ordinal or interval/ratio

IMPORTANT

Significance tables

• When you have completed a test, you will have an observed value

• Once you have completed a test we get an observed (or calculated) U value, which is then compared against a critical (table) U value, which is provided in a critical value table. This tells us whether or not results are significant

• You need an observed value, the no of pts and whether there is a directional or non-directional hypothesis

Degrees of freedom (df)

Relates to the number of pts and experimental conditions

• R = no of rows

• C = No of columns