Statistical testing: The sign test

The concept of significance

- Just because we found a difference in the mean number of words spoken (example) in the two conditions, we do not yet know if this can be referred to as a significant difference. The difference may be a coincidence or a fluke

The sign test

- To determine whether the difference we found is significant, we can use this. There a number of statistical tests, each of which have their own conditions of use

The sign test: Conditions of use

- Looking for a difference rather than an association

Need to have used a repeated measures design

Need data that is organised into categories, known as nominal data

The concept of probability

- The likelihood that certain events will occur. What we want to know is 'how likely are these findings if the null hypothesis is true?' The null hypothesis states there is no difference/association in the population. If we find a difference/association in our sample, do we accept the alternative hypothesis?

Accepted level of probability (level of significance)

- This 0.05 (5% when a percentage). This is the level at which a researcher decides that the findings are significant (meaningful) and will reject the null hypothesis. In some cases, researchers need to be confident that findings were not due to chance, so they use 0.01 (1%). Used when research may involve a human cost or one-off

The critical value

- When the statistical test has been calculated the researcher is left with a number- the calculated value, which needs to be compared with this to decide whether the result is significant or not. These are found in a table

Using the tables of critical values

- We need the following information to use this:

The significance level (0.05 or 5%)

The number of participants in the investigation (the N value or sometimes degree of freedom, df)

Whether the hypothesis is directional (one-tailed test) or non-directional (two-tailed test)

- Calculated value has to be equal to or lower than the critical value for the result to be significant