statistical testing

• statistical testing provides a way of determining whether hypotheses should be rejected or accepted & can tell us whether differences or relationships between variables that have been found during experiments are statistically significant or if they have only occurred due to chance
• a sign test can be used for a study that:

  • looked for a difference, not an association
  • used a repeated measures design
  • collected nominal data
• how to conduct a sign test:

  • state hypothesis including the alternative and null hypothesis
  • record data and work out the sign (N) - the sign will be negative if the value has decreased in the second condition but positive if it has increased
  • find the calculated value for the sign test which is the number of times the less frequent sign occurs (S.)
  • find the critical value of S - use N value & p<0.05
  • if S< critical value reject null hypothesis
  • if s> critical value accept null hypothesis
  • state conclusion
• inferential statistics


  ways of analysing data using statistical tests that allow the researcher to make conclusions about whether a hypothesis was supported by the results
• how to choose an inferential statistics test:

  • what was the design of the study - related or unrelated?
  • what level of data was collected - ordinal, nominal or interval?
  • was a difference or correlation measured?
• related design


  repeated measures or matched pairs design
• unrelated design


  independent groups design
• statistical tests are used to determine whether a significant difference or correlation exists by using the calculated and critical value
• critical value


  the value a test statistic must reach in order for the hypothesis to be accepted
• the critical value is worked out from a table of probability values and depends on: whether it was a one or two tailed test, the P value, the N value/ degrees of freedom value
• if there is an R in the name of the statistical test the calculated value has to be greater or equal to the critical value for the result to be significant
• if there is no R in the tests name then the calculated value has to be less than or equal to the critical value for it to be significant