reliability

• internal reliability


  the actual design of the research

  • does it consistently measure something

  to check for internal reliability use the split half method:

  • split results in half, compare results from one half with results from the other half. (should say the same thing, present you with the same findings)
  • spearmans test - statistical test for internal reliability

  double check the internal reliability, use a statistical test to see if there is a positive correlation between the 2 sets of results
• external reliability


  • do the results remain consistent over time

  check this by using the 'test-retest method'

  • do the test, the retest Ps in future and compare results. (should have a positive correlation between them) (+0.8)
  • spearmans test - test external reliability
• intra-researcher reliability


  • intra means within

  refers to whether a researcher is being consistent within themselves

  • if researchers font remain consistent it may affect the results they get

  easier to ensure when using standardised procedure and standardised instructions
• inter-researcher reliability


  refers to whether different researchers are being consistent between each other. if they aren't the data collected will most likely not be accurate

  • in order to check, get data from one observer and compare it with the data collected by another

  ways to improve inter-researcher reliability:

  • training
  • operationalised variables
  • standardised instructions
  • standardised procedure