Research methods - year 13

• Nominal Data

  data is put into named categories or frequency is counted
• Ordinal Data

  Data is put into order or rated on a scale
• Interval Data

  Data uses standardised measurements such as time or weight
• P=


  probability risk that the results may have occurred by chance
• Statistical test

  1. directional or non-directional
  2. lowest p value to use
  3. number of participants
• One tailed test


  directional
• Two tailed test


  non directional
• Stats test conclusion


  The results are significant as the calculated value of (CV) is higher than the critical value of (cv) for a one tailed/two tailed test when n=20 and p=0.05. This means the experimental hypothesis has been supported and the null hypothesis has been rejected and we can conclude that
• Sign test

  1. calculate difference between scores
  2. sign of difference
  3. calculated value = whichever sign is smallest
  4. decide whether its directional or non-directional
  5. level of significance
  6. identify critical value
  7. establish significance
• Type 1 error


  a researcher wrongly concludes that there is a significant difference between the IV and DV and wrongly concludes results are significant when they are not
• P value is too large

  too lenient = type 1 error
• Type 2 error


  researcher wrongly concludes there is not a significant difference between IV and DV and incorrectly concludes the results are not significant when they are
• P value is two small

  too strict = type 2 error
• how to assess reliability

  1. Test-retest reliability
  2. Inter-rater reliability
• Test-Retest Reliability


  the extent to which a procedure can be easily replicated using the same participants
• Inter-Rater Reliability


  the extent to which different researchers are likely to record similar findings using the same procedure
• How to improve reliability

  1. Remove extraneous variables
  2. Better operationalisation
• Reliability


  the extent to which a procedure produces consistently similar findings using the same participants when repeated
• Internal validity


  the extent to which a study successfully measures what is intended
• External validity


  extent to which findings of a study can be successfully generalised to other settings to wider populations and over time
• Face validity


  subjectively measures what is intended at face value (ask participants what they thought a test was measuring)
• Concurrent validity


  produces similar results when compared with another established method
• Ecological validity


  generalised to a variety of real life settings
• Temporal validity


  generalised over time
• Improving internal validity

  1. Remove extraneous variables
  2. Better operationalisation
• Improving external validity

  1. improve mundane realism
  2. increase sample cross-culturally
• Content analysis


  identifies patterns and trends in the material before turning these into operationalised coding system (quantitative data)
• Thematic analysis


  identifies themes in material before writing extensive and detailed notes (qualitative data)
• Content analysis

  1. read material and look for patterns and trends
  2. turn behaviour in operationalised categories
  3. read content again and tally categories
  4. make conclusions
• Inter-Rater Reliability


  using two or more researchers to perform the same content analysis at the same time using the same categories
• Thematic analysis

  1. read material and look for themes in behaviour
  2. write extensive and detailed qualitative notes
  3. read notes and reduce them to recurring themes
  4. make conclusions
• Features of a science

  1. Objectivity
  2. Empirical methods
  3. Replicability
  4. Falsifiability
  5. Theory construction
  6. Hypothesis testing
  7. Paradigm
  8. Paradigm shift
• Objectivity


  collecting data without bias or opinion
• Empirical methods


  recognised and objective methods used to generate observable evidence
• Replicability


  being able to repeat the procedure of a study to see if similar results occur every time
• Falsifiability


  can be tested to be correct or false
• Theory Construction


  ideas about the world become tested to explain behaviour
• Hypothesis testing


  statement including a clear IV and DV can be tested
• Paradigm


  shared set of assumptions and methods used to research and understand the world around us
• Paradigm shift


  the established paradigm has been challenged by a new piece of evidence which the existing paradigm cannot explain