Psychology: Research Methods

    Cards (55)

    • Null hypothesis
      There was no correlation between
    • Directional hypothesis

      State whether it is positive or negative correlation
    • Non-directional hypothesis

      You don't state whether it is positive or negative e.g "there is a correlation"
    • Difference hypothesis (Null)
      There is no difference
    • Difference hypothesis (Directional, one-tailed)

      State whether there is an increase or decrease in the difference
    • Difference hypothesis (Non-directional, two-tailed)

      There is a difference between the two groups but don't state what direction
    • Test of difference
      DV (what is being measured) and IV (what is being changed)
    • Test of correlation

      Co-variables (usually 2)
    • If you had Interval data and ordinal data it becomes ordinal overall
    • Test of difference graphs

      • Bar charts (same width and spaces between)
      • Histograms (different widths and no spaces)
    • Test of correlation graphs
      • Scatter Graphs
    • Types of data

      • Nominal (Categories)
      • Ordinal (Scores)
      • Interval (Specific time/ units)
    • Inferential data

      Statistical tests
    • Spearmans
      Can be for both ordinal and interval
    • Pearsons
      Can only be for interval data
    • Parametric tests

      Data needs to be normally distributed and needs to be interval data
    • Non-parametric tests

      Everything else
    • Types of data

      • Qualitative (non numerical data - opinions/words/sentences)
      • Quantitative (numerical data)
    • Significant result
      Researcher can reject the null hypothesis
    • Critical value
      The cut off point between acceptance and rejection of the null hypothesis
    • Probability
      Measure of likelihood that the event will occur
    • Sign test

      Critical value, N=number of participants, Calculated value (S)= lowest number of differences (+/-)
    • Researchers use statistical tests to see whether the effects/difference/relationships they have found have occurred due to chance
    • Type 1 error
      Incorrect rejection of a true null hypothesis (a false positive)
    • Type 2 error
      The failure to reject a false null hypothesis (a false negative)
    • We use a 5% significance level as it balances the chance of making a type 1 or type 2 error and we use 0.05 because the probability that the results are due to chance in 5 in every 100 or 1 in 20
    • Features of science
      • Objectivity - personal bias minimised so evidence not distort/no influence on anything
      • Falsifiability - something can not be scientific unless it admits the possibility of being proven untrue (false)
      • Empirical method - scientific approaches that are based on the gathering of evidence through direct observation and experience
      • Theory construction - developing an explanation for the causes of behaviour by gathering evidence and constructing a theory
      • Paradigm shift - result of scientific resolution when there is a significant change in the dominant unifying theory within a scientific discipline
    • Paradigm
      A set of shared assumptions and agreed methods within a scientific discipline
    • Replicability
      The extent to which scientific procedures and findings can be repeated by other researchers
    • Types of validity

      • Internal/External - whether something measures what it is supposed to measure
      • Ecological - generalised to the outer world
      • Temporal - whether findings are true over time
    • Ways of testing validity

      • Concurrent - do results match with previous established tests
      • Face - does the results show what it is supposed to measure
    • Ways of improving validity

      • Experimental research - double blind, single blind, use of a control group
      • Questionnaire - anonymity, eliminate social desirability bias
      • Observations - use overt observations, well defined behavioural categories
      • Qualitative research - depth and detail, use of triangulation
    • Overt observation

      Participants know they are being observed
    • Covert observation

      Participants are not aware that they are being observed
    • Triangulation
      Combining the use of multiple methods or data sources in qualitative research (theoretical frameworks) to develop a comprehensive understanding of phenomena - instead of using one theoretical framework in your research
    • Ways of assessing reliability

      • Test-retest - same test is administered to the same person at different times
      • Inter-observer - using a pilot study to see whether behavioural categories have been consistently applied
    • Measuring reliability

      Two sets of scores should correlate +0.80 is great
    • Ways of improving reliability

      • Questionnaire
      • Interviews
      • Observations - detailed categories remove use of jargon
      • Experiments
    • Content analysis

      A form of observation where communication is studied indirectly
    • Case studies
      Unusual events that can't be repeated e.g London riots, czech twins
    See similar decks