research methods

?
View mindmap
  • research methods
    • inferential statistics
      • nominal
        • most basic level of measurement. used when data is put into categories. only tells us how many pmts picked each option. doesn't give an individual score for each ppt
          • e.g tally chart
      • ordinal
        • used when data can be put into order 1st,2nd,3rd. it can't tell us the gap between 1st and 2nd or 4th and 5th etc. always used with questionnaires. does give an individual score for each ppt
          • e.g ppt A scored 25, ppt B 47, ppt 30 therefore A,C,B in 1st,2nd,3rd.
      • interval
        • distance between each score has meaning and has equal value. eg distance between 15-30 degrees is same as 45-60 degrees. 0 means nothing eg 0 degrees
          • eg the gap between 10 and 12 is 2 as is 4 and 6
      • ratio
        • distance between each score has meaning and equal value. 0 means nothing
          • eg if it takes 0 secs to complete a test it took no time
    • experimental design
      • independent groups
        • ppts are divided into groups. only 1 half in the experimental conditions. this is done through random allocation
      • repeated measures
        • all ppts take part in the conditions of the expt. results can be compared
      • matched pairs
        • different ppts are assigned to conditions of the expt but matched on characteristics that are important to the study
    • case studies
      • carried out on unique people or a small group of people. in cognitive people with brain damage.
      • to see how brain damage affects cognitive functioning eg perception, attention, processing info and memory
      • data is gathered through experiments to investigate how memory is affected. questionnaires to see what they can remember. interviews to look at memories from before and after and to see how they feel about any cog issues. observations to see how brain damage affects daily behaviour
      • evaluation
        • hard to  generalise. each individual is different so not representative of other people with brain damage
        • brain damage tends to affect number of ares of the brain rather than one specific area making it hard to identity exactly what areas do specifically when looking at brain damage.
        • different research methods are used within case studies so data can be cross checked with each other (triangulation)if data supports each other and the effects of brain damage are reliable
        • researchers are focusing on one person with variety of methods so lots of detailed data can be gathered increasing validity
        • ethical issues if researchers focus is testing the person not helping them
  • picking a stats test
    • is it looking for a difference?
      • yes= experimental
        • level of measurement?
          • nominal
            • ppt design?
              • independent grps
                • chi-square
          • ordinal
            • ppt design?
              • independent grps
                • mann whitney U test
              • repeated measures
                • wilcoxen test
      • no= correlational
        • level of measurement?
          • ordinal
            • Spearman's Rho

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Cognitive Psychology resources »