Reliability

Cards (7)

  • Reliablility is the extentto which a test produces consistent results
  • What are the two types of reliablity
    External and internal
  • External validlity is the extent to which the findings of the study can be generalised to other situations and is consistent overtime.
  • Internal relibalilty is when the testis consistent in itself
  • Issues and how to face:
    only one researcher= add another researcher
    research only done once= repeat it tosee if it was replicable
    Instrunctions not given in same way= standardise procedures
    variables not clearly defined= define the variables inmore detail
    p's not asked same questions= standardise procedures
  • To measure reliablility in a lab/ field experiement you need to replicate the study
  • To measure reliability in non experiemtnal ethods(interviews, questionaires, observations):
    • test-retest reliability=measure is administrated from the same group twice= if results are similar assume its reliable
    • splithalf reliabilty=splitting into two halves andcomparing scores= if results from both halves are similar then can assume its reliable
    • inter-rater reliability= if measure depends upon interpretation of behaviour, we can compare results from two or more raters= if there is high agreement between them thenassume measure is reliable