Describe inter-rater reliability and test-retest reliability, using any examples.

Mark scheme

Marking points from the official mark scheme (Oct/Nov 2019). The mark or mark type is shown in brackets.

1. [6] Award 1 mark for each example that is linked to reliability, up to a maximum of 3 marks for each. Maximum 4 marks for definition/detail of inter-rater reliability. Maximum 4 marks for definition/detail of test-retest reliability. Examples can include examples from studies using such measures of reliability, or of ways it could be used. Max 4 marks in total for all one side, e.g. all about inter-rater reliability. For example: Inter-rater reliability: ∞ is a measure of whether different researchers are scoring consistently (1 for definition) ∞ indicates whether two scorers are rating responses in the same way (1st mark for detail) ∞ if they are (reliable), ratings for the same data by two (different) researchers should (positively) correlate (2nd mark for detail) ∞ for example if data from an interview was interpreted in the same way (1 for example) ∞ for example Bandura et al. checked that the scoring of aggressive ratings was similar when done by different researchers (1 for example) Test-retest reliability: ∞ is a way to find out whether a measure is consistent over time/measures whether a test is consistent (1 for definition) ∞ it is done by using the same test on the same participants twice (in close succession) (1st mark for detail) ∞ two results for each participant should be (roughly) the same (2nd mark for detail) ∞ so there should be a (strong positive) correlation between the two sets of scores (3rd mark for detail) ∞ for example if an IQ test was done on the same people twice, each person should get the same score (1 for example) ∞ for example if Baron-Cohen et al. used the AQ on the same group twice, they should get the same AQ each time (2nd mark for detail) Other appropriate responses should also be credited.