Reliability (AO1) Flashcards
Reliability
is a measure of consistency
e.g. measuring a box with a ruler, you should get the same measurement each time
2 Types of Reliability
internal
external
Internal Reliability
the extent to which a measure is consistent within itself
e.g. score on first half of an OCD test should be the same on the second half
External Reliability
the extent to which a measure is consistent from one use to another
e.g. your score on an OCD test should be similar if you do it again a week later
2 Ways to Assess Reliability
test-retest reliability
inter-observer reliability
Test-Retest Reliability
administering the same test or questionnaire to the same people on different occasions
- if the test is reliable, the results should be the same or very similar + 2 sets of scores should significantly correlate
- must be enough time between tests to ppt doesn’t remember answers
Inter-Observer Reliability
ensuring 2+ observers are observing behaviour in a consistent way
- the data collected by the different observers should be the same and significantly correlate
- often involves ensuring observers are using behavioural categories correctly
4 Things to Improve Reliability of
questionnaires
interviews
experiments
observations
Questionnaires
use the test-retest method - results should correlate highly
if they don’t, identify the questions that are unreliable + remove, replace or reword them so they are clearer
Interviews
use the same interview each time or ensure each interviewer receives similar training, so questions are clear
use structured interviews where questions are pre-written
Experiments
keeping conditions tightly controlled increases reliability - bc of standardised procedures
control extraneous variables for consistency
Observations
rely on interpretations - can lack objectivity
need to operationalise any behavioural categories accurately to avoid subjective decisions
