reliability and validity Flashcards
(18 cards)
What is reliability?
A measure of consistency
When is a measurement seen as reliable?
When it can be repeated
What is the test retest method
A straightforward method of testing reliability. The same test should be given out to the same pp’s more than once. There should be a time gap big enough for pp’s to forget questions + answers but not too big that they forget their opinions. Most common types are questionnaires, interviews or psychological assessments. Results should be correlated if they are reliable
How would you improve reliability of a questionnaire?
Use the test retest method. Some items may need to be removed or rewritten. Replace open questions with closed/fixed questions that are less ambiguous
How would you improve reliability of an interview?
Use the same interviewer or used trained interviewers. Use structured interviews where interviewer uses fixed questions and their behaviour is controlled. use questions that are easy to understand and not open to interpretation
How would you improve reliability of experiments?
Procedures must be consistent and the same every time. Use standardised procedures so results can be compared
How would you improve reliability of observations?
Make sure behavioural categories have been operationalised and can be measured. Categories shouldn’t overlap. All behaviours should be covered on the checklist. Observers may need to be trained
What is validity?
Validity relies on whether psychological tests, observations, investigations etc are legitimate. It tells the researcher if the experiment is genuine and is representative of the real world. It refers to whether the researcher measures what they set out to measure. The extent to which an observed effect is genuine.
What are the two types of validity?
Internal and external
What is internal validity?
Refers to whether the manipulation of the IV is the only thing affecting the DV. has the researcher measured what they set out to measure?
What is external validity?
Relates more to factors outside of the investigation like generalisation to other settings.
What is ecological validity?
A type of external validity which concerns generalising the findings to other settings
What is temporal validity?
The issue of whether findings from a study hold true over time. The extent to which findings can be applied to other historical eras
what is inter-rater reliability?
the extent to which there is an agreement between multiple observers involved in observations of behaviour. This is measured by correlating the observations of 2 or more researchers. A general rule is if total number of agreements divided by total observations is > 0.8, data has high reliability
what is face validity?
a basic form of validity in which a measure is scrutinised to determine whether it appears to measure what it is supposed to measure
how to improve the validity of an experiment
-include a control group so researcher can be certain it is the IV affecting the DV.
-standardise procedures to minimise investigator effects etc
how to improve the validity of a questionnaire
-incorporate a lie scale to assess consistency and control social desirability bias
-assure respondents that data will remain anonymous
how to improve the validity of an observation
-observer remains undetected (covert) so behaviour is more natural
-have distinct behavioural categories that aren’t open to interpretation
