Reliability Flashcards
Define the term reliability.
Is a measure of consistency- producing the same results over and over again.
In psychological research, reliability refers to the consistency of results from a study. If the same method is used and the same results are produced, the study can be called reliable.
State the ways of accessing reliabiltiy.
Test-retest.
Inter-observer reliability.
What happens in a test-retest method?
Involves giving participants the same questionnaire, or the same tasks in an experiment, more than once. Participants are the same so results gained should be the same as well. Enough time must pass so that the participant don’t remember the first time. Once completed, the participant’s scores can be correlated and a statistical test can be done to check the degree of similarity. If there’s a significant positive correlation (a __coefficient of +.80 or more) the study is likely to be reliable.
What happens in an inter-observer reliability method?
The extent to which there is agreement between two or more observers involved in observations of a behaviour. This is measured by correlating the observations of the observers. If there is a significant positive correlation (a __coefficient of +.80 or more) the study is likely to be reliable.
What is a relevant issue of observational research?
As one researcher’s interpretation of events may differ widely from someone else’s - introducing subjectivity bias and unreliability in the data collection process.
How can you improve the reliability of observations?
Behavioural categories need to be fully operationalised (measurable/observable), should not overlap with each other- for example, ‘hitting’ and ‘striking’ would be too similar- and all possible behaviours should be covered. This makes the categories as objective as possible, and less open to individual interpretation.
How can you improve the reliability of questionnaires?
Using the test-retest method previously described. If reliability is low, the questionnaire should be modified by amending or removing questions (for example if they are unclear), before using the test-retest method again to see if it has improved.
How can you improve the reliability of interview?
Using the same interviewer each time will enhance reliability. Alternatively, the questions shouldn’t be leading or ambiguous. Structured interviews are much more likely to be reliable than unstructured interviews.
