Reliability Flashcards
What is meant by ‘internal reliability’
The extent to which a test or measure is consistent within itself e.g. the use of standardised instructions and procedures for all participants.
What is meant by ‘external reliability’
External reliability: the extent to which a test produces consistent results over several occasions.
What are issues of relability?
Lack of operationalisation of the variables is a potential issue. If you do not clearly state how you are defining and measuring your variables, this makes replication
impossible and makes inter-rater reliability issues more likely (see below for details).
Order effects can occur in a repeated measures design. Whilst completing the second
condition the participant can become practiced/better at the task the second time round, or they can underperform due to factors such as boredom. A lack of standardised procedures, inconsistency of measuring tools and an uncontrolled environment all jeopardise reliability
What are the ways of dealing with issues of reliability (three ways)
Increasing levels of control in a study raises reliability – such as using a standardised procedure and scientific measuring equipment to produce objective, and unbiased data (e.g. PET and MRI scans over self-reported ‘feelings’)
Another way of dealing with issues of reliability is by conducting the research in a laboratory helps to overcome most reliability issues as there is a greater the level of control over the study leading to a lower the risk of extraneous variables affecting results.
Yet another way of dealing with issues of reliability is using counterbalancing. This is done to remove the reliability issue of Order effects.
This involves splitting the population/sample into two halves; the first half complete condition one followed by condition two, the second half complete condition two followed by condition one.
This negates the impact of order/practice effects, asboth conditions are affected equally, and increases internal reliability.
What are the three ways of assessing reliability
Split half method - The split-half method is a technique used to assess the internal consistency of a measurement scale or test. It involves splitting the items or questions of the scale into two halves and comparing the scores obtained from each half.
Inter-rater - Inter-rater reliability refers to the consistency or agreement between two or more raters or observers when assessing or scoring the same set of data or stimuli.
Test retest – Participants are tested. And then tested again some time later to see if the results are similar.
