Designing Psychological Investigations Flashcards
Random sampling
Everyone in the target population has an equal chance of being selected
High population validity
Even if it’s random it may not be truly representative of the population
Practical limitations
Sample may not be random as ps may choose not to take part
Opportunity sampling
Selected by using people who are most readily available
High chance of biased because sample is chosen from small part of the target population
Leading to low population validity
Volunteer sampling
A sample where the participants self select - they volunteer to take part in the research
A particular type of person is likely to volunteer for research
High bias so cannot generalise to target population leading to low population validity
Reliability
This refers to the consistency of a particular measurement.
Researchers should be able to measure something time after time and get similar results
Replicability
The ability to repeat a study
External reliability
This is the ability to produce the same results every time
Assessing external reliability
Can be checked using the test - retest method. This is where the test or questionnaire is used twice on the same participants. To get a measure of reliability correlate the two scores, the higher the correlation the more reliable the test/questionnaire
Inter-rater reliability
Concerns whether two interviewers or observers can produce the same outcome and are consistent.
Assessing inter-rater reliability
Can be assessed by correlating the scores of the two observers to give an indication of reliability. The higher the correlation the the more reliable the observations
Internal reliability
This is concerned with the level of consistency within a test or observation
Assessing internal reliability
The split half method assesses the extent to which all parts of the test or questionnaire contribute equally to what is being measured. Score on half the items are correlated with scores on the other half. The higher the correlation the more reliable the yet or questionnaire. The reliability of a test could be improved through this method
Improving reliability
Taking more than one measurement from a participant and then taking an average score.
A pilot study could be done.
Essential that all of the conditions are the same otherwise any change in results could be due to a change in conditions.
Standardised procedures to allow precise replication.
Standardised behaviour checklist or coding system where the categories of behaviour are clearly defined
Validity
Validity refers to whether the test or measure is measuring what is claims to measure.
Internal validity
The ability of the study to test the hypothesis that is was designed to test, essentially does the dv measure what it was intended to measure
Potential problem for internal validity - experimental designs and controls
If a repeated measures design is used there is danger of order effects such as practice effects. If an independent groups design is used then participants variables are introduced
Controls > order effects can be overcome by counterbalancing the conditions. Participant variables can be overcome by using match pairs design