Psychology - Research Methods 1
Target population/Sample
The group of people the researcher is interested in and from which the sample is drawn.
Generalisability
The extent to which results can be generalised beyond the sample to a wider population or situation.
Representativeness
Generalisations only make sense if the sample is a representative of the target population.
Opportunity sampling
A sample of participants produced by selecting people who are most easily available at the time of the study.
Self-selected sampling
A sampling method made up of volunteers.
+ A way to find willing participants, that are less likely to drop out.
- Sample Is biased because volunteers are highly motivated and have extra time on their hands than the population in general.
Random sampling
Sample produced using random technique.
+ Unbiased - all members of TP have an equal chance of being selected.
- Takes more time/effort than other methods because you need to obtain a list of all members of your TP, then identify the sample and contact those people to ask if they will take part.
Snowball sampling
Recruitment of participants based on word of mouth or referrals from other participants.
+ Enables researcher to locate groups of people who are difficult to access.
- Sample not likely to be a good cross-section of population because its made up of freinds of friends.
Self-report
Any data collection method that involves asking people to report their thoughts, feelings, or behaviour. Could be done via questionnaires or interviews.
Closed question
Questions that require only brief answers like yes or no.
+ Easy to analyse because data is in numbers and can be easily summarised using averages.
- May not allow people to express their precise feelings, so data collected could be low validity.
- Tends not to uncover new insights.
Open question
Allows the person to respond in-depth.
+ Provides rich details of how people behave as they express what they actually think.
- More difficult to detect patterns and draw conclusions because there are likely to be a wide range of answers.
Rating scale
Respondents give an assessment of their views using a scale like 1-5.
+ Reasonably objective way to represent attitudes related to the topic being researched.
+ produces quantitative data which is easy to analyse or represent in graphs.
- Respondents may avoid ends of scales and go for the 'middle road' so answers are not true and will lack validity.
Structured interview
Predetermined questions delivered by interviewer who does not go beyond the answers received.
+ Can be easily repeated
+ Easy to analyse than unstructed interview because answers are more predictable.
- Interviewers expectations may influence the answers the person gives.
Unstructured interview
No questions decided in advance.
+ Detailed info can be obtained from each respondent than in a structured interview.
+ Can access info that may not be revealed by predetermined questions.
- More affected by interviewer bias than structured interviews because the interviewer is developing questions on the spot.
Semi-structured interview
A mix of pre-set questions and unprepared questions asked to a respondent.
+ detailed info can be obtained from each respondent than in an structured interview because questions are specially shaped for participant.
- Affected by interviewer bias than structured interviews because the interviewer is developing questions on the spot.
Questionares
A set of open or closed questions.
+ easily repeated so data can be collected from large numbers of people cheaply and quickly.
+ respondents might be more willing to reveal personal info in a questionnaire than in a interview.
Reliability of a self-report
A self-report method is considered reliable if it can be repeated and the same or similar results are found.
External reliability
Extent to which a measure varies from one use to another.
i.e is consistant overtime.
Inter-rater reliability
The consistency between two different interviewers
can be checked by comparing results from 2 or more interviewers who questioned the same person and the answers given should be the same if the interview is reable.
Internal reliability
Measure of whether is something is consistent within itself.
Split-half method
Test items are split into 2 halves and the scores on both halves are compared.
Scores should be similar if the test is reliable.