Issues of reliability, validity and sampling

HideShow resource information
Experimental research (reliability)
Reliability refers to ability to repeat study and obtain same result (replication). Essential all conditions remain the same.
1 of 10
Observational techniques (reliability)
Observations should be consistent. Inter-observer reliability: calculated by total agreements to total observations. Result of > 0.80 good i-o reliability. Can be improved via training to use coding system/ behaviour checklist
2 of 10
Self-report techniques (reliability)
Internal reliability: measure extent to which something is consistent within, e.g. Qs on IQs should all measure same thing. External validity: measure of consistency over several occasions.
3 of 10
Inter-interviewer reliability / inter-rater reliability (reliability)
Whether two interviewers produce same outcome
4 of 10
Split-half method (reliability)
Compare person's performance on two halves of questionnaire/test. Should be close correlation in scores derived from each half. - internal reliability
5 of 10
Test-retest method (reliability)
Questionnaire/test/interview repeated after reasonable interval. Outcome should be the same
6 of 10
Validity
Internal validity - researcher tested what they intended to test / External internal - extent results can be generalised
7 of 10
Experimental research (validity)
Some lab experiments have high external validity, where lab setting not relevant to behaviour observed. Important to consider: if task was artificial (rather than setting) and thus low in mundane realism - reduces generalisability
8 of 10
Observational techniques (validity)
Internal validity - observations will not be valid/reliable if coding system/ behaviour checklist is flawed. Internal validity - also affected by observer bias
9 of 10
Self-report techniques (validity)
Face validity - does the test look like it is measuring what the researcher intended. Concurrent validity - comparing performance on a new test with a previous established test on same topic; if participants' performance show high correlation.
10 of 10

Other cards in this set

Card 2

Front

Observations should be consistent. Inter-observer reliability: calculated by total agreements to total observations. Result of > 0.80 good i-o reliability. Can be improved via training to use coding system/ behaviour checklist

Back

Observational techniques (reliability)

Card 3

Front

Internal reliability: measure extent to which something is consistent within, e.g. Qs on IQs should all measure same thing. External validity: measure of consistency over several occasions.

Back

Preview of the back of card 3

Card 4

Front

Whether two interviewers produce same outcome

Back

Preview of the back of card 4

Card 5

Front

Compare person's performance on two halves of questionnaire/test. Should be close correlation in scores derived from each half. - internal reliability

Back

Preview of the back of card 5
View more cards

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »