RELIABILITY= how CONSISTENT and DEPENDABLE a test and its results are
Internal reliability= how consistently a method measures WITHIN ITSELF (different parts of test should give consistent results) Assessing: 'split half method'-splits test into 2 halves to see if each gives similar results External reliability= how CONSISTENTLY a method measures OVER TIME WHEN REPEATED (extent to which the results from a test can be generalised) Assessing: 'test-retest method'- repeating test with same ps to see if gives same results Inter-rater reliability= test gives consistent results regardless of who does the test Assessing: 'correlating scores'- comparing at least 2 researchers data to see if results are similar. Can be insured through training and checklists. |
VALIDITY= whether a test measures what it claims to
Internal validity= extent to which the results of the test are caused by the variable being measured (DV), rather than extraneous variables (whether the results are really due to variables the researcher tested) External validity= test should produce consistent results regardless of when it is used (whether the results can be generalised; if conducted in different environments or using ps) Ecological validity= extent to which the results reflect real life (representative of naturally occurring behaviour) Factors affecting Validity:
|
Whose your date? Will it be Mr Reliable or Mr Valid... Here is some info on each of them:
Evaluation of Reliability Evaluation of Validity:
Strengths:
|
Weaknesses:
|
Strengths:
|
Weaknesses:
|