Term
|
Definition
Refers to the extent to which a psychological measure measure what it's supposed to
(Accuracy) |
|
|
Term
|
Definition
whether the effects observed in a study is due to the manipulation of the IV and not another factor
-did it measure what it was intended too? |
|
|
Term
| What compromises internal validity? |
|
Definition
Confounding variables
-eg demand characteristics, researcher bias |
|
|
Term
| What improves internal validity? |
|
Definition
-controlling variables
-using standardised instructions
-counter balancing |
|
|
Term
| What is counter balancing? |
|
Definition
One group does test A then test B, the other does test B and then test A
-(repeated measures) |
|
|
Term
| What are extraneous variables? |
|
Definition
| Variables other than the IV that could have an effect on the DV and have been controlled |
|
|
Term
|
Definition
Ecological - results of study are generalisable to other settings/environments
Population - results are generalisable to other people
Temporal (historical) - results are generalisable to other times/eras |
|
|
Term
| What could be done to improve external validity? |
|
Definition
| -use a more natural setting eg field experiment or covert observation |
|
|
Term
| Define construct validity |
|
Definition
-The study allows researchers to measure constructs as they occur in real life
Qualitative data - high construct validity
Quantitative data - low construct validity |
|
|
Term
|
Definition
A way of assessing validity
-the test appears to be measuring what is it supposed to be measuring
-eg for a pilot study |
|
|
Term
|
Definition
A way of assessing validity
-ask research colleagues whether they think the test is measuring what is is supposed to be measuring |
|
|
Term
| Define predictive validity |
|
Definition
| The degree to which a test accurately predicts criteria that will occur in the future |
|
|
Term
| Define concurrent validity |
|
Definition
-test a group of people with the new test and test them again with an established one
-the results should be the same |
|
|
Term
| Case studies have ___ validity because: |
|
Definition
High
-Rich, qualitative data
Low
-ungeneralisable |
|
|
Term
| Observations have ___ validity because: |
|
Definition
Overt- low
-demand characteristics
-social desirability effect
Covert - high
-no demand characteristics |
|
|
Term
| Lab experiments have ____ validity because: |
|
Definition
Low
-demand characteristics
-researcher effects
-false environment |
|
|
Term
|
Definition
| -the consistency to which findings can be repeated |
|
|
Term
| Assessing reliability - Inter rater |
|
Definition
External validity
-more than one observer records behaviour in an observation
-consistent estimates of number of times behaviour is shown means good reliabilty |
|
|
Term
| Assessing reliability - test-retest |
|
Definition
External reliability
-test given to person twice on two separate occations
-same/similar results obtained means high reliability
Critical value for correlation - 0.7 and above |
|
|
Term
| Assessing reliability - split half method |
|
Definition
Internal validity
-assesses questionnaires
-measures the extent to which all parts of the test contribute equally to what is being measured |
|
|
Term
| What is the method of the split half method? |
|
Definition
| Questionnaire is split in half and each half is given to a participant. If each half has the same results the questionnaire is reliable |
|
|
Term
| What is the strength and limitation of the split half method? |
|
Definition
-quick and easy
-only works for large questionnaires |
|
|
