reliability + validity

Question 1

what is reliability

Answer 1

how consistent the results are

Question 2

what is internal reliability

Answer 2

results that are consistent within itself

Question 3

what is external reliability

Answer 3

consistent from one occasion to the next

Question 4

why is it important to assess the reliability of observations

Answer 4

so they can be repeated for a second time

Question 5

how do you assess the reliability of observations

Answer 5

inter-observer reliability

Question 6

explain the process of interobserver reliability

Answer 6

two observers make independent recordings

after the observation, come together to see the agreements

correlate results (>0.8)

Question 7

what are the two ways to improve the reliability of observations if the inter-observer score is low

Answer 7

operationalise behavioural categories further

provide more training for the observers

Question 8

what are the two methods used to improve reliability of self report techniques

Answer 8

test-retest and inter-interviewer

Question 9

what is the process of test-rest reliability

Answer 9

a group of people complete the test then the same test if given to the same people after a short interval of time (1-2 weeks) so answers have been forgotten

the scores across both tests for each person are then correlated, if they are the same it has high reliability

Question 10

what is the process of inter-interviewer reliability

Answer 10

you could interview the same person again after a week similar to test-retest

you could also record the interview and use two separate interviewers and see their agreements and correlate the results

Question 11

how do you improve the reliability of self report techniques

Answer 11

reduce ambiguity as some questions may be unclear and provide multiple answer types

standardisation as procedures and instructions must be the same for all participants

Question 12

what is validity

Answer 12

legitimacy, whether the results reflect reality

Question 13

what is internal validity

Answer 13

what goes on within the study that could influence results

Question 14

what are examples of things that could affect internal validity in experiments (3)

Answer 14

investigator effects: anything the investigator does which may have an affect on participants performance that wasnt intended

demand characteristics: cues that communicate the aim of the study to the participants so their true behaviour is altered

confounding variables: a variable in an experiment that changes systematically with the IV, conclusions can not be drawn about what caused changes in the DV

Question 15

what are examples of things that could affect internal validity in self report measures

Answer 15

social desirability bias: ppts may give answers that do not reflect reality as they want to give answers that are reflected in a good light

Question 16

what are examples of things that could affect internal validity in observations

Answer 16

poorly operationalised behavioural categories: observers cannot record reality because the categories are not clear

Question 17

what is population validity

Answer 17

generalising the findings of a study to other people

Question 18

what is temporal/historical validity

Answer 18

generalising the findings of a study to other time periods

Question 19

what is ecological validity

Answer 19

generalising the findings of a study to other settings

Question 20

what are the two ways of assessing validity

Answer 20

face validity and concurrent validity

Question 21

what is face validity

Answer 21

concerns the issue on whether a self report measure looks like it is measuring what the researcher intended

Question 22

what is concurrent validity

Answer 22

involves comparing the current method of measuring a topic with a previously validated method of the same topic

Question 23

if a questionnaire has poor face validity how can this be improved

Answer 23

questions should be revised so they relate more to the topic

Question 24

if a questionnaire has poor concurrent validity how can this be improved

Answer 24

researcher should remove irrelevant questions, experts can help check this

Question 25

how can you improve validity through the design of studies

Answer 25

single blind (ppts are unaware of what condition they are in) and double blind (ppts and researcher are unaware of conditions) procedures to prevent ppts guessing the aims