Reliability and Validity

Question 1

what is reliability?

Answer 1

• measure of consistency/ repeatability (correlation should exceed 80%)
• measure attitudes and emotions rather than concrete things like height

Question 2

Assessing reliability:

test re-test

Answer 2

• give same test to same person on diff occasion
• give enough time for person to forget 1st test
• test correlation/ significance

Question 3

Assessing reliability:

interobserver reliability

Answer 3

• stop researcher bias as there multiple researchers
• pilot study make sure behaviour categories consistently applied
• results then correlated

Question 4

how can questionnaires be used to test reliability?

Answer 4

• if they have low test re-test (less 80%) things need to be changed
  (like amount of open questions used)

Question 5

how can reliability be improved in interviews?

Answer 5

• use same researcher

- avoid leading and ambiguous questions

Question 6

how can reliability be improved in experiments?

Answer 6

• standardisation of processes (like lab studies)

- minimises extraneous variables

Question 7

how can reliability be improved in observations?

Answer 7

• behavioural categories operationalised
• categories shouldn’t overlap
• all possible behaviour should be covered in categories

Question 8

what is validity?

Answer 8

whether something produces legitimate result that represents behaviour in real world
- data can be wrong, unreliable or inconsistent (like broken scales)

Question 9

what is internal validity?

Answer 9

• wether effects on DV due to IV manipulation

- most threatened by demand characteristics

Question 10

what is external validity?

Answer 10

• factors outside of investigation than can effect results

- like generalising to settings, people, eras

Question 11

what is ecological validity?

Answer 11

• extent of how findings generalise from one setting to another
• can be lowered if DV isn’t “real to life”
  (mundane realism)

Question 12

what is temporal validity?

Answer 12

wether findings remain true over time

Asch done in a conformist era of America, not the same now

Question 13

Assessment of validity:

face validity

Answer 13

• Whether it measures what its supposed to on the face of it

(like looking at the instrument)

Question 14

Assessment of validity:

concurrent validity

Answer 14

• if results match with a previous test conducted

Question 15

how can validity be improved in experimental research?

Answer 15

• control groups
• standardised procedure
• single and double blind trials

Question 16

what observations have the best validity?

Answer 16

covert observation

Question 17

how can qualitative methods improve validity?

Answer 17

research with more depth/ detail increases ecological validity

(enchanted by triangulation - using different sources as evidence)

Question 18

can can validity be improved in questionnaires?

Answer 18

scales to assess consistency of repossess to control social desirability bias