reliability

Question 1

reliability - define

Answer 1

refers to how consistent a measuring device is and this includes psychological tests or observations which assess behaviour

Question 2

test - retest reliability - define

Answer 2

a method is assessing the reliability of a questionnaire or psychological test by assessing the same person on two separate occasions

Question 3

inter - observer reliability - define

Answer 3

the extent to which there is agreement between two or more observers involved in observations of a behaviour

Question 4

state 2 ways of assessing reliability

Answer 4

test retest
inter - observer reliability

Question 5

what does test retest involve ?

Answer 5

involves administering the same person on different occasions

Question 6

if the test or questionnaire is reliable in test retest than how should the results obtained be ?

Answer 6

the same or at least very similar or at least very similar each time they are administered

Question 7

why are the two sets of scored be correlated in test retest ?

Answer 7

to make sure they are similar
- if the correlation turns out to be significant the reliability is good

Question 8

what should the observer do instead of going alone ?

Answer 8

conduct their observations in terms of at least two

Question 9

How is reliability measured?

Answer 9

correlational analysis

Question 10

What are the four ways to improve reliability?

Answer 10

questionnaires
Interviews
experiments
observations

Question 11

improving reliability - questionnaires

Answer 11

Low retest reliability - require some of the items to be de-selected or rewritten

for example, if some questions are complex or ambiguous, they may be interpreted differently by the same person different occasions

solution: replace with open questions with closed, fixed choice alternatives - less ambiguous

Question 12

Improving reliability - interviews

Answer 12

use the same interviewer each time if not ensure all the interviewers are properly trained and not asking questions which are too leading or ambiguous - avoided in unstructured interviews

• unstructured interviews are more freeflowing are less likely to be reliable

Question 13

improving reliability - experiments

Answer 13

Procedures must be the same each time - concerned about standardised procedures

Question 14

improving reliability - observations

Answer 14

Behavioural categories have to be operationalised
no - overlap
yes - covered on checklist

If not, observers have to make their own judgement of what to record and end up with deferring and inconsistent records