Reliability

Question 1

What is reliability?

Answer 1

A measure of consistency. If an experiment is repeated twice, results should be similar or the same.

Question 2

Name 3 ways of assessing reliability.

Answer 2

Test retest
Inter observer reliability
Correlation coefficients

Question 3

What is the test retest method?

Answer 3

Administering the same test to the same people again after a short period of time, on different occasions.

Question 4

Describe what researchers should consider when deciding on the time period before retesting.

Answer 4

The test must be administered long enough so PPTs have not remembered their answers, but not long enough that their opinions have changed.

Question 5

What research method typically uses the test retest technique?

Answer 5

Questionnaire

Question 6

After the re test has happened, what do researchers do to compare the results to the original findings?

Answer 6

The 2 sets of scores are correlated. If the correlation is significant (+0.8) and positive, the reliability is good

Question 7

What are the issues with observations being subjective?

Answer 7

Observers may perceive things in a different way. As a result, they may only record things they see as important, that may not be seen as important by others (and vice versa)

Question 10

Which type of observation aims to establish inter rater reliability?

Answer 10

Content analysis

Question 10

What might the issue be with a questionnaire that has a correlation coefficient below +0.8?

Answer 10

Questions might be ambiguous and could have been interpreted differently by the same person on different occasions.

Question 11

How might researchers try to solve issues with ambiguity in questionnaires?

Answer 11

They might try to use more open ended questions where PPTs can respond how they like.

Question 12

How can reliability be improved in interviews?

Answer 12

Use the same interviewer each time

Question 13

How do non leading questions improve reliability in interviews?

Answer 13

Allows the PPT to respond in a natural way, therefore their response given will be most authentic and not influenced by the researcher.

Question 14

Which type of interview is likely to produce the most reliable data and why?

Answer 14

Structured. Questions asked will always be the same so answers are likely to be similar each time.

Question 15

What should researchers do if reliability of observations is low?

Answer 15

They should meet and discuss how they can apply their behavioural categories more consistently.

Question 16

How might the reliability of experiments be improved?

Answer 16

Use standardised procedures each time. This means procedures are consistent and the same each time. This also ensures that findings will definitely be a result of procedures and not individual PPT differences.