RELIABILITY AND VALIDITY

Question 1

Define reliability

Answer 1

• reliability is a measure of consistency.

Question 2

Test-retest reliability

Answer 2

• test the same person twice
• If test Is reliable the results should be the same or very similar

Question 3

inter observer relability

Answer 3

• compares observations from diff oservers.
• E.g two or more oservers compare data by conducting a pilot study.

Question 4

How Is reliability measured

Answer 4

• the corrlation co-efficent should exceed +.80 for reliability.

Question 5

How can you improve reliabity questionares

Answer 5

• re write questions
• Replace open with closed questions.

Question 6

How to improve reliability of interviews

Answer 6

• improved training
• use the same interviewer.

Question 7

How to improve reliablity of observations

Answer 7

• Operationalisation of behavioural categories
• Categoties shouldnt overlap

Question 8

how to improve reliability of experiment

Answer 8

• the procedure must be consisent standardised procedure.

Question 9

Define validity

Answer 9

• is the result legitimate

Question 10

How can data be reliable but not valid

Answer 10

• people are tested but not measured what it intends on measuring
• e.g an IQ test produces the same results every time.

Question 11

internal validity

Answer 11

• control within study reduces demand characteristics

Question 12

External validity

Answer 12

Generalised to other settIngs, areas etc

Question 13

Ecological validity

Answer 13

• refers to the extent to which the findings can be generalized to a real-life setting.

Question 14

Temporal validity

Answer 14

• findings should be consistent over time

Question 15

2 ways of accessing validity

Answer 15

• concurrent validity
• Face validity

Question 16

Face validity

Answer 16

• way of assessing whether or not something measures what it claims to measure.

Question 17

Concurrent validity

Answer 17

• comparing a new test with an existing test (of the same nature) to see if they produce similar results.
• If both tests produce similar results, then the new test concurrent validity.

Question 18

How can we improve validity of experiments

Answer 18

• a control group means able to identify changes of dv were due to the effect of iv

Question 19

Improve validity of questionnaires

Answer 19

• a lie scale and confidentiality

Question 20

How to improve validity of observations

Answer 20

• good categories that are well defined

Question 21

How to improve validity of qualitative research.

Answer 21

• interpretive validity shown through the coherence of narrative and direct quotes from pp

Question 22

Double bind

Answer 22

• where neither research or pp know who’s reciving treatment
• prevents bias