Reliability

Question 1

Reliability

Answer 1

The consistency of a measure. The measure is considered reliable if we get the same result repeatedly

Question 2

When assessing reliability of a study you have to ask…

Answer 2

Can the study be replicated

If so, will the results be consistent?

Question 3

Measuring reliability (test-retest)

Answer 3

Measure is administered to the same group of ppl twice - if results on two tests are similar, we can assume the test is reliable

Question 4

Measuring reliability (inter-rater)

Answer 4

If the measure depends upon interpretation of behaviour, we can compare results from two or more raters - if there is high agreement between raters then the measure is reliable

Question 5

Measuring reliability (split half)

Answer 5

Splitting a test into two halves and comparing the scores in both halves - if the results on two halves are similar then we can assume the test is reliable

Question 7

Types of reliability (internal)

Answer 7

Measure of the extent to which something is consistent within Itself e.g all IQ tests should be measuring something

Question 8

Types of reliability (external)

Answer 8

Measure of consistency over several diff occasions e.g if an interviewer conducted the same interview with the same interviewee a week later, the outcome should be similar

Question 9

Self reports

Answer 9

In case of

Low internal rel, q’s can be removed to see if the split half test returns a high rel score
Low external rel, poorly written q’s may cause confusion and need to be rewritten

If interview has low rel then interviewer may need to be retrained in order to be more consistent

Question 10

Observations

Answer 10

Should be consistent (two or more observers should be watching same behav to produce same data)
Inter eater/observer reliability is the extent to which observers agree - it is measure using a correlation. If more than 80% agreement then the data is said to be reliable