Reliability

Question 1

Define ‘reliability’

Answer 1

• Refers to how consistent a measuring device is
• & this includes psychological tests or observations which assess behaviour

Question 2

What is test-retest reliability?

Answer 2

• A method of assessing the reliability of a questionairre or psychological test by assessing the same person on two separate occasions
• This shows to what extent the test (or other measure) produces same answers
  i.e is consistent or reliable

e.g. Sandra Bems androgony scale had good test-retest reliability

Question 3

What is inter-observer reliability?

Answer 3

• The extent to which there is an agreement between two or more observers involved in observations of a behaviour
• This is measured by correlating the observations of two or more observers

Question 4

What is the reliability figure?

Answer 4

+.8

Question 5

How can you improve reliability?

Answer 5

• Questionaires
• Interviews
• Observations
• Experiments

Question 6

How can you improve reliability for a questionaire?

Answer 6

• Reliability over time should be tested using test-retest measure
• A questionairre that produces low test/retest reliability may require some of the items to be rewritten
• A solution might be to place open questions with closed fixed choice alternatives which may be less ambiguous

Question 7

How can you ensure reliability for an interview?

Answer 7

• To use the same interviewer each time
• If not possible interviewers must be trained
• Not asking leading questions is avoided in in structured interviews where interviewer’s behaviour is more controlled by the fixed questions
• Interviews that are unstructured & more flowing are less likely to be reliable

Question 8

How can you improve reliability of observations?

Answer 8

• Making sure behavioural categories have been properly operationalised & that they are measureable & self evident
  (e.gthe category ‘pushing’ is much less open to interpretation than ‘aggression’)
• Categories shld not over lap (e.g. hugging/cuddling) if so observers may have to make their own jugements of what to record where & may end up w differing & inconsistent records
• If reliability is low observers may need further training

Question 9

How can you improve reliability of experiments?

Answer 9

• Procedures must be the same (consistent) everytime
• Therefore in terms of reliability an experimenter is concerned about standardised procedures