Reliability

Question 1

Reliability

Answer 1

measure of consistency.

Question 2

Types of Reliability

Answer 2

External - extent that measuring instruments is consistent across different times and situations.
Internal - extent that a measuring instrument is consistent within itself.

Question 3

Most Reliable -> Least Reliable

Answer 3

Experiment, Correlation, Content Analysis, Observation, Self-Report, Case Study.

Question 4

Ways Assessing Reliability

Answer 4

Test-Retest Method - use same measure on two different occasions to same people to see if get same results. If reliable results should be the same. Check using statistical correlation test if correlation coefficient 0.8 or more then reliable.
Inter-Rater/Observer/Interviewer Reliability - Compare findings between researchers using same measure and if high level of agreement then reliable. Check using statistical correlation test if correlation coefficient 0.8 or more then reliable.

Question 5

Improving Reliability of Questionnaires

Answer 5

Closed ended questions = less room for misinterpretation.

Question 6

Improving Reliability of Interviews

Answer 6

Same interviewer or train the same = don’t ask leading or ambiguous questions.
Structured Interview = more controlled by fixed questions.

Question 7

Improving Reliability of Observations

Answer 7

Operationalised behavioural categories = don’t overlap and all behaviours covered so observers don’t make own judgements leading to inconsistencies.

Question 8

Improving Reliability of Experiments

Answer 8

Lab experiments = introduce lots of control making it easier to be replicated.