Reliability

Question 1

Reliability

Answer 1

• Consistency and stability of a measure across time, settings and individuals
• Valid tests must be reliable, but reliable tests are not necessarily valid

Question 2

True score theory

Answer 2

• Reflects a person’s genuine ability, characteristic or potential.
• Observable score = true score + error (individual differences and measurement error)

Question 3

lasting and temporary characteristics

Answer 3

• Lasting characteristics are permanent traits that apply to all dimensions
• Temporary characteristics are general and may affect tests e.g. fatigue and mood etc

Question 4

Measurement error

Answer 4

• Random error: temporary condition e.g. hungry, tired, nervous etc. is not felt by all participants creating variability in the data.
• Systematic error: Consistent e.g. fire alarm during exam alters everyone’s data

Question 5

Test-retest reliability

Answer 5

• The same test is given to the same people on multiple occasions. The correlation between these scores is then computed to show how susceptible the test is to changes in conditions e.g. level of motivation.
• longer time = more room for natural change to occur.
• Shorter time = carry over/practice effects.

Question 6

Parallel forms of reliability

Answer 6

-Develop multiple versions of the test then correlate the results to check reliability.

Question 7

Split half reliability

Answer 7

-Administer the test to a group of participants, split the test in half and compute the correlation between the two halves.

Question 8

Reliability of change measures

Answer 8

-Measures the reliability of scores that change over time e.g. depression Two people that are equally depressed should have similar scores.