Classical Test Theory

Question 1

CTT

Answer 1

Measurement theory that defines conceptual basis of reliability

Question 2

Reliability in CTT

Answer 2

The extent to which differences in respondents’ observed scores are consistent with the differences in the true scores

Derived from observed scores, true scores and measurement error

Question 3

Observed scores

Answer 3

Values obtained from measureming a characteristic, eg scores on a test

Question 4

True scores

Answer 4

Real amounts of that characteristic in that sample. Difficult for psychological construct, as your measurement must measure a construct

Question 5

Measurement error

Answer 5

Random noise, extend to which other attributes than the one being measured effect the observed scores

Question 6

Signal

Answer 6

Strong signal implies high reliability, it’s the amount of explained var

Question 7

Two key assumptions

Answer 7

1) Observed scores are a function of true scores plus measurement error
2) Measurement error is random

Question 8

Properties of error

Answer 8

Mean is always 0
Correlation between error and true scores is always 0 (measures something else)