Furr, & Bacharach (2014) – Psychometrics An introduction
Chapter 1 – Psychometrics and the Importance of Psychological
Measurement
Observable behaviour and unobservable psychological attributes
- Validity: making a reasonable inference (from observable behaviour to unobservable
psychological attribute).
- Measurement usually linked to theory.
- Hypothetical constructs/latent variables: theoretical psychological
characteristics/attributes/processes/states that cannot be directly observed.
Procedures used to measure these are operational definitions.
Psychological tests: definition and types:
- What is a psychological test: ‘…systematic procedure for comparing behaviour of ≥ 2
people’.
o Tests involve behavioural samples of some kind (e.g., MC test)
o Behavioural samples must be collected in systematic way.
o Purpose of tests is to compare behaviours of ≥ 2 people.
Interindividual differences: comparing different people.
Intra-individual differences: behaviour of same individuals at different
points in time/under different circumstances.
- Types of tests:
o Variation in content, with regard to response required + according to methods.
o Distinction based on intended purpose:
Criterion/domain referenced: decision about person’s skill level based
on predetermined criterion.
Norm referenced: based on comparison to others.
Overlapping/blurred distinction: criterion based on expected level (= norm)
+ norm referenced as cutoff scores (e.g., SAT need certain score).
o Speeded tests vs. power tests:
Speeded: scored by counting number of questions answered in time
period but easy questions.
Power: not time limited but counting correct answers, questions
ranging in difficulty.
o Test/measure/instrument/scale/inventory/assessment/battery (intended to be
used together but not necessarily designed to measure single psychological
attribute).
- What is psychometrics: science concerned with evaluating attributes of psychological
tests. Attributes usually conceptualised as hypothetical constructs.
o Type of information generated by psychological tests.
o Reliability of data.
o Validity of data.
- Galton: anthropometrics (= measurement of human features) + psychometry (=
measurement of mental features). Also differential psychology (= studying individual
differences).
1
Chapter 1 – Psychometrics and the Importance of Psychological
Measurement
Observable behaviour and unobservable psychological attributes
- Validity: making a reasonable inference (from observable behaviour to unobservable
psychological attribute).
- Measurement usually linked to theory.
- Hypothetical constructs/latent variables: theoretical psychological
characteristics/attributes/processes/states that cannot be directly observed.
Procedures used to measure these are operational definitions.
Psychological tests: definition and types:
- What is a psychological test: ‘…systematic procedure for comparing behaviour of ≥ 2
people’.
o Tests involve behavioural samples of some kind (e.g., MC test)
o Behavioural samples must be collected in systematic way.
o Purpose of tests is to compare behaviours of ≥ 2 people.
Interindividual differences: comparing different people.
Intra-individual differences: behaviour of same individuals at different
points in time/under different circumstances.
- Types of tests:
o Variation in content, with regard to response required + according to methods.
o Distinction based on intended purpose:
Criterion/domain referenced: decision about person’s skill level based
on predetermined criterion.
Norm referenced: based on comparison to others.
Overlapping/blurred distinction: criterion based on expected level (= norm)
+ norm referenced as cutoff scores (e.g., SAT need certain score).
o Speeded tests vs. power tests:
Speeded: scored by counting number of questions answered in time
period but easy questions.
Power: not time limited but counting correct answers, questions
ranging in difficulty.
o Test/measure/instrument/scale/inventory/assessment/battery (intended to be
used together but not necessarily designed to measure single psychological
attribute).
- What is psychometrics: science concerned with evaluating attributes of psychological
tests. Attributes usually conceptualised as hypothetical constructs.
o Type of information generated by psychological tests.
o Reliability of data.
o Validity of data.
- Galton: anthropometrics (= measurement of human features) + psychometry (=
measurement of mental features). Also differential psychology (= studying individual
differences).
1
