L1: Intro to Psychological Measurement Flashcards
Psychological Construct - Definition
a theoretical & unobservable concept of psychological differences between people
Psychological Test - Definition
measurement instrument to quantify the individual differences in a psychological construct
What are ways in which psychological tests differ? (6)
1) Content
2) Response Required (open ended vs closed)
3) Method of Administration (individual vs group)
4) Use (criterion vs norm referenced)
5) Timing (speed vs power)
6) The meaning of indicators (formative vs reflective)
Criterion Referenced Test - Definition
Test designed to compare test result to predetermined criteria
Norm Referenced Test - Definition
Test designed to compare individual test results to the results of other test-takers
What is an example of a criterion referenced test?
High school exam
What is an example of a norm referenced test?
IQ test
Speed Test - Definition
time limited test where questions are relatively easy, and it is measured how many questions someone answers in a period of time
Power Test - Definition
test with no time limit where items are of different difficulty levels, and it is measured how far of a level of items someone can answer
Reflective Measurement - Definition
- the construct is assumed to be causing the differences in test scores.
- all items are correlated with each other, as higher values of the construct will necessarily also increase scores on all of the test items.
- arrows go construct to item
Formative Measurement - Definition
- the item response defines the construct
- items are not necessarily correlated, as they can all independently contribute to the construct
- arrows go item to construct
List the challenges in psychological measurement (6)
1) Complexity - how can be isolate & measure complex and high-dimensional concepts?
2) Reactivity - people respond differently when they know they are being observed
3) Observer Bias - expectations of the researcher influences the test
4) Composite Scores - does the composite score adequately reflect the construct?
5) Sensitivity - how sensitive should a test scale be?
6) Awareness - many administrators of psychological tests do not know the psychometric qualities of it
Challenge - Reactivity
people tend to respond differently when they know they are being observed
Observer Bias - Definition
the expectations of the researcher have the potential to influence the test
Challenge - Composite Scores
Does the composite score adequately reflect the construct being measured?
Challenge - Test Sensitivity
It is not possible to know beforehand how sensitive a tests scale should be. If you include too few response categories, you miss our on individual differences. If you include too many response categories, you cannot distinguish between different categories
Challenge - Test Awareness
Many test administrators do not know the psychometric properties & qualities of the test they are using
Dimensionality of a test - Definition
the number of constructs that a test measures
Unidimensional Test - Definition
a test that measures 1 construct
Multidimensional Test - Definition
a test that measures multiple constructs