Lecture 1: Introduction
Psychological measurement
Constructs
● Psychological construct- unobservable!
○ This is a theoretical concept of psychological differences between individuals
○ Examples: working memory, perceptual organization, extraversion, neuroticism,
etc.
● Psychological test- observable
○ This is a measurement instrument to quantify individual differences in the
construct
○ Operational definitions / Operationalizations of psychological constructs
How psychological tests differ
Test differ in their:
● Content
○ Depends on the construct to be measured
● Response that is required
○ Open-ended vs close-ended
● The method of administration
○ Individual vs group
● The use
○ Criterion referenced: determines if someone passes a certain cut-off
■ E.g., exam
○ Norm referenced: compare the score to the population
■ E.g., intelligence test: you get an IQ score which tells you what the person
is scoring in comparison to the population
● Timing
○ Speed test (time limit): relatively easy questions & you see how far a person
comes
VS
○ Power tests (no time limit): the questions have different difficulty levels & you
see what one can solve
■ You can say something about someone's skill
● The meaning of indicator
○ Formative vs reflective
,Formative vs reflective measurement
Most psychological tests involve reflective measurement!
● The construct is assumed to cause the differences in the test score
● The item responses are indicators of the construct
● Items are necessarily correlated
○ Higher values on the construct increase scores on all items
● E.g., differences in arithmetic skill cause differences in 30+20,
18+6, 24-7
Formative measurement
● The item responses (indicators) define the construct
● Items are not necessarily correlated
○ Items dependently contribute to the construct
● E.g., the value of a car (construct) is determined by its age,
condition, size (items)
Challenges in psychological measurement
● Complexity: Psychological phenomena are complex, high-dimensional concepts
○ You can't catch people by a number
○ You need to carefully isolate a psychological dimension (construct) and measure
it
● Reactivity: people react differently as they know that they are being measured
● Observer bias: expectation of the researcher may affect the test
● Composite scores: multiple items scores need to be combined into one final score
○ Can you just sum it up or weigh more difficult items more heavily?
● Sensitivity: its unknown beforehand how sensitive the scale should be
○ E.g., “How do you like the lecture until now?”
■ Too few response categories: you may miss out on individual differences
■ Too many response categories: people can't meaningfully distinguish the
different categories
● Awareness: many test administrators do not know about the psychometric quality of the
test
,Dimensionality analysis
The dimensionality of a test is an important issue
● Dimensionality: the number of constructs a test measures
Unidimensional test: the test measures a single construct
● E.g., D2: test of attention, primary school arithmetic test
Multidimensional test: the test measures multiple constructs
● Big-5 personality inventory
○ This is a 5 dimensional test
○ OCEAN
● WAIS-IV intelligence test
○ Verbal comprehension, perceptual reasoning, working memory, processing speed
Use this distinction
→ This never happens in psychology
→ One construct is correlated to the others
→ General construct is general intelligence
, How do you establish this dimensionality?
Factor analysis: a statistical method to study the dimensionality of a test - how many
dimensions are in the data
Exploratory factor analysis (EFA):
There is no theory about the factor structure (this course)
Confirmatory factor analysis (CFA):
There is a clear theory about the factor structure (not covered in this course)
The aim is: we have a psychological test, we don't know the factor structure and we want to find
out
A first EFA
Step 1: Correlation matrix
● EFA is commonly conducted on the correlation matrix
● The correlation matrix summarises all possible correlations between the constructs of
your tests
○ For example: the grades for all school subjects
This might indicate that there is a dimension in the data for ‘exact’ and for ‘language’ skill
Step 2: Check the Eigenvalues
● Tells you something about how much variability this factor can account for in the data
- The first factor always has the highest eigenvalue
“Which factors are high enough to speak of an important dimension”
Step 3: Select a number of factors
Psychological measurement
Constructs
● Psychological construct- unobservable!
○ This is a theoretical concept of psychological differences between individuals
○ Examples: working memory, perceptual organization, extraversion, neuroticism,
etc.
● Psychological test- observable
○ This is a measurement instrument to quantify individual differences in the
construct
○ Operational definitions / Operationalizations of psychological constructs
How psychological tests differ
Test differ in their:
● Content
○ Depends on the construct to be measured
● Response that is required
○ Open-ended vs close-ended
● The method of administration
○ Individual vs group
● The use
○ Criterion referenced: determines if someone passes a certain cut-off
■ E.g., exam
○ Norm referenced: compare the score to the population
■ E.g., intelligence test: you get an IQ score which tells you what the person
is scoring in comparison to the population
● Timing
○ Speed test (time limit): relatively easy questions & you see how far a person
comes
VS
○ Power tests (no time limit): the questions have different difficulty levels & you
see what one can solve
■ You can say something about someone's skill
● The meaning of indicator
○ Formative vs reflective
,Formative vs reflective measurement
Most psychological tests involve reflective measurement!
● The construct is assumed to cause the differences in the test score
● The item responses are indicators of the construct
● Items are necessarily correlated
○ Higher values on the construct increase scores on all items
● E.g., differences in arithmetic skill cause differences in 30+20,
18+6, 24-7
Formative measurement
● The item responses (indicators) define the construct
● Items are not necessarily correlated
○ Items dependently contribute to the construct
● E.g., the value of a car (construct) is determined by its age,
condition, size (items)
Challenges in psychological measurement
● Complexity: Psychological phenomena are complex, high-dimensional concepts
○ You can't catch people by a number
○ You need to carefully isolate a psychological dimension (construct) and measure
it
● Reactivity: people react differently as they know that they are being measured
● Observer bias: expectation of the researcher may affect the test
● Composite scores: multiple items scores need to be combined into one final score
○ Can you just sum it up or weigh more difficult items more heavily?
● Sensitivity: its unknown beforehand how sensitive the scale should be
○ E.g., “How do you like the lecture until now?”
■ Too few response categories: you may miss out on individual differences
■ Too many response categories: people can't meaningfully distinguish the
different categories
● Awareness: many test administrators do not know about the psychometric quality of the
test
,Dimensionality analysis
The dimensionality of a test is an important issue
● Dimensionality: the number of constructs a test measures
Unidimensional test: the test measures a single construct
● E.g., D2: test of attention, primary school arithmetic test
Multidimensional test: the test measures multiple constructs
● Big-5 personality inventory
○ This is a 5 dimensional test
○ OCEAN
● WAIS-IV intelligence test
○ Verbal comprehension, perceptual reasoning, working memory, processing speed
Use this distinction
→ This never happens in psychology
→ One construct is correlated to the others
→ General construct is general intelligence
, How do you establish this dimensionality?
Factor analysis: a statistical method to study the dimensionality of a test - how many
dimensions are in the data
Exploratory factor analysis (EFA):
There is no theory about the factor structure (this course)
Confirmatory factor analysis (CFA):
There is a clear theory about the factor structure (not covered in this course)
The aim is: we have a psychological test, we don't know the factor structure and we want to find
out
A first EFA
Step 1: Correlation matrix
● EFA is commonly conducted on the correlation matrix
● The correlation matrix summarises all possible correlations between the constructs of
your tests
○ For example: the grades for all school subjects
This might indicate that there is a dimension in the data for ‘exact’ and for ‘language’ skill
Step 2: Check the Eigenvalues
● Tells you something about how much variability this factor can account for in the data
- The first factor always has the highest eigenvalue
“Which factors are high enough to speak of an important dimension”
Step 3: Select a number of factors
