Factor Analysis
Structural Equation Modeling
Item Response Theory
100
What is a "factor" in factor analysis?
An unobserved or latent variable representing a construct.
100
What is the main reason to use SEM methods?
Issues with NHST alone
100
What is the equation for classical test theory?
Xi = Ti + ei
200
In this picture, what do H1, H2, and H3 represent?
Individual observed test items loading onto the "happiness" factor.
200
What type of model is displayed here?
Correlated Causes Model
200
What are the two broad predictors of observed responses, according to item response theory?
Individual Differences
and
Characteristics of Test Items
300
What is being described by "...how strongly an observed variable is predicted by an unobserved (latent) factor..." ?
Factor Loading
300
What type of model is displayed here?
Indirect Effects Model
300
If classical test theory is deterministic, then item response theory is __________.
...probabilistic
400
What does it mean if two factors are said to be orthogonal?
They are uncorrelated
400
What are the three requirements for establishing causality?
1. Establish association
2. Establish temporal precedence
3. Rule out alternative explanations
400
Assume test item "A" has a item difficulty of 1.
Person 1 has an ability of -3 and Person 2 has an ability of 2 on the construct for which test item A is measuring.
For which person (1 or 2) would item response theory be able to generate a more accurate prediction, regarding their accuracy on test item A?
Person 1
500
What are the five steps in the factor analysis process to publication?
500
In an indirect effects model, what does c' represent?
The direct effect
(X predicting outcome controlling for the indirect effect)
500
What is the item response theory assumption of local independence?
There is no statistical relation (e.g., correlation) between people's responses to pairs of items on a test once the primary trait being measured is held constant.