10 topics
Social sciences
Psychometrics
Psychometrics develops mathematical models to measure psychological attributes. Item response theory and factor analysis sit at its core; the same mathematics appears in econometrics, biostatistics, and ML.
Topics in this field
Classical Test Theory
The foundational framework decomposing an observed score into a true score and random measurement error.
Confirmatory Factor Analysis
A theory-driven factor model in which loadings and factor correlations are constrained a priori and tested against data.
Differential Item Functioning
Detection of items that favor one group over another after matching on the latent trait being measured.
Factor Analysis
Dimensionality-reduction technique decomposing observed variables into latent common factors and unique error terms.
Item Response Theory
Probabilistic models linking latent trait levels to item response probabilities via item characteristic curves.
Latent Class Analysis
A model-based clustering technique that identifies unobserved subgroups from patterns of categorical observed indicators.
Rasch Model
A one-parameter IRT model placing persons and items on a common logit scale via the log-odds of success.
Reliability and Validity
Core psychometric properties evaluating the consistency and accuracy of a measurement instrument.
Structural Equation Modeling
A framework combining a measurement model and a structural model to test hypothesized relationships among latent variables.
Test Equating
Statistical procedures that place scores from different test forms onto a common scale so they can be used interchangeably.