Psicometría: Modelo de medición y estructura factorial

Introduction to Measurement Models in Psychometrics

Understanding Latent Variables

• The discussion begins with the concept of latent variables or psychological constructs, which are intangible and cannot be directly accessed.

• Measurement instruments are utilized to derive scores that represent observed behaviors, allowing for inferences about these latent variables.

Classical Test Theory

• The classical test theory establishes a relationship between observed variables and true scores, emphasizing that true scores cannot be directly observed.

• A basic linear relationship is proposed where the observed score equals the true score plus an error term.

Item Selection and Representation

• It is crucial to define the construct clearly to select appropriate indicators (items), ensuring they accurately reflect the latent variable.

• Items should correlate well with each other statistically, indicating a strong representation of the underlying factor.

Measurement Model Components

• The measurement model consists of a true variable explaining item responses, acknowledging that this explanation includes measurement error.

• Mathematically, item scores depend on both the factor and associated errors, forming the basis for statistical analysis in psychometrics.

Validity of Content

• When constructing items or validating content, critical inspection is necessary to ensure items genuinely reflect the intended latent variable.