Skip to contents

Classification Quality Metrics

Source: R/diagnostics.R
classification_quality.Rd

Computes detailed classification quality metrics including average posterior probabilities, entropy, and odds of correct classification.

Usage

classification_quality(object)

Arguments

object

A SurveyMixr object

Value

List containing classification quality metrics

Details

Classification quality metrics assess how well individuals are assigned to latent classes:

  • Entropy: Overall classification certainty (0-1, higher better)

  • AvePP: Average posterior probability of assigned class

  • OCC: Odds of correct classification (AvePP / (1 - AvePP))

  • mcaAvePP: Minimum classification accuracy based on AvePP

Guidelines:

  • Entropy > 0.80: Excellent separation

  • Entropy 0.60-0.80: Good separation

  • Entropy < 0.60: Poor separation, consider fewer classes

  • AvePP > 0.90 for each class indicates reliable classification

  • OCC > 5 is desirable

References

Clark, S. L., & Muthén, B. (2009). Relating latent class analysis results to variables not included in the analysis. Unpublished manuscript.

Examples

if (FALSE) { # \dontrun{
fit <- gmm_survey(data = mcs_simulated, id = "id", time = "age",
                  outcome = "selfcontrol", n_classes = 3)

classification_quality(fit)
} # }