Pairwise likelihood estimation and limited‐information goodness‐of‐fit test statistics for binary factor analysis models under complex survey sampling
2025

Estimating Factor Models for Binary Data

publication Evidence: moderate

Author Information

Author(s): Haziq Jamil, Irini Moustaki, Chris Skinner

Primary Institution: Universiti Brunei Darussalam

Hypothesis

Can pairwise likelihood estimation and limited-information goodness-of-fit test statistics be effectively applied to binary factor analysis models under complex survey sampling?

Conclusion

The study successfully extends pairwise likelihood estimation and introduces limited-information test statistics for binary factor analysis models, demonstrating good performance under complex sampling designs.

Supporting Evidence

  • The study evaluates the performance of the proposed estimation methods using simulated data.
  • Limited-information test statistics showed good performance in terms of Type I error and power.
  • The introduction of sampling weights improved the accuracy of parameter estimates.

Takeaway

This study helps researchers analyze data from surveys where answers are yes or no, making it easier to understand people's opinions and behaviors.

Methodology

The study uses pairwise likelihood estimation and introduces limited-information goodness-of-fit test statistics for binary data.

Potential Biases

Potential bias may arise from ignoring sampling weights in the estimation process.

Limitations

The performance of the Wald test statistic can be unstable, especially with complex sampling designs and small sample sizes.

Digital Object Identifier (DOI)

10.1111/bmsp.12358

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication