Sequential Confirmatory Factor Analysis: A Novel Approach to Latent Variable Measurement
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915455687458816 |
|---|---|
| author | Johnson, Zachary Esses |
| author_facet | Johnson, Zachary Esses |
| contents | Factor score estimation in small sample sizes often encounters parameter bias and convergence failures when constructing hierarchical national/sub-national indices. This paper proposes a novel method for hierarchical factor analysis called "sequential Confirmatory Factor Analysis". Instead of estimating multiple levels of factors at the same time, this approach calculates factor scores sequentially from the lowest to highest levels. This sequential estimation keeps the original sample size in each step and also removes cross-level covariance estimation. Using a series of Monte Carlo simulations, we isolate the difference between sequential Confirmatory Factor Analysis and traditional Confirmatory Factor Analysis by comparing their resulting factor scores to the true latent variables under varying conditions. We also estimate the WJP Rule of Law Index using traditional Confirmatory Factor Analysis, Bayesian Confirmatory Factor Analysis, and sequential Confirmatory Factor Analysis to test performance. Our findings demonstrate that sequential Confirmatory Factor Analysis significantly outperforms the traditional model for indices with simple/moderate complexity. Traditional Confirmatory Factor Analysis performs better where the data are skewed. Where the hierarchical model becomes complex, the two methods perform similarly. Finally, sequential Confirmatory Factor Analysis can provide valid estimates where traditional or Bayesian Confirmatory Factor Analysis fail to converge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_15611 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Sequential Confirmatory Factor Analysis: A Novel Approach to Latent Variable Measurement Johnson, Zachary Esses Methodology Factor score estimation in small sample sizes often encounters parameter bias and convergence failures when constructing hierarchical national/sub-national indices. This paper proposes a novel method for hierarchical factor analysis called "sequential Confirmatory Factor Analysis". Instead of estimating multiple levels of factors at the same time, this approach calculates factor scores sequentially from the lowest to highest levels. This sequential estimation keeps the original sample size in each step and also removes cross-level covariance estimation. Using a series of Monte Carlo simulations, we isolate the difference between sequential Confirmatory Factor Analysis and traditional Confirmatory Factor Analysis by comparing their resulting factor scores to the true latent variables under varying conditions. We also estimate the WJP Rule of Law Index using traditional Confirmatory Factor Analysis, Bayesian Confirmatory Factor Analysis, and sequential Confirmatory Factor Analysis to test performance. Our findings demonstrate that sequential Confirmatory Factor Analysis significantly outperforms the traditional model for indices with simple/moderate complexity. Traditional Confirmatory Factor Analysis performs better where the data are skewed. Where the hierarchical model becomes complex, the two methods perform similarly. Finally, sequential Confirmatory Factor Analysis can provide valid estimates where traditional or Bayesian Confirmatory Factor Analysis fail to converge. |
| title | Sequential Confirmatory Factor Analysis: A Novel Approach to Latent Variable Measurement |
| topic | Methodology |
| url | https://arxiv.org/abs/2508.15611 |