---
title: "Confirmatory Factor Analysis Workflow"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Confirmatory Factor Analysis Workflow}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
```

# Confirmatory Factor Analysis Workflow

This vignette outlines a small CFA workflow using the bundled continuous
dataset.

```{r setup}
library(PsychoMatic)
data(psychomatic_continuous)
```

## Specify A Model

```{r model}
model <- "
factor1 =~ item1 + item2 + item3
factor2 =~ item4 + item5 + item6
"
```

## Compare Candidate Models

`compare_models_auto()` uses `lavaan` directly and returns a compact table for
alternative CFA specifications. ML-type estimators report information criteria
when available, whereas WLSMV-style workflows compare models through scaled or
robust fit indices when available. The code is not evaluated during vignette
checks to keep package installation fast on constrained machines.

```{r compare-models, eval = FALSE}
models <- list(
  one_factor = "general =~ item1 + item2 + item3 + item4 + item5 + item6",
  two_factor = model
)
comparison <- compare_models_auto(
  psychomatic_continuous,
  models = models,
  estimator = "ML",
  language = "eng"
)
comparison$fit
```

## Full Automated CFA

The higher-level `cfa_auto()` function adds estimator selection, reliability,
fit interpretation, warnings, and reporting tables. It is shown but not run by
default in this vignette because it requires optional diagnostic packages.

```{r cfa-auto, eval = FALSE}
cfa_result <- cfa_auto(
  psychomatic_continuous,
  model = model,
  estimator = "ML",
  language = "eng"
)
summary(cfa_result)
```