Flexible Cutoffs for Model Fit Evaluation in Covariance-Based Structural Models

A toolbox to derive flexible cutoffs for fit indices in 'Covariance-based Structural Equation Modeling' based on the paper by 'Niemand & Mai (2018)' . Flexible cutoffs are an alternative to fixed cutoffs - rules-of-thumb - regarding an appropriate cutoff for fit indices such as 'CFI' or 'SRMR'. It has been demonstrated that these flexible cutoffs perform better than fixed cutoffs in grey areas where misspecification is not easy to detect. The package provides an alternative to the tool at < https://flexiblecutoffs.org> as it allows to tailor flexible cutoffs to a given dataset and model, which is so far not available in the tool. The package simulates fit indices based on a given dataset and model and then estimates the flexible cutoffs. Some useful functions, e.g., to determine the 'GoF-' or 'BoF-nature' of a fit index, are provided. So far, additional options for a relative use (is a model better than another?) are provided in an exploratory manner.

Package FCO - Flexible Cutoffs for Model Fit Evaluation in Covariance-Based Structural Models

Changes in FCO

FCO 2.0.2

• Small changes to solve CRAN check errors on some platforms.

FCO 2.0.1

• Skewness and kurtosis now estimated by an internal function as semTools may be archived.

FCO 2.0.0

• New release including new functions (e.g., gen_fit2, flex_co2, plot_fit2)
• The new functions now incorporate multiple extensions of the tool for multple decision rules, Type I and II error control and convenience
• Old functions remain available for compatibility

FCO 0.8.0

• Changed vignette after release on CRAN
• Fixed issue in index_guess
• Allows to specify sample size n instead of dataset x

FCO 0.7.2

• New seed argument in gen_fit for reproducible cutoffs

FCO 0.7.1

• Added a NEWS.md file to track changes to the package.
• Minor revisions to tests

FCO 0.7.0

• Speed improvements in the vignette
• New naming scheme
• Minor revisions to the descriptions and references
• Added contributor

FCO 0.69

• Bug fixes in gen_fit for OS compatibility
• Improvements in the vignette

FCO 0.67

• First stable release

Description

The goal of FCO is to to derive flexible cutoffs for fit indices in Covariance-based Structural Equation Modeling based on the paper by Niemand & Mai (2018). Flexible cutoffs are an alternative to fixed cutoffs - rules-of-thumb - regarding an appropriate cutoff for fit indices such as CFI or SRMR. It has been demonstrated that these flexible cutoffs perform better than fixed cutoffs in grey areas where misspecification is not easy to detect. The package provides an alternative to the tool at flexiblecutoffs.org as it allows to tailor flexible cutoffs to a given dataset and model, which is so far not available in the tool. The package simulates fit indices based on a given dataset and model and then estimates the flexible cutoffs. Some useful functions, e.g., to determine the GoF or BoF-nature of a fit index, are provided. So far, additional options for a relative use (is a model better than another?) are provided in an exploratory manner. Starting with version 2, we offer a lot improvements and additional new decision rules as well as many flexibility options.

Installation

You can install the FCO from CRAN CRAN with:

install.packages("FCO")
library(FCO)

Example

This is the basic usage for FCO in case of deriving flexible cutoffs for a single model:

library(FCO)
library(lavaan)
HS.model <- ' visual  =~ x1 + x2 + x3
              textual =~ x4 + x5 + x6
              speed   =~ x7 + x8 + x9 '

fit <- cfa(
  HS.model,
  data = HolzingerSwineford1939
)

#Fit for the model
fitmeasures(fit)

#Simulation
#Note: Demonstration only! Please use higher numbers of replications for your applications (>= 500).
fits <- gen_fit2(fit = fit, rep = 100)

#Obtain and plot cutoffs
flex_co2(fits)
plot_fit2(fits)