Help for package pathmodelfit

Type:

Package

Title:

Path Component Fit Indices for Latent Structural Equation Models

Version:

1.0.5

Description:

Functions for computing fit indices for evaluating the path component of latent variable structural equation models. Available fit indices include RMSEA-P and NSCI-P originally presented and evaluated by Williams and O'Boyle (2011) <doi:10.1177/1094428110391472> and demonstrated by O'Boyle and Williams (2011) <doi:10.1037/a0020539> and Williams, O'Boyle, & Yu (2020) <doi:10.1177/1094428117736137>. Also included are fit indices described by Hancock and Mueller (2011) <doi:10.1177/0013164410384856>.

License:

GPL-3

Depends:

lavaan

Encoding:

UTF-8

LazyData:

true

RoxygenNote:

7.1.1

NeedsCompilation:

Packaged:

2020-09-02 03:23:21 UTC; steve

Author:

Steven Andrew Culpepper

[aut, cre], Larry Williams [aut]

Maintainer:

Steven Andrew Culpepper <sculpepp@illinois.edu>

Repository:

CRAN

Date/Publication:

2020-09-02 05:40:10 UTC

pathmodelfit: Path Component Fit Indices for Latent Structural Equation Models

Description

Author(s)

Maintainer: Steven Andrew Culpepper sculpepp@illinois.edu (ORCID)

Authors:

Larry Williams larry.williams@ttu.edu

References

Hancock, G. R., & Mueller, R. O. (2011). The reliability paradox in assessing structural relations within covariance structure models. Educational and Psychological Measurement, 71(2), 306-324.

McNeish, D., & Hancock, G. R. (2018). The effect of measurement quality on targeted structural model fit indices: A comment on Lance, Beck, Fan, and Carter (2016). Psychological Methods, 23(1), 184–190. https://doi.org/10.1037/met0000157

O'Boyle, E. H., Jr., & Williams, L. J. (2011). Decomposing model fit: Measurement vs. theory in organizational research using latent variables. Journal of Applied Psychology, 96(1), 1–12. https://doi.org/10.1037/a0020539

Williams, L. J., & O’Boyle, E. H. (2011). The myth of global fit indices and alternatives for assessing latent variable relations. Organizational Research Methods, 14, 350-369.

Williams, L. J., O’Boyle, E. H., & Yu, J. (2020). Condition 9 and 10 tests of model confirmation: A review of James, Mulaik, and Brett (1982) and contemporary alternatives. Organizational Research Methods, 23, 1, 6-29.

Examples


library(lavaan)

model4 <- '
Ldrrew =~ LdrrewI1 + LdrrewI2 + LdrrewI3
Jobcom =~ JobcomI1 + JobcomI2 + JobcomI3
Jobsat =~ JobsatI1 + JobsatI2 + JobsatI3
Orgcom =~ OrgcomI1 + OrgcomI2 + OrgcomI3
Jobsat ~ Ldrrew + Jobcom
Orgcom ~ Jobsat'

data(mediationVC)

fit <- sem(model4, sample.cov = mediationVC, sample.nobs = 232)
pathmodelfit(fit)

Williams and Anderson (1994) Mediated Multifoci Model Dataset

Description

This data set is from Williams and Anderson (1994) on the study of methods effects in organizational research using latent-variable models.

Usage

mediationVC

Format

A variance-covariance matrix for 232 observations and 12 variables. The variables are indicators of four constructss: 1) job satisfaction (Jobsat; 10 items), 2) organizational committment (Orgcom; 8 items), 3) leader-contingent reward behavior (Ldrrew; 10 items), and 4) job complexity (Jobcom; 6 items). The individual item responses were used to create three, total-score indicators for each construct defined as follows:

JobsatI1: Job satisfaction indicator 1
JobsatI2: Job satisfaction indicator 2
JobsatI3: Job satisfaction indicator 3
OrgcomI1: Organizational committment indicator 1
OrgcomI2: Organizational committment indicator 2
OrgcomI3: Organizational committment indicator 3
LdrrewI1: Leader-contingent reward behavior indicator 1
LdrrewI2: Leader-contingent reward behavior indicator 2
LdrrewI3: Leader-contingent reward behavior indicator 3
JobcomI1: Job complexity indicator 1
JobcomI2: Job complexity indicator 2
JobcomI3: Job complexity indicator 3

Author(s)

Steven Culpepper and Larry Williams

Source

Williams, L. J. & Anderson, S. E. (1994). An alternative approach to method effects by using latent-variable models: Applications in organizational behavior research. Journal of Applied Psychology, 79, 323-331.

Compute fit indices for the path component of latent variable structural equation models.

Description

pathmodelfit computes fit indices for evaluating the path component of latent variable structural equation models. Available fit indices include RMSEA-P and NSCI-P originally presented and evaluated by Williams and O'Boyle (2011) and demonstrated by O'Boyle and Williams (2011) and Williams, O'Boyle, & Yu, (2019). Also included are fit indices described by Hancock and Mueller (2011).

Usage

pathmodelfit(lavaanoutput)

Arguments

lavaanoutput

A lavaan sem object.

Value

A vector with RMSEA-P, a p-value for the chi-square test comparing the theoretical and saturated model, a 90 percent confidence interval for RMSEA-P, NSCI-P, and SRMRs, RMSEAs, TLIs, and CFIs.

References

Hancock, G. R., & Mueller, R. O. (2011). The reliability paradox in assessing structural relations within covariance structure models. Educational and Psychological Measurement, 71(2), 306-324.

Williams, L. J., & O’Boyle, E. H. (2011). The myth of global fit indices and alternatives for assessing latent variable relations. Organizational Research Methods, 14, 350-369.

Examples

library(lavaan)

model4 <- '
Ldrrew =~ LdrrewI1 + LdrrewI2 + LdrrewI3
Jobcom =~ JobcomI1 + JobcomI2 + JobcomI3
Jobsat =~ JobsatI1 + JobsatI2 + JobsatI3
Orgcom =~ OrgcomI1 + OrgcomI2 + OrgcomI3
Jobsat ~ Ldrrew + Jobcom
Orgcom ~ Jobsat'

data(mediationVC)

fit <- sem(model4, sample.cov = mediationVC, sample.nobs = 232)
pathmodelfit(fit)