2.50
Hdl Handle:
http://hdl.handle.net/10755/169893
Type:
Research Study
Title:
Simple Rules for Calculating Degrees of Freedom for Structural Equation Models
Abstract:
Simple Rules for Calculating Degrees of Freedom for Structural Equation Models
Structural equation models (SEM's) are being increasingly used in nursing research to test theoretical models of multivariate relationships. Graphically oriented software programs (e.g., OS) have made SEM statistical techniques easier to use. However, even with these new programs, issues of model specification and identification remain difficult for many nurse researchers. Understanding how to calculate degrees of freedom is fundamental for addressing these issues, and is also essential for determining sample size adequacy and power. Accurately calculating degrees of freedom in SEM has been difficult because current guidelines are complex, and not easily generalizable to a broad array of models. In this paper, we describe a simple procedure for computing degrees of freedom in SEM's. We show that once the usual SEM constants are taken into account, a few simple rules will allow correct calculation of degrees of freedom in a wide variety of SEM models. We illustrate the use of this procedure in several types of SEM's including, confirmatory factory analysis, path analysis, and latent variable models. An additional advantage of using this calculation procedure is that it ensures that the model is correctly specified. Thus, this method facilitates the translation of a proposed model into a form that is readily testable with graphically oriented SEM software popular today.
Research Data

Ending Year:
Design:
Study Type:
Theoretical Framework:
Description of Sample:
Sample Size:
Number of Groups:
Sampling Plan:
Gender:
Minimum Age:
Maximum Age:
Data Collection Settings(s):

Repository Posting Date:
27-Oct-2011
Date of Publication:
17-Oct-2011

Full metadata record

DC FieldValue Language
dc.typeResearch Studyen_GB
dc.titleSimple Rules for Calculating Degrees of Freedom for Structural Equation Modelsen_GB
dc.identifier.urihttp://hdl.handle.net/10755/169893-
dc.description.abstract<table><tr><td colspan="2" class="item-title">Simple Rules for Calculating Degrees of Freedom for Structural Equation Models</td></tr><tr><td colspan="2" class="item-abstract">Structural equation models (SEM's) are being increasingly used in nursing research to test theoretical models of multivariate relationships. Graphically oriented software programs (e.g., OS) have made SEM statistical techniques easier to use. However, even with these new programs, issues of model specification and identification remain difficult for many nurse researchers. Understanding how to calculate degrees of freedom is fundamental for addressing these issues, and is also essential for determining sample size adequacy and power. Accurately calculating degrees of freedom in SEM has been difficult because current guidelines are complex, and not easily generalizable to a broad array of models. In this paper, we describe a simple procedure for computing degrees of freedom in SEM's. We show that once the usual SEM constants are taken into account, a few simple rules will allow correct calculation of degrees of freedom in a wide variety of SEM models. We illustrate the use of this procedure in several types of SEM's including, confirmatory factory analysis, path analysis, and latent variable models. An additional advantage of using this calculation procedure is that it ensures that the model is correctly specified. Thus, this method facilitates the translation of a proposed model into a form that is readily testable with graphically oriented SEM software popular today.</td></tr><tr><td colspan="2" class="researcher-header">Research Data</td></tr><tr><td colspan="2" class="researcher-data"><hr/></td></tr><tr class="data"><td class="label">Ending Year:</td><td class="value"></td></tr><tr class="data"><td class="label">Design:</td><td class="value"></td></tr><tr class="data"><td class="label">Study Type:</td><td class="value"></td></tr><tr class="data"><td class="label">Theoretical Framework:</td><td class="value"></td></tr><tr class="data"><td class="label">Description of Sample:</td><td class="value"></td></tr><tr class="data"><td class="label">Sample Size:</td><td class="value"></td></tr><tr class="data"><td class="label">Number of Groups:</td><td class="value"></td></tr><tr class="data"><td class="label">Sampling Plan:</td><td class="value"></td></tr><tr class="data"><td class="label">Gender:</td><td class="value"></td></tr><tr class="data"><td class="label">Minimum Age:</td><td class="value"></td></tr><tr class="data"><td class="label">Maximum Age:</td><td class="value"></td></tr><tr class="data"><td class="label">Data Collection Settings(s):</td><td class="value"></td></tr><tr><td colspan="2" class="researcher-data"><hr/></td></tr></table>en_GB
dc.date.available2011-10-27T21:30:56Z-
dc.date.issued2011-10-17en_GB
dc.date.accessioned2011-10-27T21:30:56Z-
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.