The Effect of Estimation Methods on SEM Fit Indices
We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI...
Gespeichert in:
| Veröffentlicht in: | Educational and psychological measurement 2020-06, Vol.80 (3), p.421-445 |
|---|---|
| Hauptverfasser: | , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 445 |
|---|---|
| container_issue | 3 |
| container_start_page | 421 |
| container_title | Educational and psychological measurement |
| container_volume | 80 |
| creator | Shi, Dexin Maydeu-Olivares, Alberto |
| description | We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). We considered different types and levels of misspecification in factor analysis models: misspecified dimensionality, omitting cross-loadings, and ignoring residual correlations. Estimation methods had substantial impacts on the RMSEA and CFI so that different cutoff values need to be employed for different estimators. In contrast, SRMR is robust to the method used to estimate the model parameters. The same criterion can be applied at the population level when using the SRMR to evaluate model fit, regardless of the choice of estimation method. |
| doi_str_mv | 10.1177/0013164419885164 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmed_primary_32425213</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ericid>EJ1253305</ericid><sage_id>10.1177_0013164419885164</sage_id><sourcerecordid>2401740013</sourcerecordid><originalsourceid>FETCH-LOGICAL-c550t-53664d591c71bb96ec75a1e17499d1686952ec5ca227770a271d9f3bd764a2283</originalsourceid><addsrcrecordid>eNp1kT1PwzAQhi0EouVjZwFFYmEJ-PwRJwsSqlIoasVAmS3HcdpUaVziFIl_j6OWApXw4tO9z73n8yF0AfgWQIg7jIFCxBgkccx9cID6wDkJaRzHh6jfyWGn99CJcwvsDwM4Rj1KGOEEaB_R6dwEaVEY3Qa2CFLXlkvVlrYOJqad29wFPnxNJ8GwbINRnZfauDN0VKjKmfPtfYrehul08BSOXx5Hg4dxqDnHbchpFLGcJ6AFZFkSGS24AgOCJUkOURwlnBjNtSJECIEVEZAnBc1yETGfi-kput_4rtbZ0uTa1G2jKrlq_BObT2lVKf8qdTmXM_shBSHAEvAGlxsD05R6V5c-A-GUYu71m22Dxr6vjWvlsnTaVJWqjV07SRhmEfNW1KPXe-jCrpvaj99Rfqburz2FN5RurHONKXZNActuY3J_Y77k6veQu4LvFXkg3ABOzcxP138NvwBEF5jZ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2401740013</pqid></control><display><type>article</type><title>The Effect of Estimation Methods on SEM Fit Indices</title><source>ERIC - Full Text Only (Discovery)</source><source>SAGE Complete A-Z List</source><source>EZB-FREE-00999 freely available EZB journals</source><source>PubMed Central</source><creator>Shi, Dexin ; Maydeu-Olivares, Alberto</creator><creatorcontrib>Shi, Dexin ; Maydeu-Olivares, Alberto</creatorcontrib><description>We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). We considered different types and levels of misspecification in factor analysis models: misspecified dimensionality, omitting cross-loadings, and ignoring residual correlations. Estimation methods had substantial impacts on the RMSEA and CFI so that different cutoff values need to be employed for different estimators. In contrast, SRMR is robust to the method used to estimate the model parameters. The same criterion can be applied at the population level when using the SRMR to evaluate model fit, regardless of the choice of estimation method.</description><identifier>ISSN: 0013-1644</identifier><identifier>EISSN: 1552-3888</identifier><identifier>DOI: 10.1177/0013164419885164</identifier><identifier>PMID: 32425213</identifier><language>eng</language><publisher>Los Angeles, CA: SAGE Publications</publisher><subject>Computation ; Error of Measurement ; Estimating techniques ; Factor Analysis ; Goodness of Fit ; Least Squares Statistics ; Maximum likelihood method ; Maximum Likelihood Statistics ; Structural equation modeling ; Structural Equation Models</subject><ispartof>Educational and psychological measurement, 2020-06, Vol.80 (3), p.421-445</ispartof><rights>The Author(s) 2019</rights><rights>The Author(s) 2019.</rights><rights>The Author(s) 2019 2019 SAGE Publications</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c550t-53664d591c71bb96ec75a1e17499d1686952ec5ca227770a271d9f3bd764a2283</citedby><cites>FETCH-LOGICAL-c550t-53664d591c71bb96ec75a1e17499d1686952ec5ca227770a271d9f3bd764a2283</cites><orcidid>0000-0002-4120-6756</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC7221491/pdf/$$EPDF$$P50$$Gpubmedcentral$$H</linktopdf><linktohtml>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC7221491/$$EHTML$$P50$$Gpubmedcentral$$H</linktohtml><link.rule.ids>230,314,690,727,780,784,885,21819,27924,27925,43621,43622,53791,53793</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/32425213$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink><backlink>$$Uhttp://eric.ed.gov/ERICWebPortal/detail?accno=EJ1253305$$DView record in ERIC$$Hfree_for_read</backlink></links><search><creatorcontrib>Shi, Dexin</creatorcontrib><creatorcontrib>Maydeu-Olivares, Alberto</creatorcontrib><title>The Effect of Estimation Methods on SEM Fit Indices</title><title>Educational and psychological measurement</title><addtitle>Educ Psychol Meas</addtitle><description>We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). We considered different types and levels of misspecification in factor analysis models: misspecified dimensionality, omitting cross-loadings, and ignoring residual correlations. Estimation methods had substantial impacts on the RMSEA and CFI so that different cutoff values need to be employed for different estimators. In contrast, SRMR is robust to the method used to estimate the model parameters. The same criterion can be applied at the population level when using the SRMR to evaluate model fit, regardless of the choice of estimation method.</description><subject>Computation</subject><subject>Error of Measurement</subject><subject>Estimating techniques</subject><subject>Factor Analysis</subject><subject>Goodness of Fit</subject><subject>Least Squares Statistics</subject><subject>Maximum likelihood method</subject><subject>Maximum Likelihood Statistics</subject><subject>Structural equation modeling</subject><subject>Structural Equation Models</subject><issn>0013-1644</issn><issn>1552-3888</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GA5</sourceid><recordid>eNp1kT1PwzAQhi0EouVjZwFFYmEJ-PwRJwsSqlIoasVAmS3HcdpUaVziFIl_j6OWApXw4tO9z73n8yF0AfgWQIg7jIFCxBgkccx9cID6wDkJaRzHh6jfyWGn99CJcwvsDwM4Rj1KGOEEaB_R6dwEaVEY3Qa2CFLXlkvVlrYOJqad29wFPnxNJ8GwbINRnZfauDN0VKjKmfPtfYrehul08BSOXx5Hg4dxqDnHbchpFLGcJ6AFZFkSGS24AgOCJUkOURwlnBjNtSJECIEVEZAnBc1yETGfi-kput_4rtbZ0uTa1G2jKrlq_BObT2lVKf8qdTmXM_shBSHAEvAGlxsD05R6V5c-A-GUYu71m22Dxr6vjWvlsnTaVJWqjV07SRhmEfNW1KPXe-jCrpvaj99Rfqburz2FN5RurHONKXZNActuY3J_Y77k6veQu4LvFXkg3ABOzcxP138NvwBEF5jZ</recordid><startdate>20200601</startdate><enddate>20200601</enddate><creator>Shi, Dexin</creator><creator>Maydeu-Olivares, Alberto</creator><general>SAGE Publications</general><general>SAGE PUBLICATIONS, INC</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>ERI</scope><scope>GA5</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-4120-6756</orcidid></search><sort><creationdate>20200601</creationdate><title>The Effect of Estimation Methods on SEM Fit Indices</title><author>Shi, Dexin ; Maydeu-Olivares, Alberto</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c550t-53664d591c71bb96ec75a1e17499d1686952ec5ca227770a271d9f3bd764a2283</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Computation</topic><topic>Error of Measurement</topic><topic>Estimating techniques</topic><topic>Factor Analysis</topic><topic>Goodness of Fit</topic><topic>Least Squares Statistics</topic><topic>Maximum likelihood method</topic><topic>Maximum Likelihood Statistics</topic><topic>Structural equation modeling</topic><topic>Structural Equation Models</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shi, Dexin</creatorcontrib><creatorcontrib>Maydeu-Olivares, Alberto</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>ERIC</collection><collection>ERIC - Full Text Only (Discovery)</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Educational and psychological measurement</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shi, Dexin</au><au>Maydeu-Olivares, Alberto</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><ericid>EJ1253305</ericid><atitle>The Effect of Estimation Methods on SEM Fit Indices</atitle><jtitle>Educational and psychological measurement</jtitle><addtitle>Educ Psychol Meas</addtitle><date>2020-06-01</date><risdate>2020</risdate><volume>80</volume><issue>3</issue><spage>421</spage><epage>445</epage><pages>421-445</pages><issn>0013-1644</issn><eissn>1552-3888</eissn><abstract>We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). We considered different types and levels of misspecification in factor analysis models: misspecified dimensionality, omitting cross-loadings, and ignoring residual correlations. Estimation methods had substantial impacts on the RMSEA and CFI so that different cutoff values need to be employed for different estimators. In contrast, SRMR is robust to the method used to estimate the model parameters. The same criterion can be applied at the population level when using the SRMR to evaluate model fit, regardless of the choice of estimation method.</abstract><cop>Los Angeles, CA</cop><pub>SAGE Publications</pub><pmid>32425213</pmid><doi>10.1177/0013164419885164</doi><tpages>25</tpages><orcidid>https://orcid.org/0000-0002-4120-6756</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0013-1644 |
| ispartof | Educational and psychological measurement, 2020-06, Vol.80 (3), p.421-445 |
| issn | 0013-1644 1552-3888 |
| language | eng |
| recordid | cdi_pubmed_primary_32425213 |
| source | ERIC - Full Text Only (Discovery); SAGE Complete A-Z List; EZB-FREE-00999 freely available EZB journals; PubMed Central |
| subjects | Computation Error of Measurement Estimating techniques Factor Analysis Goodness of Fit Least Squares Statistics Maximum likelihood method Maximum Likelihood Statistics Structural equation modeling Structural Equation Models |
| title | The Effect of Estimation Methods on SEM Fit Indices |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-06T04%3A52%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=The%20Effect%20of%20Estimation%20Methods%20on%20SEM%20Fit%20Indices&rft.jtitle=Educational%20and%20psychological%20measurement&rft.au=Shi,%20Dexin&rft.date=2020-06-01&rft.volume=80&rft.issue=3&rft.spage=421&rft.epage=445&rft.pages=421-445&rft.issn=0013-1644&rft.eissn=1552-3888&rft_id=info:doi/10.1177/0013164419885164&rft_dat=%3Cproquest_pubme%3E2401740013%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2401740013&rft_id=info:pmid/32425213&rft_ericid=EJ1253305&rft_sage_id=10.1177_0013164419885164&rfr_iscdi=true |