Modelling method effects as individual causal effects
Identifieur interne : 001B19 ( Istex/Corpus ); précédent : 001B18; suivant : 001B20Modelling method effects as individual causal effects
Auteurs : Steffi Pohl ; Rolf Steyer ; Katrin KrausSource :
- Journal of the Royal Statistical Society: Series A (Statistics in Society) [ 0964-1998 ] ; 2008-01.
English descriptors
- KwdEn :
Abstract
Summary. Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the method effect model, that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method j instead of k. They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as ‘feel bad’) as compared with positively formulated items (such as ‘feel good’) measuring mood states.
Url:
DOI: 10.1111/j.1467-985X.2007.00517.x
Links to Exploration step
ISTEX:D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Modelling method effects as individual causal effects</title><author><name sortKey="Pohl, Steffi" sort="Pohl, Steffi" uniqKey="Pohl S" first="Steffi" last="Pohl">Steffi Pohl</name><affiliation><mods:affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</mods:affiliation></affiliation></author><author><name sortKey="Steyer, Rolf" sort="Steyer, Rolf" uniqKey="Steyer R" first="Rolf" last="Steyer">Rolf Steyer</name><affiliation><mods:affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</mods:affiliation></affiliation></author><author><name sortKey="Kraus, Katrin" sort="Kraus, Katrin" uniqKey="Kraus K" first="Katrin" last="Kraus">Katrin Kraus</name><affiliation><mods:affiliation>University of Uppsala, Sweden</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974</idno><date when="2008" year="2008">2008</date><idno type="doi">10.1111/j.1467-985X.2007.00517.x</idno><idno type="url">https://api.istex.fr/document/D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">001B19</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001B19</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Modelling method effects as individual causal effects</title><author><name sortKey="Pohl, Steffi" sort="Pohl, Steffi" uniqKey="Pohl S" first="Steffi" last="Pohl">Steffi Pohl</name><affiliation><mods:affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</mods:affiliation></affiliation></author><author><name sortKey="Steyer, Rolf" sort="Steyer, Rolf" uniqKey="Steyer R" first="Rolf" last="Steyer">Rolf Steyer</name><affiliation><mods:affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</mods:affiliation></affiliation></author><author><name sortKey="Kraus, Katrin" sort="Kraus, Katrin" uniqKey="Kraus K" first="Katrin" last="Kraus">Katrin Kraus</name><affiliation><mods:affiliation>University of Uppsala, Sweden</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Journal of the Royal Statistical Society: Series A (Statistics in Society)</title><idno type="ISSN">0964-1998</idno><idno type="eISSN">1467-985X</idno><imprint><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><date type="published" when="2008-01">2008-01</date><biblScope unit="volume">171</biblScope><biblScope unit="issue">1</biblScope><biblScope unit="page" from="41">41</biblScope><biblScope unit="page" to="63">63</biblScope></imprint><idno type="ISSN">0964-1998</idno></series><idno type="istex">D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974</idno><idno type="DOI">10.1111/j.1467-985X.2007.00517.x</idno><idno type="ArticleID">RSSA517</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0964-1998</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Causality</term><term>Method effect</term><term>Multitrait</term><term>Negative item formulation</term><term>Structural equation modelling</term><term>multimethod</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract">Summary. Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the method effect model, that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method j instead of k. They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as ‘feel bad’) as compared with positively formulated items (such as ‘feel good’) measuring mood states.</div></front></TEI><istex><corpusName>wiley</corpusName><author><json:item><name>Steffi Pohl</name><affiliations><json:string>Friedrich‐Schiller‐Universität, Jena, Germany</json:string></affiliations></json:item><json:item><name>Rolf Steyer</name><affiliations><json:string>Friedrich‐Schiller‐Universität, Jena, Germany</json:string></affiliations></json:item><json:item><name>Katrin Kraus</name><affiliations><json:string>University of Uppsala, Sweden</json:string></affiliations></json:item></author><subject><json:item><lang><json:string>eng</json:string></lang><value>Causality</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Method effect</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Multitrait</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>multimethod</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Negative item formulation</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Structural equation modelling</value></json:item></subject><articleId><json:string>RSSA517</json:string></articleId><language><json:string>eng</json:string></language><originalGenre><json:string>article</json:string></originalGenre><abstract>Summary. Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the method effect model, that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method j instead of k. They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as ‘feel bad’) as compared with positively formulated items (such as ‘feel good’) measuring mood states.</abstract><qualityIndicators><score>8.5</score><pdfVersion>1.4</pdfVersion><pdfPageSize>484.7 x 697.3 pts</pdfPageSize><refBibsNative>true</refBibsNative><abstractCharCount>1656</abstractCharCount><pdfWordCount>11037</pdfWordCount><pdfCharCount>63819</pdfCharCount><pdfPageCount>23</pdfPageCount><abstractWordCount>259</abstractWordCount></qualityIndicators><title>Modelling method effects as individual causal effects</title><refBibs><json:item><author><json:item><name>L. F. Barrett</name></json:item><json:item><name>J. A. Russell</name></json:item></author><host><volume>74</volume><pages><last>984</last><first>967</first></pages><author></author><title>J. Personlty Socl Psychol.</title></host><title>Independence and bipolarity in the structure of current affect</title></json:item><json:item><host><author></author><title>Bollen, K. A. (1989) Structural Equations with Latent Variables. Oxford: Wiley.</title></host></json:item><json:item><author><json:item><name>M. W. Browne</name></json:item><json:item><name>R. Cudeck</name></json:item></author><host><pages><last>162</last><first>136</first></pages><author></author><title>Testing Structural Equation Models</title></host><title>Alternative ways of assessing model fit</title></json:item><json:item><author><json:item><name>D. T. Campbell</name></json:item><json:item><name>D. W. Fiske</name></json:item></author><host><volume>56</volume><pages><last>105</last><first>81</first></pages><author></author><title>Psychol. Bull.</title></host><title>Convergent and discriminant validation by multitrait‐multimethod matrix</title></json:item><json:item><author><json:item><name>D. T. Campbell</name></json:item><json:item><name>J. C Stanley</name></json:item></author><host><pages><last>246</last><first>171</first></pages><author></author><title>Handbook on Research on Teaching</title></host><title>Experimental and quasiexperimental designs for research on teaching</title></json:item><json:item><author><json:item><name>D. A. Cole</name></json:item><json:item><name>J. M. Martin</name></json:item><json:item><name>B. Powers</name></json:item><json:item><name>R. Truglio</name></json:item></author><host><volume>105</volume><pages><last>270</last><first>258</first></pages><author></author><title>J. Abnorm. Psychol.</title></host><title>Modeling causal relations between academic and social competence and depression: a multitrait‐multimethod longitudinal study of children</title></json:item><json:item><author><json:item><name>J. M. Conway</name></json:item><json:item><name>F. Lievens</name></json:item><json:item><name>S. E. Scullen</name></json:item><json:item><name>C. E. Lance</name></json:item></author><host><volume>11</volume><pages><last>559</last><first>535</first></pages><author></author><title>Struct. Equn Modlng</title></host><title>Bias in the correlated uniqueness model for MTMM data</title></json:item><json:item><host><author></author><title>Cook, T. D. and Campbell, D. T. (1979) Quasi‐experimentation: Design and Analysis Issues for Field Settings. Boston: Houghton Mifflin.</title></host></json:item><json:item><author><json:item><name>M. Eid</name></json:item></author><host><volume>65</volume><pages><last>261</last><first>241</first></pages><author></author><title>Psychometrika</title></host><title>A multitrait‐multimethod model with minimal assumptions</title></json:item><json:item><author><json:item><name>M. Eid</name></json:item><json:item><name>T. Lischetzke</name></json:item><json:item><name>F. Nussbeck</name></json:item><json:item><name>L. I. Trierweiler</name></json:item></author><host><volume>8</volume><pages><last>60</last><first>38</first></pages><author></author><title>Psychol. Meth.</title></host><title>Separating trait effects from trait‐specific method effects in multitrait‐multimethod models: a multiple‐indicator CT‐C(M‐1) model</title></json:item><json:item><host><author></author><title>Analyzing multitrait‐multimethod data with structural equation modeling: some guidelines for selecting an appropriate model</title></host></json:item><json:item><host><author></author><title>Fahrenberg, J., Hampel, R. and Selg, H. (1984) Das Freiburger Persönlichkeitsinventar (FPI und FPI‐R): Handbuch, 4th edn. Göttingen: Hogrefe.</title></host></json:item><json:item><author><json:item><name>G. E. Gignac</name></json:item></author><host><volume>34</volume><pages><last>46</last><first>29</first></pages><author></author><title>Intelligence</title></host><title>Evaluating subtest ‘g’ saturation levels via the single trait‐correlated uniqueness (STCU) SEM approach: evidence in favor of crystallized subtests as the best indicators of ‘g’</title></json:item><json:item><author><json:item><name>R. L. Holzbach</name></json:item></author><host><volume>63</volume><pages><last>588</last><first>579</first></pages><author></author><title>J. Appl. Psychol.</title></host><title>Rater bias in performance ratings: superior, self‐, and peer ratings</title></json:item><json:item><author><json:item><name>P. M. Horan</name></json:item><json:item><name>C. DiStefano</name></json:item><json:item><name>R. W. Motl</name></json:item></author><host><volume>10</volume><pages><last>455</last><first>435</first></pages><author></author><title>Equn Modlng</title></host><title>Wording effects in self‐esteem scales: methodological artifact or response style? Struc</title></json:item><json:item><author><json:item><name>K. G. Jöreskog</name></json:item></author><host><volume>36</volume><pages><last>426</last><first>409</first></pages><author></author><title>Psychometrika</title></host><title>Statistical analysis of sets of congeneric tests</title></json:item><json:item><author><json:item><name>K. G. Jöreskog</name></json:item></author><host><pages><last>56</last><first>1</first></pages><author></author><title>Contemporary Developments in Mathematical Psychology</title></host><title>Analyzing psychological data by structural analysis of covariance matrices</title></json:item><json:item><host><author></author><title>Jöreskog, K. G. and Sörbom, D. (2004) LISREL 8.7 for Windows. Lincolnwood: Scientific Software International.</title></host></json:item><json:item><author><json:item><name>D. A. Kenny</name></json:item></author><host><volume>12</volume><pages><last>252</last><first>247</first></pages><author></author><title>J. Expertl Socl Psychol.</title></host><title>An empirical application of confirmatory factor analysis to the multitrait‐multimethod matrix</title></json:item><json:item><host><author></author><title>Kraus, K. (2006) Analysis of multitrait‐multimethod data based on the theory of individual causal effects. Master's Thesis . Friedrich‐Schiller‐Universität, Jena.</title></host></json:item><json:item><author><json:item><name>A. I. Kraut</name></json:item><json:item><name>A. D. Wolfson</name></json:item><json:item><name>A. Rothenberg</name></json:item></author><host><volume>60</volume><pages><last>776</last><first>774</first></pages><author></author><title>J. Appl. Psychol.</title></host><title>Some effects of position on opinion survey items</title></json:item><json:item><author><json:item><name>L. M. Lewin</name></json:item><json:item><name>H. Hops</name></json:item><json:item><name>B. Davis</name></json:item><json:item><name>T. J. Dishion</name></json:item></author><host><volume>29</volume><pages><last>969</last><first>963</first></pages><author></author><title>Devlpmntl Psychol.</title></host><title>Multimethod comparison of similarity in school adjustment of siblings and unrelated children</title></json:item><json:item><host><author></author><title>Lischetzke, T., Eid, M. and Nussbeck, F. (2002) Unterschiedliche Definitionen von Methodeneffekten in MTMM Modellen und ihre Implikationen für die Analyse der Validität. 43rd Meet . German Psychological Association , Berlin .</title></host></json:item><json:item><host><author></author><title>Lord, F. M. and Novick, M. R. (1968) Statistical Theories of Mental Test Scores. Reading: Addison‐Wesley.</title></host></json:item><json:item><author><json:item><name>H. W. Marsh</name></json:item></author><host><volume>13</volume><pages><last>361</last><first>335</first></pages><author></author><title>Appl. Psychol. Measmnt</title></host><title>Confirmatory factor analyses of multitrait‐multimethod data: many problems and a few solutions</title></json:item><json:item><author><json:item><name>H. W. Marsh</name></json:item></author><host><volume>70</volume><pages><last>819</last><first>810</first></pages><author></author><title>J. Personlty Socl Psychol.</title></host><title>Positive and negative global self‐esteem: a substantively meaningful distinction or artifactors?</title></json:item><json:item><author><json:item><name>H. W. Marsh</name></json:item><json:item><name>B. M. Byrne</name></json:item></author><host><volume>45</volume><pages><last>58</last><first>49</first></pages><author></author><title>Aust. J. Psychol.</title></host><title>Do we see ourselves as others infer: a comparison of self‐other agreement on multiple dimensions of self‐concept from two continents</title></json:item><json:item><author><json:item><name>H. W. Marsh</name></json:item><json:item><name>R. G. Craven</name></json:item></author><host><volume>83</volume><pages><last>404</last><first>393</first></pages><author></author><title>J. Educ. Psychol.</title></host><title>Self‐other agreement on multiple dimensions of preadolescent self‐concept: inferences by teacher, mothers, and fathers</title></json:item><json:item><author><json:item><name>W. H. Marsh</name></json:item><json:item><name>D. Grayson</name></json:item></author><host><pages><last>198</last><first>177</first></pages><author></author><title>Structural Equation Modeling: Concepts, Issues and Applications</title></host><title>Latent variable models of multitrait‐multimethod data</title></json:item><json:item><host><author></author><title>Matthews, G. and Deary, I. J. (1998) Personality Traits. Cambridge: Cambridge University Press.</title></host></json:item><json:item><author><json:item><name>A. R. McConnell</name></json:item><json:item><name>J. M. Leibold</name></json:item></author><host><volume>37</volume><pages><last>442</last><first>435</first></pages><author></author><title>J. Exptl Socl Psychol.</title></host><title>Relations among the Implicit Association Test, discriminatory behavior, and explicit measures of racial attitudes</title></json:item><json:item><author><json:item><name>R. W. Motl</name></json:item><json:item><name>C. DiStefano</name></json:item></author><host><volume>9</volume><pages><last>578</last><first>562</first></pages><author></author><title>Struct. Equn Modlng</title></host><title>Longitudinal invariance of self‐esteem and method effects associated with negatively worded items</title></json:item><json:item><author><json:item><name>M. K. Mount</name></json:item></author><host><volume>37</volume><pages><last>702</last><first>687</first></pages><author></author><title>Personn. Psychol.</title></host><title>Psychometric properties of subordinate ratings of managerial performance</title></json:item><json:item><author><json:item><name>J. Neyman</name></json:item></author><host><volume>5</volume><pages><last>472</last><first>465</first></pages><author></author><title>Statist. Sci.</title></host><title>On the application of probability theory to agricultural experiments: essay on principles, sect. 9</title></json:item><json:item><author><json:item><name>M. R. Novick</name></json:item></author><host><volume>3</volume><pages><last>18</last><first>1</first></pages><author></author><title>J. Math. Psychol.</title></host><title>The axioms and principal results of classical test theory</title></json:item><json:item><author><json:item><name>T. M. Orthner</name></json:item></author><host><volume>46</volume><pages><last>476</last><first>446</first></pages><author></author><title>Psychol. Sci.</title></host><title>On changing the positions of items in personality questionnaires analysing effects of item sequence using IRT</title></json:item><json:item><author><json:item><name>D. B. Rubin</name></json:item></author><host><volume>66</volume><pages><last>701</last><first>688</first></pages><author></author><title>J. Educ. Psychol.</title></host><title>Estimating causal effects of treatments in randomized and nonrandomized studies</title></json:item><json:item><author><json:item><name>D. B. Rubin</name></json:item></author><host><volume>6</volume><pages><last>58</last><first>34</first></pages><author></author><title>Ann. Statist.</title></host><title>Bayesian inference for causal effects: the role of randomization</title></json:item><json:item><author><json:item><name>J. A. Russell</name></json:item><json:item><name>J. M. Carroll</name></json:item></author><host><volume>125</volume><pages><last>30</last><first>3</first></pages><author></author><title>Psychol. Bull.</title></host><title>On the bipolarity of positive and negative affect</title></json:item><json:item><author><json:item><name>M. Schmitt</name></json:item></author><host><pages><last>25</last><first>9</first></pages><author></author><title>Handbook of Multimethod Measurement in Psychology</title></host><title>Conceptual, theoretical and historical foundations of multimethod assessment</title></json:item><json:item><host><author></author><title>Shadish, W. R., Cook, T. D. and Campbell, D. T. (2002) Experimental and Quasi‐experimental Designs for Generalized Causal Inference. Boston: Houghton Mifflin.</title></host></json:item><json:item><author><json:item><name>R. Steyer</name></json:item></author><host><volume>3</volume><pages><last>60</last><first>25</first></pages><author></author><title>Methodika</title></host><title>Models of classical psychometric test theory as stochastic measurement models: representation, uniqueness, meaningfulness, identifiability, and testifiability</title></json:item><json:item><author><json:item><name>R. Steyer</name></json:item></author><host><pages><last>520</last><first>481</first></pages><author></author><title>International Encyclopedia of the Social and Behavioural Sciences: Logic of Inquiry and Research Design</title></host><title>Classical Test Theory</title></json:item><json:item><author><json:item><name>R. Steyer</name></json:item></author><host><volume>1</volume><pages><last>54</last><first>39</first></pages><author></author><title>Methodology</title></host><title>Analyzing individual and average causal effects via structural equation models</title></json:item><json:item><author><json:item><name>R. Steyer</name></json:item><json:item><name>M. Eid</name></json:item><json:item><name>P Schwenkmezger</name></json:item></author><host><author></author><title>Meth. Psychol. Res.</title></host><title>Modeling true intraindividual change: true change as a latent variable</title></json:item><json:item><author><json:item><name>R. Steyer</name></json:item><json:item><name>D. Ferring</name></json:item><json:item><name>M. Schmitt</name></json:item></author><host><volume>8</volume><pages><last>98</last><first>79</first></pages><author></author><title>Eur. J. Psychol. Assessmnt</title></host><title>States and traits in psychological assessment</title></json:item><json:item><host><author></author><title>Steyer, R., Partchev, I., Kröhne, U., Nagengast, B. and Fiege, C. (2007) Causal effects in between‐group experiments and quasi‐experiments: theory. Manuscript. Friedrich‐Schiller‐Universität Jena, Jena. (Available from http://www.causal‐effects.de/.)</title></host></json:item><json:item><author><json:item><name>R. Steyer</name></json:item><json:item><name>K. Riedl</name></json:item></author><host><pages><last>220</last><first>197</first></pages><author></author><title>Recent Developments on Structural Equation Modeling: Theory and Applications</title></host><title>Is it possible to feel good and bad at the same time?: new evidence on the bipolarity of mood‐state dimensions</title></json:item><json:item><author><json:item><name>R. Steyer</name></json:item><json:item><name>M. Schmitt</name></json:item><json:item><name>M. Eid</name></json:item></author><host><volume>13</volume><pages><last>408</last><first>389</first></pages><author></author><title>Eur. J. Personlty</title></host><title>Latent state‐trait theory and research in personality and individual differences</title></json:item><json:item><host><author></author><title>Steyer, R., Schwenkmezger, P., Eid, M. and Notz, P. (1991) Befindlichkeitsmessung und Latent‐State‐Trait‐Modelle. Arbeitsbericht, DFG‐Projekt STE 411/3‐1 . University of Trier, Trier.</title></host></json:item><json:item><host><author></author><title>Steyer, R., Schwenkmezger, P., Notz, P. and Eid, M. (1997) Der Mehrdimensionale Befindlichkeitsfragebogen (MDBF). Göttingen: Hogrefe.</title></host></json:item><json:item><host><author></author><title>Steyer, R., Schwenkmezger, P., Notz, P. and Eid, M. (2004) Entwicklung des Mehrdimensionalen Befindlichkeitsfragebogens (MDBF): Primärdatensatz. Trier: Psychologisches Datenarchiv PsychData des Zentrums für Psychologische Information und Dokumentation.</title></host></json:item><json:item><host><author></author><title>Vautier, S. and Pohl, S. (2007) Bipolarity and method effects in STAI scores. Manuscript. To be published.</title></host></json:item><json:item><author><json:item><name>S. Vautier</name></json:item><json:item><name>R. Steyer</name></json:item><json:item><name>A Boomsma</name></json:item></author><host><author></author><title>Br. J. Math. Statist. Psychol.</title></host><title>A true‐change model with individual method effects: reliability issues</title></json:item><json:item><author><json:item><name>S. Vautier</name></json:item><json:item><name>R. Steyer</name></json:item><json:item><name>S. Jmel</name></json:item><json:item><name>E. Raufaste</name></json:item></author><host><volume>12</volume><pages><last>410</last><first>391</first></pages><author></author><title>Struct. Equn Modlng</title></host><title>Imperfect or perfect dynamic bipolarity?: the case of antonymous affective judgments</title></json:item><json:item><author><json:item><name>P. Villar</name></json:item><json:item><name>M. A. Luengo</name></json:item><json:item><name>J. A. Gómez‐Fraguela</name></json:item><json:item><name>E. Romero</name></json:item></author><host><volume>22</volume><pages><last>68</last><first>59</first></pages><author></author><title>Eur. J. Psychol. Assessmnt</title></host><title>Assessment of validity of parenting constructs using the multitrait‐multimethod model</title></json:item><json:item><author><json:item><name>K. F. Widaman</name></json:item></author><host><volume>9</volume><pages><last>26</last><first>1</first></pages><author></author><title>Appl. Psychol. Measmnt</title></host><title>Hierarchically nested covariance structure models for multitrait‐multimethod data</title></json:item><json:item><author><json:item><name>D. W. Zimmerman</name></json:item></author><host><volume>40</volume><pages><last>412</last><first>395</first></pages><author></author><title>Psychometrika</title></host><title>Probability spaces, Hilbert spaces, and the axioms of test theory</title></json:item></refBibs><genre><json:string>article</json:string></genre><host><volume>171</volume><publisherId><json:string>RSSA</json:string></publisherId><pages><total>23</total><last>63</last><first>41</first></pages><issn><json:string>0964-1998</json:string></issn><issue>1</issue><genre><json:string>journal</json:string></genre><language><json:string>unknown</json:string></language><eissn><json:string>1467-985X</json:string></eissn><title>Journal of the Royal Statistical Society: Series A (Statistics in Society)</title><doi><json:string>10.1111/(ISSN)1467-985X</json:string></doi></host><categories><wos><json:string>social science</json:string><json:string>social sciences, mathematical methods</json:string><json:string>science</json:string><json:string>statistics & probability</json:string></wos><scienceMetrix><json:string>natural sciences</json:string><json:string>mathematics & statistics</json:string><json:string>statistics & probability</json:string></scienceMetrix></categories><publicationDate>2008</publicationDate><copyrightDate>2008</copyrightDate><doi><json:string>10.1111/j.1467-985X.2007.00517.x</json:string></doi><id>D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974</id><score>0.02233306</score><fulltext><json:item><extension>pdf</extension><original>true</original><mimetype>application/pdf</mimetype><uri>https://api.istex.fr/document/D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974/fulltext/pdf</uri></json:item><json:item><extension>zip</extension><original>false</original><mimetype>application/zip</mimetype><uri>https://api.istex.fr/document/D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a" type="main" xml:lang="en">Modelling method effects as individual causal effects</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><availability><p>WILEY</p></availability><date>2008</date></publicationStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a" type="main" xml:lang="en">Modelling method effects as individual causal effects</title><author xml:id="author-1"><persName><forename type="first">Steffi</forename><surname>Pohl</surname></persName><affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</affiliation></author><author xml:id="author-2"><persName><forename type="first">Rolf</forename><surname>Steyer</surname></persName><affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</affiliation></author><author xml:id="author-3"><persName><forename type="first">Katrin</forename><surname>Kraus</surname></persName><affiliation>University of Uppsala, Sweden</affiliation></author></analytic><monogr><title level="j">Journal of the Royal Statistical Society: Series A (Statistics in Society)</title><idno type="pISSN">0964-1998</idno><idno type="eISSN">1467-985X</idno><idno type="DOI">10.1111/(ISSN)1467-985X</idno><imprint><publisher>Blackwell Publishing Ltd</publisher><pubPlace>Oxford, UK</pubPlace><date type="published" when="2008-01"></date><biblScope unit="volume">171</biblScope><biblScope unit="issue">1</biblScope><biblScope unit="page" from="41">41</biblScope><biblScope unit="page" to="63">63</biblScope></imprint></monogr><idno type="istex">D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974</idno><idno type="DOI">10.1111/j.1467-985X.2007.00517.x</idno><idno type="ArticleID">RSSA517</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>2008</date></creation><langUsage><language ident="en">en</language></langUsage><abstract><p>Summary. Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the method effect model, that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method j instead of k. They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as ‘feel bad’) as compared with positively formulated items (such as ‘feel good’) measuring mood states.</p></abstract><textClass xml:lang="en"><keywords scheme="keyword"><list><head>keywords</head><item><term>Causality</term></item><item><term>Method effect</term></item><item><term>Multitrait</term></item><item><term>multimethod</term></item><item><term>Negative item formulation</term></item><item><term>Structural equation modelling</term></item></list></keywords></textClass></profileDesc><revisionDesc><change when="2008-01">Published</change></revisionDesc></teiHeader></istex:fulltextTEI><json:item><extension>txt</extension><original>false</original><mimetype>text/plain</mimetype><uri>https://api.istex.fr/document/D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974/fulltext/txt</uri></json:item></fulltext><metadata><istex:metadataXml wicri:clean="Wiley, elements deleted: body"><istex:xmlDeclaration>version="1.0" encoding="UTF-8" standalone="yes"</istex:xmlDeclaration><istex:document><component version="2.0" type="serialArticle" xml:lang="en"><header><publicationMeta level="product"><publisherInfo><publisherName>Blackwell Publishing Ltd</publisherName><publisherLoc>Oxford, UK</publisherLoc></publisherInfo><doi origin="wiley" registered="yes">10.1111/(ISSN)1467-985X</doi><issn type="print">0964-1998</issn><issn type="electronic">1467-985X</issn><idGroup><id type="product" value="RSSA"></id><id type="publisherDivision" value="ST"></id></idGroup><titleGroup><title type="main" sort="JOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES A (STATISTICS IN SOCIETY)">Journal of the Royal Statistical Society: Series A (Statistics in Society)</title></titleGroup></publicationMeta><publicationMeta level="part" position="01001"><doi origin="wiley">10.1111/rssa.2008.171.issue-1</doi><numberingGroup><numbering type="journalVolume" number="171">171</numbering><numbering type="journalIssue" number="1">1</numbering></numberingGroup><coverDate startDate="2008-01">January 2008</coverDate></publicationMeta><publicationMeta level="unit" type="article" position="4" status="forIssue"><doi origin="wiley">10.1111/j.1467-985X.2007.00517.x</doi><idGroup><id type="unit" value="RSSA517"></id></idGroup><countGroup><count type="pageTotal" number="23"></count></countGroup><titleGroup><title type="tocHeading1"><i>Original articles</i></title></titleGroup><eventGroup><event type="firstOnline" date="2007-11-19"></event><event type="publishedOnlineFinalForm" date="2007-11-19"></event><event type="xmlConverted" agent="Converter:BPG_TO_WML3G version:2.3.2 mode:FullText source:FullText result:FullText" date="2010-03-16"></event><event type="xmlConverted" agent="Converter:WILEY_ML3G_TO_WILEY_ML3GV2 version:4.0.1" date="2014-03-20"></event><event type="xmlConverted" agent="Converter:WML3G_To_WML3G version:4.1.7 mode:FullText,remove_FC" date="2014-10-31"></event></eventGroup><numberingGroup><numbering type="pageFirst" number="41">41</numbering><numbering type="pageLast" number="63">63</numbering></numberingGroup><correspondenceTo>Steffi Pohl, Lehrstuhl für Methodenlehre und Evaluationsforschung, Friedrich‐Schiller‐Universität Jena, Am Steiger 3, Haus 1, 07743 Jena, Germany.
E‐mail: <email normalForm="steffi.pohl@uni-jena.de">steffi.pohl@uni‐jena.de</email></correspondenceTo><objectNameGroup><objectName elementName="appendix">Appendices</objectName></objectNameGroup><linkGroup><link type="toTypesetVersion" href="file:RSSA.RSSA517.pdf"></link></linkGroup></publicationMeta><contentMeta><unparsedEditorialHistory>[Received August 2006. Final revision August 2007]</unparsedEditorialHistory><countGroup><count type="figureTotal" number="6"></count><count type="tableTotal" number="3"></count></countGroup><titleGroup><title type="main">Modelling method effects as individual causal effects</title><title type="shortAuthors"><i>S. Pohl, R. Steyer and K. Kraus</i></title><title type="short"><i>Modelling Method Effects</i></title></titleGroup><creators><creator creatorRole="author" xml:id="cr1" affiliationRef="#a1"><personName><givenNames>Steffi</givenNames><familyName>Pohl</familyName></personName></creator><creator creatorRole="author" xml:id="cr2" affiliationRef="#a1"><personName><givenNames>Rolf</givenNames><familyName>Steyer</familyName></personName></creator><creator creatorRole="author" xml:id="cr3" affiliationRef="#a2"><personName><givenNames>Katrin</givenNames><familyName>Kraus</familyName></personName></creator></creators><affiliationGroup><affiliation xml:id="a1" countryCode="DE"><unparsedAffiliation>Friedrich‐Schiller‐Universität, Jena, Germany</unparsedAffiliation></affiliation><affiliation xml:id="a2" countryCode="SE"><unparsedAffiliation>University of Uppsala, Sweden</unparsedAffiliation></affiliation></affiliationGroup><keywordGroup xml:lang="en"><keyword xml:id="k1">Causality</keyword><keyword xml:id="k2">Method effect</keyword><keyword xml:id="k3">Multitrait</keyword><keyword xml:id="k4">multimethod</keyword><keyword xml:id="k5">Negative item formulation</keyword><keyword xml:id="k6">Structural equation modelling</keyword></keywordGroup><abstractGroup><abstract type="main" xml:lang="en"><p><b>Summary. </b> Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the <i>method effect model</i>, that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method <i>j</i> instead of <i>k</i>. They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as ‘feel bad’) as compared with positively formulated items (such as ‘feel good’) measuring mood states.</p></abstract></abstractGroup></contentMeta></header></component></istex:document></istex:metadataXml><mods version="3.6"><titleInfo lang="en"><title>Modelling method effects as individual causal effects</title></titleInfo><titleInfo type="abbreviated" lang="en"><title>Modelling Method Effects</title></titleInfo><titleInfo type="alternative" contentType="CDATA" lang="en"><title>Modelling method effects as individual causal effects</title></titleInfo><name type="personal"><namePart type="given">Steffi</namePart><namePart type="family">Pohl</namePart><affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">Rolf</namePart><namePart type="family">Steyer</namePart><affiliation>Friedrich‐Schiller‐Universität, Jena, Germany</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">Katrin</namePart><namePart type="family">Kraus</namePart><affiliation>University of Uppsala, Sweden</affiliation><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="article" displayLabel="article"></genre><originInfo><publisher>Blackwell Publishing Ltd</publisher><place><placeTerm type="text">Oxford, UK</placeTerm></place><dateIssued encoding="w3cdtf">2008-01</dateIssued><edition>[Received August 2006. Final revision August 2007]</edition><copyrightDate encoding="w3cdtf">2008</copyrightDate></originInfo><language><languageTerm type="code" authority="rfc3066">en</languageTerm><languageTerm type="code" authority="iso639-2b">eng</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType><extent unit="figures">6</extent><extent unit="tables">3</extent></physicalDescription><abstract>Summary. Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the method effect model, that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method j instead of k. They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as ‘feel bad’) as compared with positively formulated items (such as ‘feel good’) measuring mood states.</abstract><subject lang="en"><genre>keywords</genre><topic>Causality</topic><topic>Method effect</topic><topic>Multitrait</topic><topic>multimethod</topic><topic>Negative item formulation</topic><topic>Structural equation modelling</topic></subject><relatedItem type="host"><titleInfo><title>Journal of the Royal Statistical Society: Series A (Statistics in Society)</title></titleInfo><genre type="journal">journal</genre><identifier type="ISSN">0964-1998</identifier><identifier type="eISSN">1467-985X</identifier><identifier type="DOI">10.1111/(ISSN)1467-985X</identifier><identifier type="PublisherID">RSSA</identifier><part><date>2008</date><detail type="volume"><caption>vol.</caption><number>171</number></detail><detail type="issue"><caption>no.</caption><number>1</number></detail><extent unit="pages"><start>41</start><end>63</end><total>23</total></extent></part></relatedItem><identifier type="istex">D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974</identifier><identifier type="DOI">10.1111/j.1467-985X.2007.00517.x</identifier><identifier type="ArticleID">RSSA517</identifier><recordInfo><recordContentSource>WILEY</recordContentSource><recordOrigin>Blackwell Publishing Ltd</recordOrigin></recordInfo></mods></metadata><serie></serie></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Rhénanie/explor/UnivTrevesV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001B19 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 001B19 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Rhénanie
|area= UnivTrevesV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:D0BFF03C6DA96E4BCAD5ED77126F71FDE13C5974
|texte= Modelling method effects as individual causal effects
}}
| This area was generated with Dilib version V0.6.31. |  |