Skip to main content

Heteroscedastic Latent Trait Models for Dichotomous Data

Psychometrika volume 80, pages 625–644 (2015)Cite this article

Abstract

Effort has been devoted to account for heteroscedasticity with respect to observed or latent moderator variables in item or test scores. For instance, in the multi-group generalized linear latent trait model, it could be tested whether the observed (polychoric) covariance matrix differs across the levels of an observed moderator variable. In the case that heteroscedasticity arises across the latent trait itself, existing models commonly distinguish between heteroscedastic residuals and a skewed trait distribution. These models have valuable applications in intelligence, personality and psychopathology research. However, existing approaches are only limited to continuous and polytomous data, while dichotomous data are common in intelligence and psychopathology research. Therefore, in present paper, a heteroscedastic latent trait model is presented for dichotomous data. The model is studied in a simulation study, and applied to data pertaining alcohol use and cognitive ability.

This is a preview of subscription content, access via your institution.

Access options

Buy single article

Instant access to the full article PDF.

43,34 €

Price includes VAT (Finland)
Tax calculation will be finalised during checkout.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Notes

  1. A sensible designation as ‘skedasis’ is the Greek word for scatter or dispersion.

  2. Truly continuous observed scores are rare in psychological and educational measurement. It has been shown that pragmatically, a scale with 7 or more ordered levels can be treated as continuous (Dolan, 1994). Therefore, in this paper, the term ‘continuous variable’ is used to denote variables with 7 or more ordered levels.

  3. In fact, Millsap and Yun-Tein (2004) use the restriction of \(\hbox {VAR}(\hbox {y}_{pi}^{*}) = 1\) instead of fixing \(\sigma _{\varepsilon i}^{2} = 1\) as is done here. Because, \(\hbox {VAR}(\hbox {y}_{pi}^{*})=\alpha _{i}^{2} \times \sigma _{\theta }^{2}+\sigma _{\varepsilon i}^{2}\) in which \(\sigma _{\theta }^{2}\) is already identified by fixing \(\sigma _{\theta }^{2 }= 1\) (or \(\alpha _{i} = 1\)), fixing \(\hbox {VAR}(\hbox {y}_{pi}) = 1\) will result in \(\sigma _{\varepsilon i}^{2} = 1 - \alpha _{i}^{2}\) (or \(\sigma _{\varepsilon i}^{2} = 1- \sigma _{\theta }^{2})\) which thus implies a fixed \(\sigma _{\varepsilon i}^{2}\). The opposite holds as well, that is, fixing \(\sigma _{\varepsilon i}^{2} = 1\) implies a fixed \(\hbox {VAR}(\hbox {y}_{pi}^{*})\).

  4. Note that Molenaar et al. (2012) used \(\delta _{0i}\) = 1.5 instead of \(\delta _{0i} = 1,\); therefore, in the present case, \(\delta _{1i}\) is rescaled to correspond to the effect size in Molenaar et al (i.e. due to the difference in \(\delta _{0i}\), there is no one-to-one correspondence).

  5. The opinions expressed in this article are those of the author and do not necessarily reflect the views of the ICPSR. 

  6. RMSEA values smaller than 0.05 are generally considered to indicate good model fit. CFI and TLI values larger than 0.95 are considered to indicate good model fit, see Hu and Bentler (1999)

References

  • Anderson, M. J. (2006). Distance-based tests for homogeneity of multivariate dispersions. Biometrics, 62, 245–253.

    Article  PubMed  Google Scholar 

  • Andrich, D. (1978). A rating formulation for ordered response categories. Psychometrika, 43, 561–573.

    Article  Google Scholar 

  • Austin, E. J., Deary, I. J., & Gibson, G. J. (1997). Relationships between ability and personality: Three hypotheses tested. Intelligence, 25(1), 49–70.

    Article  Google Scholar 

  • Azevedo, C. L. N., Bolfarine, H., & Andrade, D. F. (2011). Bayesian inference for a skew-normal IRT model under the centred parameterization. Computational Statistics & Data Analysis, 55, 353–365.

  • Azzalini, A. (1985). A class of distributions which includes the normal ones. Scandinavian Journal of Statistics, 12, 171–178.

    Google Scholar 

  • Azzalini, A. (1986). Further results on a class of distributions which includes the normal ones. Statistica, 46, 199–208.

    Google Scholar 

  • Azzalini, A., & Capatanio, A. (1999). Statistical applications of the multivariate skew normal distribution. Journal of the Royal Statistical Society, Series B, 61, 579–602.

    Article  Google Scholar 

  • Bakker, M., & Wicherts, J. M. (2013). Outlier removal, sum scores, and the inflation of the type I error rate in independent samples t tests: The power of alternatives and recommendations. Psychological Methods. doi:10.1037/met0000014.

  • Bartholomew, D. J., Knott, M., & Moustaki, I. (2011). Latent variable models and factor analysis: A unified approach. UK: Wiley.

    Book  Google Scholar 

  • Bauer, D. J., & Hussong, A. M. (2009). Psychometric approaches for developing commensurate measures across independent studies: Traditional and new models. Psychological Methods, 14(2), 101–125.

  • Bazán, J. L., Branco, M. D., & Bolfarine, H. (2006). A skew item response model. Bayesian Analysis, 1, 861–892.

    Article  Google Scholar 

  • Birnbaum, A. (1968). Some latent trait models and their use in inferring an examinee’s ability. In E. M. Lord & M. R. Novick (Eds.), Statistical theories of mental test scores (chap (pp. 17–20). Reading, MA: Addison Wesley.

    Google Scholar 

  • Bock, R. D. (1972). Estimating item parameters and latent ability when responses are scored in two or more nominal categories. Psychometrika, 37, 29–51.

    Article  Google Scholar 

  • Bock, R. D., & Aitkin, M. (1981). Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm. Psychometrika, 46, 443–459.

    Article  Google Scholar 

  • Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.

    Book  Google Scholar 

  • Bollen, K. A. (1996). A limited-information estimator for LISREL models with or without heteroscedastic errors. In G. A. Marcoulides & R. E. Schumacker (Eds.), Advanced structural equation modeling: Issues and techniques (pp. 227–241). Mahwah, NJ: Erlbaum.

    Google Scholar 

  • Brunner, E., Dette, H., & Munk, A. (1997). Box-Type Approximations in Nonparametric Factorial Designs. Journal of the American Statistical Association, 92, 1494–1502.

    Article  Google Scholar 

  • Christoffersson, A. (1975). Factor analysis of dichotomized variables. Psychometrika, 40, 5–32.

    Article  Google Scholar 

  • Dobson, A. J. (2010). An introduction to generalized linear models. Boca Raton, FL: CRC Press.

    Google Scholar 

  • Dolan, C. V. (1994). Factor analysis of variables with 2, 3, 5 and 7 response categories: A comparison of categorical variable estimators using simulated data. British Journal of Mathematical and Statistical Psychology, 47(2), 309–326.

  • Dolan, C. V., Colom, R., Abad, F. J., Wicherts, J. M., Hessen, D. J., & van de Sluis, S. (2006). Multi-group covariance and mean structure modeling of the relationship between the WAIS-III common factors and sex and educational attainment in Spain. Intelligence, 34, 193–210.

    Article  Google Scholar 

  • Dolan, C. V., & van der Maas, H. L. (1998). Fitting multivariage normal finite mixtures subject to structural equation modeling. Psychometrika, 63, 227–253.

    Article  Google Scholar 

  • Dolan, C. V., & Molenaar, P. (1991). A comparison of four methods of calculating standard errors of maximum-likelihood estimates in the analysis of covariance structure. British Journal of Mathematical and Statistical Psychology, 44, 359–368.

    Article  Google Scholar 

  • Fisher, R.A. (1928). The general sampling distribution of the multiple correlation coefficient. Proceedings of the Royal Society of London. Series A, 121, 654–673.

  • Fischer, G. H. (1983). Logistic latent trait models with linear constraints. Psychometrika, 48, 3–26.

    Article  Google Scholar 

  • Fokkema, M., Smits, N., Kelderman, H., & Cuijpers, P. (2013). Response shifts in mental health interventions: An illustration of longitudinal measurement invariance. Psychological Assessment, 25, 520–531.

    Article  PubMed  Google Scholar 

  • Greene, W. (2011). Econometric analysis (7th ed.). New York: Prentice Hall.

    Google Scholar 

  • Grzywacz, J. G., & Marks, N. F. (1999). Family solidarity and health behaviors: Evidence from the National Survey of Midlife Development in the United States. Journal of Family Issues, 20(2), 243–268.

  • Harvey, A.C. (1976). Estimating regression models with multiplicative heteroscedasticity. Econometrica: Journal of the Econometric Society, 44, 461–465.

  • Hessen, D. J., & Dolan, C. V. (2009). Heteroscedastic one-factor models and marginal maximum likelihood estimation. British Journal of Mathematical and Statistical Psychology, 62, 57–77.

    Article  PubMed  Google Scholar 

  • Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55.

  • Jansen, B. R., & van der Maas, H. L. (1997). Statistical test of the rule assessment methodology by latent class analysis. Developmental Review, 17, 321–357.

    Article  Google Scholar 

  • Jarque, C. M., & Bera, A. K. (1980). Efficient tests for normality, homoscedasticity and serial independence of regression residuals. Economics Letters, 6, 255–259.

    Article  Google Scholar 

  • Jedidi, K., Jagpal, H. S., & DeSarbo, W. S. (1997). Finite mixture structural equation models for response based segmentation and unobserved heterogeneity. Marketing Science, 16, 39–59.

    Article  Google Scholar 

  • Jöreskog, K. G. (1971). Simultaneous factor analysis in several populations. Psychometrika, 36, 409–426.

    Article  Google Scholar 

  • Jöreskog, K. G., & Moustaki, I. (2001). Factor analysis of ordinal variables: A comparison of three approaches. Multivariate Behavioral Research, 36, 347–387.

    Article  Google Scholar 

  • Kenny, D. A., & Judd, C. M. (1984). Estimating the nonlinear and interactive effects of latent variables. Psychological Bulletin, 96, 201–210.

    Article  Google Scholar 

  • Kirisci, L., Hsu, T., & Yu, L. (2001). Robustness of item parameter estimation programs to assumptions of unidimensionality and normality. Applied Psychological Measurement, 25, 146–162.

    Article  Google Scholar 

  • Klein, A., & Moosbrugger, H. (2000). Maximum likelihood estimation of latent interaction effects with the LMS method. Psychometrika, 65, 457–474.

    Article  Google Scholar 

  • Lee, S. Y., Poon, W. Y., & Bentler, P. M. (1989). Simultaneous analysis of multivariate polytomous variates in several groups. Psychometrika, 54, 63–73.

    Article  Google Scholar 

  • Lewin-Koh, S., & Amemiya, Y. (2003). Heteroscedastic factor analysis. Biometrika, 90, 85–97.

    Article  Google Scholar 

  • Long, J. S., & Ervin, L. H. (2000). Using heteroscedasticity consistent standard errors in the linear regression model. The American Statistician, 54, 217–224.

    Google Scholar 

  • Lord, F. M. (1952). A theory of test scores. New York: Psychometric Society.

    Google Scholar 

  • Lucke, J.F. (2012). Positive Trait Item Response Models. Proceedings of the 2012 Joint Statistical Meetings of the American Statistical Association, the International Biometric Society, the Institute of Mathematical Statistics, and Statistica Canada. San Diego, CA. 2012. Retrieved from http://works.bepress.com/joseph_lucke/35

  • Masters, G. N. (1982). A Rasch model for partial credit scoring. Psychometrika, 47, 149–174.

    Article  Google Scholar 

  • Mehta, P. D., & Neale, M. C. (2005). People are variables too: Multilevel structural equations modeling. Psychological Methods, 10, 259–284.

    Article  PubMed  Google Scholar 

  • Mehta, P. D., Neale, M. C., & Flay, B. R. (2004). Squeezing interval change from ordinal panel data: Latent growth curves with ordinal outcomes. Psychological Methods, 9, 301.

    Article  PubMed  Google Scholar 

  • Meijer, E., & Mooijaart, A. (1996). Factor analysis with heteroscedastic errors. British Journal of Mathematical and Statistical Psychology, 49, 189–202.

    Article  Google Scholar 

  • Merkle, E. C., Fan, J., & Zeileis, A. (2013). Testing for measurement invariance with respect to an ordinal variable. Psychometrika, 1–16. doi:10.1007/s11336-013-9376-7.

  • Merkle, E. C., & Zeileis, A. (2013). Tests of measurement invariance without subgroups: A generalization of classical methods. Psychometrika, 78, 59–82.

    Article  PubMed  Google Scholar 

  • Mellenbergh, G. J. (1994a). A unidimensional latent trait model for continuous item responses. Multivariate Behavioral Research, 29, 223–236.

    Article  Google Scholar 

  • Mellenbergh, G. J. (1994b). Generalized linear item response theory. Psychological Bulletin, 115, 300–307.

    Article  Google Scholar 

  • Meredith, W. (1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58, 525–543.

    Article  Google Scholar 

  • Millsap, R. E., & Yun-Tein, J. (2004). Assessing factorial invariance in ordered-categorical measures. Multivariate Behavioral Research, 39, 479–515.

    Article  Google Scholar 

  • Mislevy, R. J., & Verhelst, N. (1990). Modeling item responses when different subjects employ different solution strategies. Psychometrika, 55, 195–215.

    Article  Google Scholar 

  • Molenaar, D., Dolan, C. V., & de Boeck, P. (2012a). The heteroscedastic graded response model with a skewed latent trait: Testing statistical and substantive hypotheses related to skewed item category functions. Psychometrika, 77, 455–478.

    Article  Google Scholar 

  • Molenaar, D., Dolan, C. V., & van der Maas, H. L. (2011). Modeling ability differentiation in the second order factor model. Structural Equation Modeling, 18, 578–594.

    Article  Google Scholar 

  • Molenaar, D., Dolan, C. V., & Verhelst, N. D. (2010a). Testing and modeling non-normality within the one factor model. British Journal of Mathematical and Statistical Psychology, 63, 293–317.

  • Molenaar, D., van der Sluis, S., Boomsma, D. I., & Dolan, C. V. (2012b). Detecting specific genotype by environment interaction using marginal maximum likelihood estimation in the classical twin design. Behavior Genetics, 42, 483–499.

    PubMed Central  Article  PubMed  Google Scholar 

  • Molenaar, D., Dolan, C. V., Wicherts, J. M., & van der Maas, H. L. J. (2010b). Modeling differentiation of cognitive abilities within the higher-order factor model using moderated factor analysis. Intelligence, 38, 611–624.

    Article  Google Scholar 

  • Moustaki, I., & Knott, M. (2000). Generalized latent trait models. Psychometrika, 65, 391–411.

    Article  Google Scholar 

  • Murray, A. L., Dixon, H., & Johnson, W. (2013). Spearman’s law of diminishing returns: A statistical artifact? Intelligence, 41, 439–451.

    Article  Google Scholar 

  • Muthén, B. O. (1978). Contributions to factor analysis of dichotomous variables. Psychometrika, 43, 551–560.

    Article  Google Scholar 

  • Muthén, B. O., & Christoffersson, A. (1981). Simultaneous factor analysis of dichotomous variables in several groups. Psychometrika, 46, 407–419.

    Article  Google Scholar 

  • Muthén, L. K., & Muthén, B. O. (2007). Mplus user’s guide (5th edn.). Los Angeles, CA: Muthén & Muthén.

  • Neale, M. C., Aggen, S. H., Maes, H. H., Kubarych, T. S., & Schmitt, J. E. (2006a). Methodological issues in the assessment of substance use phenotypes. Addicitive Behavior, 31, 1010–34.

    Article  Google Scholar 

  • Neale, M. C., Boker, S. M., Xie, G., & Maes, H. H. (2006b). Mx: Statistical modeling (7th ed.). VCU, Richmond, VA: Author.

    Google Scholar 

  • Nocedal, J., & Wright, S. J. (2006). Numerical optimization (2nd ed.). New York: Springer.

    Google Scholar 

  • Olsson, U. (1979). Maximum likelihood estimation of the polychoric correlation coefficient. Psychometrika, 44, 443–460.

    Article  Google Scholar 

  • Purcell, S. (2002). Variance components models for gene-environment interaction in twin analysis. Twin Research, 5, 554–571.

    Article  PubMed  Google Scholar 

  • Rabe-Hesketh, S., Skrondal, A., & Pickles, A. (2004). Generalized multilevel structural equation modeling. Psychometrika, 69, 167–190.

    Article  Google Scholar 

  • R Core Team. (2012). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.

  • Raven, J. (2000). The Raven’s progressive matrices: Change and stability over culture and time. Cognitive Psychology, 41, 1–48.

    Article  PubMed  Google Scholar 

  • Ree, M. J. (1979). Estimating item characteristic curves. Applied Psychological Measurement, 3, 371–385.

    Article  Google Scholar 

  • Rost, J. (1990). Rasch models in latent classes: An integration of two approaches to item analysis. Applied Psychological Measurement, 14, 271–282.

    Article  Google Scholar 

  • Samejima, F. (1969). Estimation of ability using a response pattern of graded scores (Psychometric Monograph No. 17). Richmond, VA: The Psychometric Society.

  • Samejima, F. (1997). Departure from normal assumptions: A promise for future psychometrics with substantive mathematical modeling. Psychometrika, 62, 471–493.

    Article  Google Scholar 

  • Samejima, F. (2000). Logistic positive exponent family of models: Virtue of asymmetric item characteristic curves. Psychometrika, 65, 319–335.

    Article  Google Scholar 

  • Samejima, F. (2008). Graded response model based on the logistic positive exponent family of models for dichotomous responses. Psychometrika, 73, 561–578.

    Article  Google Scholar 

  • SAS Institute. (2011). SAS/STAT 9.3 user’s guide. SAS Institute.

  • Satorra, A., & Saris, W. E. (1985). The power of the likelihood ratio test in covariance structure analysis. Psychometrika, 50, 83–90.

    Article  Google Scholar 

  • Selzer, M. L. (1971). The michigan alcohol screening test: The quest for a new diagnostic instrument. American Journal of Psychiatry, 127, 89–94.

    Article  Google Scholar 

  • Seong, T. J. (1990). Sensitivity of marginal maximum likelihood estimation of item and ability parameters to the characteristics of the prior ability distributions. Applied Psychological Measurement, 14, 299–311.

    Article  Google Scholar 

  • Skrondal, A. (1996). Latent trait, multilevel and repeated measurement modeling with incomplete data of mixed measurement levels. Oslo: UiO .

    Google Scholar 

  • Skrondal, A., & Rabe-Hesketh, S. (2004). Generalized latent variable modeling: Multilevel, longitudinal, and structural equation models. Boca Raton, FL: Chapman and Hall/CRC.

    Book  Google Scholar 

  • Slutsky, E. E. (1913). On the criterion of goodness of fit of the regression lines and on the best method of fitting them to the data. Journal of the Royal Statistical Society, 77, 78–84.

    Article  Google Scholar 

  • Stevens, J. (2009). Applied multivariate statistics for the social sciences. USA: Taylor & Francis.

    Google Scholar 

  • Stone, C. A. (1992). Recovery of marginal maximum likelihood estimates in the two-parameter logistic response model: An evaluation of MULTILOG. Applied Psychological Measurement, 16, 1–16.

    Article  Google Scholar 

  • Swaminathan, H., & Gifford, J. (1983). Estimation of parameters in the three- parameter latent trait model. In D. J. Weiss (Ed.), New horizons in testing: Latent trait test theory and computerized adaptive testing (pp. 13–30). New York: Academic Press.

    Google Scholar 

  • Takane, Y., & de Leeuw, J. (1987). On the relationship between item response theory and factor analysis of discretized variables. Psychometrika, 52, 393–408.

    Article  Google Scholar 

  • Thissen, D., & Steinberg, L. (1984). A response model for multiple choice items. Psychometrika, 49, 501–519.

    Article  Google Scholar 

  • Thissen, D., & Steinberg, L. (1986). A taxonomy of item response models. Psychometrika, 51, 567–577.

    Article  Google Scholar 

  • Turkheimer, E., Haley, A., Waldron, M., D’Onofrio, B., & Gottesman, I. I. (2003). Socioeconomic status modifies heritability of IQ in young children. Psychological science, 14(6), 623–628.

    Article  PubMed  Google Scholar 

  • van der Sluis, S., Dolan, C. V., Neale, M. C., Boomsma, D. I., & Posthuma, D. (2006). Detecting genotype-environment interaction in monozygotic twin data: Comparing the Jinks & Fulker test and a new test based on marginal maximum likelihood estimation. Twin Research and Human Genetics, 9, 377–392.

    Article  PubMed  Google Scholar 

  • van der Sluis, S., Posthuma, D., & Dolan, C. V. (2012). A note on false positives and power in G\(\times \) E modelling of twin data. Behavior Genetics, 42, 170–186.

    PubMed Central  Article  PubMed  Google Scholar 

  • Wicherts, J. M., & Bakker, M. (2012). Publish (your data) or (let the data) perish! Why not publish your data too? Intelligence, 40, 73–76.

    Article  Google Scholar 

  • Zwinderman, A. H., & van den Wollenberg, A. L. (1990). Robustness of marginal maximum likelihood estimation in the Rasch model. Applied Psychological Measurement, 14, 73–81.

    Article  Google Scholar 

Download references

Acknowledgments

In memory of Roger Millsap whose comments and guidance were of great help in preparing the final version of this paper. I am also indebted to Mijke Rhemtulla, Conor Dolan, and two anonymous reviewers for their valuable comments on a previous draft of the manuscript.

Author information

Affiliations

  1. Psychological Methods, Department of Psychology, University of Amsterdam, Weesperplein 4, 1018 XA , Amsterdam, The Netherlands

    Dylan Molenaar

Authors
  1. Dylan Molenaar
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Dylan Molenaar.

Additional information

Correspondence should be sent to Dylan Molenaar, Psychological Methods, Department of Psychology, University of Amsterdam, Weesperplein 4, 1018 XA Amsterdam, The Netherlands. E-mail: D.Molenaar@uva.nl

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Molenaar, D. Heteroscedastic Latent Trait Models for Dichotomous Data. Psychometrika 80, 625–644 (2015). https://doi.org/10.1007/s11336-014-9406-0

Download citation

  • Received:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11336-014-9406-0

Keywords

  • heteroscedasticity
  • latent trait models
  • item response theory
  • two-parameter model
  • non-normal latent variables.