Skip to main content
SpringerLink
Log in

Comparing two maximum likelihood algorithms for mixture Rasch models

  • Original Paper
  • Published:
Behaviormetrika Aims and scope Submit manuscript

Abstract

The mixture Rasch model is gaining popularity as it allows items to perform differently across subpopulations and hence addresses the violation of the unidimensionality assumption with traditional Rasch models. This study focuses on comparing two common maximum likelihood methods for estimating such models using Monte Carlo simulations. The conditional maximum likelihood (CML) and joint maximum likelihood (JML) estimations, as implemented in three popular R packages are compared by evaluating parameter recovery and class accuracy. The results suggest that in general, CML is preferred in parameter recovery and JML is preferred in identifying the correct number of classes. A set of guidelines is also provided regarding how sample sizes, test lengths or actual class probabilities affect the accuracy of estimation and number of classes, as well as how different information criteria compare in achieving class accuracy. Specific issues regarding the performance of particular R packages are highlighted in the study as well.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Akaike H (1973) Information theory and an extension of the maximum likelihood principle. In: Petroy BN, Csaki F (eds) The second international symposium on information theory. Akadmiai Kiado, Budapest

    MATH  Google Scholar 

  • Amthauer R (1953) Intelligenz–Struktur-test (IST) [Intelligence structure test IST]. Hogrefe, Göttingen

    Google Scholar 

  • Amthauer R, Brocke B, Liepmann D, Beauducel A (2001) Intelligenz–Struktur-Test 2000 R (I-S-T 2000 R) [Intelligence structure test IST 2000 R]. Hogrefe, Göttingen

    Google Scholar 

  • Andersen EB (1972) The numerical solution of a set of conditional estimation equations. J R Stat Soc B 34:42–54

    MathSciNet  MATH  Google Scholar 

  • Baghaei P, Carstensen C (2013) Fitting the mixed Rasch model to a reading comprehension test:identifying reader types. Pr Assess Res Eval 18(5):1–13

    Google Scholar 

  • Birnbaum A (1968) Some latent trait models and their use in inferring an examinee’s ability. In: Lord FM, Novick MR (eds) Statistical theories of mental test scores. Addison-Wesley, Reading, pp 395–479

    Google Scholar 

  • Bolt DM, Cohen AS, Wollack JA (2002) Item parameter estimation under conditions of test speededness: application of a Mixture Rasch model with ordinal constraints. J Educ Meas 39:331–348

    Article  Google Scholar 

  • Choi I-H, Paek I, Cho S-J (2017) The impact of various class-distinction features on model selection in the mixture Rasch model. J Exp Educ 85(3):411–424

    Article  Google Scholar 

  • Cohen AS, Bolt DM (2005) A mixture model analysis of differential item functioning. J Educ Meas 42:133–148

    Article  Google Scholar 

  • Dai Y (2013) A mixture Rasch model with a covariate: a simulation study via Bayesian Markov Chain Monte Carlo estimation. Appl Psychol Meas 37:375–396

    Article  Google Scholar 

  • Dempster A, Laird N, Rubin D (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc B 39(1):1–38

    MathSciNet  MATH  Google Scholar 

  • Eid M, Zickar MJ (2007) Detecting response styles and faking in personality and organizational assessments by mixed Rasch models. In: von Davier M, Carstensen CH (eds) Multivariate and mixture distribution Rasch models: extensions and applications. Springer, New York, pp 255–270

    Chapter  Google Scholar 

  • Finch WH, French BF (2012) Parameter estimation with mixture item response theory models: a Monte Carlo comparison of maximum likelihood and Bayesian methods. J Mod Appl Stat Methods 11(1):167–178

    Article  Google Scholar 

  • Fischer GH (1981) On the existence and uniqueness of maximum likelihood estimates in the Rasch model. Psychometrika 46:59–77

    Article  MathSciNet  MATH  Google Scholar 

  • Frick H, Strobl C, Leisch F, Zeileis A (2012) Flexible Rasch mixture models with package psychomix. J Stat Softw 48(7):1–25

    Article  Google Scholar 

  • Frick H, Strobl C, Zeileis A (2015) Rasch mixture models for DIF detection: a comparison of old and new score specifications. Educ Psychol Meas 75(2):208–234

    Article  Google Scholar 

  • Gilula Z, Haberman SJ (1995) Prediction functions for categorical panel data. Ann Stat 23:1130–1142

    Article  MathSciNet  MATH  Google Scholar 

  • Glück J, Machat R, Jirasko M, Rollett B (2002) Training-related changes in solution strategy in a spatial test: an application of item response models. Learn Individ Differ 13:1–22

    Article  Google Scholar 

  • Haberman SJ (1977) Log-linear models and frequency tables with small expected cell counts. Ann Stat 5:1148–1169

    Article  MathSciNet  MATH  Google Scholar 

  • Haberman SJ (2004) Joint and conditional maximum likelihood estimation for the Rasch model with binary responses. ETS research report RR-04-20

  • Henson JM, Reise SP, Kim KH (2007) Detecting mixtures from structural model differences using latent variable mixture modeling: a comparison of relative model fit statistics. Struct Equ Model 14:202–226

    Article  MathSciNet  Google Scholar 

  • Hong S, Min S (2007) Mixed Rasch modeling of the self-rating depression scale: incorporating latent class and Rasch rating scale models. Educ Psychol Meas 67(2):280–299

    Article  MathSciNet  Google Scholar 

  • Hosenfeld I, Strauss B, Köller O (1997) Gender differences in spatial ability tasks—a question of strategy? Zeitschrift für Pädagogische Psychol 11:84–94

    Google Scholar 

  • Janssen AB, Geiser C (2010) On the relationship between solution strategies in two mental rotation tasks. Learn Individ Differ 20(5):473–478

    Article  Google Scholar 

  • Jiao H, Lissitz R, Macready G, Wang S, Liang S (2011) Exploring levels of performance using the mixture Rasch model for standard setting. Psychol Test Assess Model 53:499–522

    Google Scholar 

  • Kaiser FG, Keller C (2001) Disclosing situational constraints to ecological behavior: a confirmatory application of the mixed Rasch model. Eur J Psychol Assess 17(3):212–221

    Article  Google Scholar 

  • Kim S-H (2001) An evaluation of a Markov chain Monte Carlo method for the Rasch model. Appl Psychol Meas 25:163–176

    Article  MathSciNet  Google Scholar 

  • Köller O, Rost J, Köller M (1994) Individuell differences in solving spatial tasks from the IST- and IST-70 subtest “cube comparison”. Zeitschrift für Psychol 202:65–85

    Google Scholar 

  • Leisch F (2004) FlexMix: a general framework for finite mixture models and latent class regression in R. J Stat Softw 11(8):1–18

    Article  Google Scholar 

  • Li F, Cohen AS, Kim S-H, Cho S-J (2009) Model selection methods for mixture dichotomous IRT models. Appl Psychol Meas 33:353–373

    Article  MathSciNet  Google Scholar 

  • Lu R, Jiao H (2009) Detecting DIF using mixture Rasch model. Paper presented at the annual meeting of the National Council on Measurement in Education, San Diego

  • Maij-de Meij A, Kelderman H, van der Flier H (2008) Fitting a mixture item response theory model to personality questionnaire data: characterizing latent classes and investigating possibilities for improving prediction. Appl Psychol Meas 32(8):611–631

    Article  MathSciNet  Google Scholar 

  • Masters GN (1982) A Rasch model for partial credit scoring. Psychometrika 47:149–174

    Article  MATH  Google Scholar 

  • McCutcheon AC (1987) Latent class analysis. Sage, Beverly Hills

    Book  Google Scholar 

  • McLachlan G, Krishnan T (1997) The EM algorithm and extensions. Wiley, New York

    MATH  Google Scholar 

  • Nylund KL, Asparouhov T, Muthen BO (2007) Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct Equ Model 14:535–569

    Article  MathSciNet  Google Scholar 

  • Preinerstorfer D, Formann A (2012) Parameter recovery and model selection in mixed Rasch models. Br J Math Stat Psychol 65(2):251–262

    Article  MathSciNet  MATH  Google Scholar 

  • Putz-Osterloh W (1977) Über Problemlöseprozesse bei dem Test Würfelaufgaben aus dem Intelligenzstrukturtest IST und IST-70 von Amthauer [On solution processes in the test cube comparisons from Amthauer’s Intelligence Structure Test IST and IST-70]. Diagnostica 23:252–265

    Google Scholar 

  • R Core Team (2016) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. https://www.R-project.org/. Accessed 9 Jan 2019

  • Rasch G (1960) Probabilistic models for some intelligence and attainment tests. The University of Chicago Press, Chicago

    Google Scholar 

  • Rasch G (1961) On general laws and the meaning of measurement in psychology. In Neyman J (ed) Proceedings of the 4th Berkeley symposium on mathematical statistics and probability, vol IV. University of California Press, Berkeley, pp 321–333

  • Rice K (2004) Equivalence between conditional and mixture approaches to the Rasch model and matched case–control studies, with applications. J Am Stat Assoc 99(466):510–522

    Article  MathSciNet  MATH  Google Scholar 

  • Rost J (1990) Rasch models in latent classes: an integration of two approaches to item analysis. Appl Psychol Meas 14:271–282

    Article  Google Scholar 

  • Rost J (1991) A logistic mixture distribution model for polychotomous item responses. Br J Math Stat Psychol 44:75–92

    Article  Google Scholar 

  • Rost J, von Davier M (1994) A conditional item-fit index for Rasch models. Appl Psychol Meas 18(2):171–182

    Article  Google Scholar 

  • Schwarz G (1978) Estimating the dimension of a model. Ann Stat 6:461–464

    Article  MathSciNet  MATH  Google Scholar 

  • Sclove LS (1987) Application of model-selection criteria to some problems in multivariate analysis. Psychometrika 52:333–343

    Article  Google Scholar 

  • Sen S (2018) Spurious latent class problem in the mixed Rasch model: a comparison of three maximum likelihood estimation methods under different ability distributions. Int J Test 18:71–100

    Article  Google Scholar 

  • Sinharay S (2003) Practical applications of posterior predictive model checking for assessing fit of common item response theory models. (ETS research Rep. No. RR-03-33). Educational Testing Service, Princeton

  • Thissen D (1982) Marginal maximum likelihood estimation for the one-parameter logistic model. Psychometrika 47:175–186

    Article  MATH  Google Scholar 

  • Tofighi D, Enders CK (2008) Identifying the correct number of classes in growth mixture models. In: Hancock GR, Samuelsen KM (eds) Advances in latent variable mixture models. Information Age Publications, Charlotte, pp 317–341

    Google Scholar 

  • von Davier M (1994) WINMIRA (a windows program for analyses with the Rasch model, with the latent class analysis and with the mixed Rasch model) [Computer software]. Institute for Science Education, Kiel

    Google Scholar 

  • von Davier M (1997) Methoden zur Prufung probabilistischer Testmodelle, vol 157. IPN, Kiel

    Google Scholar 

  • von Davier M, Carstensen CH (eds) (2007) Multivariate and mixture distribution Rasch models: extensions and applications. Springer, New York

    MATH  Google Scholar 

  • Willse JT (2011) Mixture Rasch models with joint maximum likelihood estimation. Psychol Meas 71:5–19

    Article  Google Scholar 

  • Zickar M, Gibby R, Robie C (2004) Uncovering faking samples in applicant, incumbent, and experimental data sets: an application of mixed-model item response theory. Organ Res Methods 7(2):168–190

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Mathematical Sciences Department, Worcester Polytechnic Institute, 100 Institute Road, Worcester, MA, 01609-2280, USA

    Yevgeniy Ptukhin

  2. Southern Illinois University – Carbondale, Carbondale, IL, USA

    Yanyan Sheng

Authors
  1. Yevgeniy Ptukhin
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yanyan Sheng
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yevgeniy Ptukhin.

Ethics declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Communicated by Marie Wiberg.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ptukhin, Y., Sheng, Y. Comparing two maximum likelihood algorithms for mixture Rasch models. Behaviormetrika 46, 101–119 (2019). https://doi.org/10.1007/s41237-019-00076-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s41237-019-00076-6

Keywords

Search

Navigation