Convergence: Statistics or Substance?

Estimation frequently requires iterative procedures: the more iterations, the more accurate estimates. But when are estimates accurate enough? When can iteration cease? My the rule has become "Convergence is reached when more iterations do not change my interpretation of the estimates".

There is a trade-off between accuracy and speed. Greater accuracy requires more iterations - more time and computer resources. The specification of estimation accuracy is a compromise. Frequently, squeezing that last bit of inaccuracy out of estimates only affects the least significant digits of printed output, has no noticeable effect on model-data fit, and does not alter interpretation. Three numerical convergence rules are often employed:

1) Estimates are pronounced "accurate enough" when a predetermined "maximum" number of iterations have been performed.

2) Estimates are deemed converged when no estimate changes more than a small pre-set "tolerance" value during an iteration.

3) Estimates have converged when there is less residual difference between the observed data and that expected than can actually be observed.

Be wary! In a recent analysis of responses to a set of math tests, linked in block diagonal matrix form, I set these three convergence criteria to reasonable values. The computer program BIGSTEPS ran smoothly. All appeared well. The outcome is shown in Figure 1. As most of us would expected, both the 2995 children and the 1031 math items appear close to normally distributed. The children were from 9 grades, so the spread of 7 logits across the examinees could be right.

A question arose, however, when I went back and inspected the linking design. Children in the lower and higher grades had been deliberately over-sampled in order to get good child measures and item calibrations at the extremes. Yet this bias towards the extremes does not appear in Figure 1!

After eliminating other theories for this unexpected result, suspicion focussed on the analysis itself. Perhaps the familiar values for the convergence criteria were not stringent enough in this case. Accordingly, the criteria were made more stringent, and estimates were again obtained. The initial run used 50 iterations. The revised run, 263 iterations. The second outcome is shown in Figure 2. Now both the child and item distributions are clearly bimodal. The range of child abilities is about 9 logits, an increase of 2 logits. This result makes much better sense.

Establishing convergence is more than a statistical nicety. It can have profound substantive implications.

Ong Kim Lee

                       CHILDREN     MATH ITEMS
 5                             . +
                               . +
                               . +
 4                             . + #.
                             .## + #
                            .### + ####
 3                       .###### + ######.
                     .########## + ##############.
                   .############ + ##############.
 2              .############### + ##################.
       .######################## + ###############.
      .######################### + #################
 1   .########################## + #######################.
    .########################### + ######################.
   ############################# + ##########################
 0     .######################## + #########################
      .######################### + ######################
     .########################## + ####################.
-1        .##################### + ###############.
                .############### + ##################.
                   .############ + ###################.
-2                    .######### + ##################.
                          .##### + ############.
                            .### + ############
-3                            .# + #####.
                               . + ####
                                 + ##.
-4                               + #.
                                 + #
                                 + .
-5                               +
                                 +
                                 +
-6                               + .
                       CHILDREN  +  MATH ITEMS

Figure 1. Statistically converged estimates.

                         PERSONS +   ITEMS
 5                           ### + ###
                            .### + ##
                          .##### + #########
 4                   .########## + ################.
                     .########## + #################.
                 .############## + #####################.
 3             ################# + ######################
          .##################### + ################
          .##################### + ##################.
 2         .#################### + #####################
         ####################### + ######################
              .################# + ###################
 1              .############### + ################.
                .############### + #############.
                    .########### + #####################.
 0          .################### + ######################.
          .##################### + ###################.
        .####################### + #####################.
-1      .####################### + #######################
         .###################### + ###################
      .######################### + ########################
-2       .###################### + #######################
               .################ + #######################.
                  .############# + ####################.
-3                   .########## + ##########################
                        .####### + ###############.
                          .##### + ##############
-4                            .# + ######
                               # + ########
                                 + ###
-5                               + #.
                                 + ###
                                 + #
-6                               +
                         PERSONS +   ITEMS

Figure 2. Substantively converged estimates.

Convergence: Statistics or Substance?, O K Lee … Rasch Measurement Transactions, 1991, 5:3 p. 172

Rasch Books and Publications
Invariant Measurement: Using Rasch Models in the Social, Behavioral, and Health Sciences, 2nd Edn. George Engelhard, Jr. & Jue Wang	Applying the Rasch Model (Winsteps, Facets) 4th Ed., Bond, Yan, Heene	Advances in Rasch Analyses in the Human Sciences (Winsteps, Facets) 1st Ed., Boone, Staver	Advances in Applications of Rasch Measurement in Science Education, X. Liu & W. J. Boone	Rasch Analysis in the Human Sciences (Winsteps) Boone, Staver, Yale
Introduction to Many-Facet Rasch Measurement (Facets), Thomas Eckes	Statistical Analyses for Language Testers (Facets), Rita Green	Invariant Measurement with Raters and Rating Scales: Rasch Models for Rater-Mediated Assessments (Facets), George Engelhard, Jr. & Stefanie Wind	Aplicação do Modelo de Rasch (Português), de Bond, Trevor G., Fox, Christine M	Appliquer le modèle de Rasch: Défis et pistes de solution (Winsteps) E. Dionne, S. Béland
Exploring Rating Scale Functioning for Survey Research (R, Facets), Stefanie Wind	Rasch Measurement: Applications, Khine	Winsteps Tutorials - free Facets Tutorials - free	Many-Facet Rasch Measurement (Facets) - free, J.M. Linacre	Fairness, Justice and Language Assessment (Winsteps, Facets), McNamara, Knoch, Fan
Other Rasch-Related Resources: Rasch Measurement YouTube Channel
Rasch Measurement Transactions & Rasch Measurement research papers - free	An Introduction to the Rasch Model with Examples in R (eRm, etc.), Debelak, Strobl, Zeigenfuse	Rasch Measurement Theory Analysis in R, Wind, Hua	Applying the Rasch Model in Social Sciences Using R, Lamprianou	El modelo métrico de Rasch: Fundamentación, implementación e interpretación de la medida en ciencias sociales (Spanish Edition), Manuel González-Montesinos M.
Rasch Models: Foundations, Recent Developments, and Applications, Fischer & Molenaar	Probabilistic Models for Some Intelligence and Attainment Tests, Georg Rasch	Rasch Models for Measurement, David Andrich	Constructing Measures, Mark Wilson	Best Test Design - free, Wright & Stone Rating Scale Analysis - free, Wright & Masters
Virtual Standard Setting: Setting Cut Scores, Charalambos Kollias	Diseño de Mejores Pruebas - free, Spanish Best Test Design	A Course in Rasch Measurement Theory, Andrich, Marais	Rasch Models in Health, Christensen, Kreiner, Mesba	Multivariate and Mixture Distribution Rasch Models, von Davier, Carstensen

Forum

Rasch Measurement Forum to discuss any Rasch-related topic

Go to Top of Page
Go to index of all Rasch Measurement Transactions
AERA members: Join the Rasch Measurement SIG and receive the printed version of RMT
Some back issues of RMT are available as bound volumes
Subscribe to Journal of Applied Measurement

Go to Institute for Objective Measurement Home Page. The Rasch Measurement SIG (AERA) thanks the Institute for Objective Measurement for inviting the publication of Rasch Measurement Transactions on the Institute's website, www.rasch.org.

The URL of this page is www.rasch.org/rmt/rmt53l.htm

Website: www.rasch.org/rmt/contents.htm