Detecting Multidimensional DIF in Polytomous Items with IRT Methods and Estimation Approaches

The purpose of this study was to investigate multidimensional DIF with a simple and nonsimple structure in the context of multidimensional Graded Response Model (MGRM). This study examined and compared the performance of the IRT-LR and Wald test using MML-EM and MHRM estimation approaches with different test factors and test structures in simulation studies and applying real data sets. When the test structure included two dimensions, the IRT-LR (MML-EM) generally performed better than the Wald test and provided higher power rates. If the test included three dimensions, the methods provided similar performance in DIF detection. In contrast to these results, when the number of dimensions in the test was four, MML-EM estimation completely lost precision in estimating the nonuniform DIF, even with large sample sizes. The Wald with MHRM estimation approaches outperformed the Wald test (MML-EM) and IRT-LR (MML-EM). The Wald test had higher power rate and acceptable type I error rates for nonuniform DIF with the MHRM estimation approach.The small and/or unbalanced sample sizes, small DIF magnitudes, unequal ability distributions between groups, number of dimensions, estimation methods and test structure were evaluated as important test factors for detecting multidimensional DIF.     

A Comparison of Estimation Techniques for IRT Models With Small Samples

The usefulness of item response theory (IRT) models depends, in large part, on the accuracy of item and person parameter estimates. For the standard 3 parameter logistic model, for example, these parameters include the item parameters of difficulty, discrimination, and pseudo-chance, as well as the person ability parameter. Several factors impact traditional marginal maximum likelihood (ML) estimation of IRT model parameters, including sample size, with smaller samples generally being associated with lower parameter estimation accuracy, and inflated standard errors for the estimates. Given this deleterious impact of small samples on IRT model performance, use of these techniques with low-incidence populations, where it might prove to be particularly useful, estimation becomes difficult, especially with more complex models. Recently, a Pairwise estimation method for Rasch model parameters has been suggested for use with missing data, and may also hold promise for parameter estimation with small samples. This simulation study compared item difficulty parameter estimation accuracy of ML with the Pairwise approach to ascertain the benefits of this latter method. The results support the use of the Pairwise method with small samples, particularly for obtaining item location estimates.     

Relating Unidimensional IRT Parameters to a Multidimensional Response Space: A Review of Two Alternative Projection IRT Models for Scoring Subscales

A practical concern for many existing tests is that subscore test lengths are too short to provide reliable and meaningful measurement. A possible method of improving the subscale reliability and validity would be to make use of collateral information provided by items from other subscales of the same test. To this end, the purpose of this article is to compare two different formulations of an alternative Item Response Theory (IRT) model developed to parameterize unidimensional projections of multidimensional test items: Analytical and Empirical formulations. Two real data applications are provided to illustrate how the projection IRT model can be used in practice, as well as to further examine how ability estimates from the projection IRT model compare to external examinee measures. The results suggest that collateral information extracted by a projection IRT model can be used to improve reliability and validity of subscale scores, which in turn can be used to provide diagnostic information about strength and weaknesses of examinees helping stakeholders to link instruction or curriculum to assessment results.     

Detecting Potentially Biased Test Items: Comparison of IRT Area and Mantel-Haenszel Methods

The purpose of this study was to compare the IRT-based area method and the Mantel-Haenszel method for investigating differential item functioning (DIF), to determine the degree of agreement between the methods in identifying potentially biased items, and, when the two methods led to different results, to identify possible reasons for the discrepancies. Data for the study were the item responses of Anglo American and Native American students who took the 1982 New Mexico High School Proficiency Exam. Two samples of 1,000 students from each group were studied. The major findings were that (a) the consistency of classifications of items into "biased" and "not-biased" categories across replications was 75% to 80% for both methods and (b) when the unreliability of the statistics was taken into account, the two methods led to very similar results. Discrepancies between methods were due to the presence of nonuniform DIF (the Mantel-Haenszel method could not identify these items) and the choice of interval over which DIF was assessed (the IRT method results depended on the choice of interval). The implications for practitioners seem clear: The Mantel-Haenszel method in general provides an acceptable approximation to the IRT-based methods.     

An Empirical Investigation Demonstrating the Multidimensional DIF Paradigm: A Cognitive Explanation for DIF

Differential Item Functioning (DIF) is traditionally used to identify different item performance patterns between intact groups, most commonly involving race or sex comparisons. This study advocates expanding the utility of DIF as a step in construct validation. Rather than grouping examinees based on cultural differences, the reference and focal groups are chosen from two extremes along a distinct cognitive dimension that is hypothesized to supplement the dominant latent trait being measured. Specifically, this study investigates DIF between proficient and non-proficient fourth- and seventh-grade writers on open-ended mathematics test items that require students to communicate about mathematics. It is suggested that the occurrence of DIF in this situation actually enhances, rather than detracts from, the construct validity of the test because, according to the National Council of Teachers of Mathematics (NCTM), mathematical communication is an important component of mathematical ability, the dominant construct being assessed. However, the presence of DIF influences the validity of inferences that can be made from test scores and suggests that two scores should be reported, one for general mathematical ability and one for mathematical communication. The fact that currently only one test score is reported, a simple composite of scores on multiple-choice and open-ended items, may lead to incorrect decisions being made about examinees.     

A Comparison of Item Fit Statistics for Mixed IRT Models

In this study we examined procedures for assessing model-data fit of item response theory (IRT) models for mixed format data. The model fit indices used in this study include PARSCALE's G² , Orlando and Thissen's S − X² and S − G² , and Stone's χ²* and G²* . To investigate the relative performance of the fit statistics at the item level, we conducted two simulation studies: Type I error and power studies. We evaluated the performance of the item fit indices for various conditions of test length, sample size, and IRT models. Among the competing measures, the summed score-based indices S − X² and S − G² were found to be the sensible and efficient choice for assessing model fit for mixed format data. These indices performed well, particularly with short tests. The pseudo-observed score indices, χ²* and G²* , showed inflated Type I error rates in some simulation conditions. Consistent with the findings of current literature, the PARSCALE's G² index was rarely useful, although it provided reasonable results for long tests.     

A Comparison of Alternative Matching Strategies for DIF Detection in Tests That Are Multidimensional

Most currently accepted approaches for identifying differentially functioning test items compare performance across groups after first matching examinees on the ability of interest. The typical basis for this matching is the total test score. Previous research indicates that when the test is not approximately unidimensional, matching using the total test score may result in an inflated Type I error rate. This study compares the results of differential item functioning (DIF) analysis with matching based on the total test score, matching based on subtest scores, or multivariate matching using multiple subtest scores. Analysis of both actual and simulated data indicate that for the dimensionally complex test examined in this study, using the total test score as the matching criterion is inappropriate. The results suggest that matching on multiple subtest scores simultaneously may be superior to using either the total test score or individual relevant subtest scores.     

基于IRT的一种测验等值问题的参数估计方法

基于项目反应理论,文章介绍了测验等值问题的意义和模型,然后分析了测验等值的原理,并采用最小二乘估计法对其中涉及到的转换系数进行了参数估计,真正实现了项目反应理论中的项目参数等值和真分数等值.     

A Comparison of Two Procedures for Computing IRT Equating Coefficients

In order to equate tests under Item Response Theory (IRT), one must obtain the slope and intercept coefficients of the appropriate linear transformation. This article compares two methods for computing such equating coefficients–Loyd and Hoover (1980) and Stocking and Lord (1983). The former is based upon summary statistics of the test calibrations; the latter is based upon matching test characteristic curves by minimizing a quadratic loss function. Three types of equating situations: horizontal, vertical, and that inherent in IRT parameter recovery studies–were investigated. The results showed that the two computing procedures generally yielded similar equating coefficients in all three situations. In addition, two sets of SAT data were equated via the two procedures, and little difference in the obtained results was observed. Overall, the results suggest that the Loyd and Hoover procedure usually yields acceptable equating coefficients. The Stocking and Lord procedure improves upon the Loyd and Hoover values and appears to be less sensitive to atypical test characteristics. When the user has reason to suspect that the test calibrations may be associated with data sets that are typically troublesome to calibrate, the Stocking and Lord procedure is to be preferred.     

Factor Models for Ordinal Variables With Covariate Effects on the Manifest and Latent Variables: A Comparison of LISREL and IRT Approaches

We consider a general type of model for analyzing ordinal variables with covariate effects and 2 approaches for analyzing data for such models, the item response theory (IRT) approach and the PRELIS-LISREL (PLA) approach. We compare these 2 approaches on the basis of 2 examples, 1 involving only covariate effects directly on the ordinal variables and 1 involving covariate effects on the latent variables in addition.     

A Nonparametric Approach for Assessing Goodness-of-Fit of IRT Models in a Mixed Format Test

Investigating the fit of a parametric model plays a vital role in validating an item response theory (IRT) model. An area that has received little attention is the assessment of multiple IRT models used in a mixed-format test. The present study extends the nonparametric approach, proposed by Douglas and Cohen (2001), to assess model fit of three IRT models (three- and two-parameter logistic model, and generalized partial credit model) used in a mixed-format test. The statistical properties of the proposed fit statistic were examined and compared to S-X² and PARSCALE’s G². Overall, RISE (Root Integrated Square Error) outperformed the other two fit statistics under the studied conditions in that the Type I error rate was not inflated and the power was acceptable. A further advantage of the nonparametric approach is that it provides a convenient graphical inspection of the misfit.     

Evaluating the Wald Test for Item‐Level Comparison of Saturated and Reduced Models in Cognitive Diagnosis

This article used the Wald test to evaluate the item‐level fit of a saturated cognitive diagnosis model (CDM) relative to the fits of the reduced models it subsumes. A simulation study was carried out to examine the Type I error and power of the Wald test in the context of the G‐DINA model. Results show that when the sample size is small and a larger number of attributes are required, the Type I error rate of the Wald test for the DINA and DINO models can be higher than the nominal significance levels, while the Type I error rate of the A‐CDM is closer to the nominal significance levels. However, with larger sample sizes, the Type I error rates for the three models are closer to the nominal significance levels. In addition, the Wald test has excellent statistical power to detect when the true underlying model is none of the reduced models examined even for relatively small sample sizes. The performance of the Wald test was also examined with real data. With an increasing number of CDMs from which to choose, this article provides an important contribution toward advancing the use of CDMs in practical educational settings.     

A Comparison of IRT Proficiency Estimation Methods Under Adaptive Multistage Testing

This inquiry is an investigation of item response theory (IRT) proficiency estimators’ accuracy under multistage testing (MST). We chose a two‐stage MST design that includes four modules (one at Stage 1, three at Stage 2) and three difficulty paths (low, middle, high). We assembled various two‐stage MST panels (i.e., forms) by manipulating two assembly conditions in each module, such as difficulty level and module length. For each panel, we investigated the accuracy of examinees’ proficiency levels derived from seven IRT proficiency estimators. The choice of Bayesian (prior) versus non‐Bayesian (no prior) estimators was of more practical significance than the choice of number‐correct versus item‐pattern scoring estimators. The Bayesian estimators were slightly more efficient than the non‐Bayesian estimators, resulting in smaller overall error. Possible score changes caused by the use of different proficiency estimators would be nonnegligible, particularly for low‐ and high‐performing examinees.     

Differential Item Functioning Assessment in Cognitive Diagnostic Modeling: Application of the Wald Test to Investigate DIF in the DINA Model

Analyzing examinees’ responses using cognitive diagnostic models (CDMs) has the advantage of providing diagnostic information. To ensure the validity of the results from these models, differential item functioning (DIF) in CDMs needs to be investigated. In this article, the Wald test is proposed to examine DIF in the context of CDMs. This study explored the effectiveness of the Wald test in detecting both uniform and nonuniform DIF in the DINA model through a simulation study. Results of this study suggest that for relatively discriminating items, the Wald test had Type I error rates close to the nominal level. Moreover, its viability was underscored by the medium to high power rates for most investigated DIF types when DIF size was large. Furthermore, the performance of the Wald test in detecting uniform DIF was compared to that of the traditional Mantel‐Haenszel (MH) and SIBTEST procedures. The results of the comparison study showed that the Wald test was comparable to or outperformed the MH and SIBTEST procedures. Finally, the strengths and limitations of the proposed method and suggestions for future studies are discussed.     

Bayesian Model Selection Methods for Multilevel IRT Models: A Comparison of Five DIC‐Based Indices

Model selection is important in any statistical analysis, and the primary goal is to find the preferred (or most parsimonious) model, based on certain criteria, from a set of candidate models given data. Several recent publications have employed the deviance information criterion (DIC) to do model selection among different forms of multilevel item response theory models (MLIRT). The majority of the practitioners use WinBUGS for implementing MCMC algorithms for MLIRT models, and the default version of DIC provided by WinBUGS focused on the measurement‐level parameters only. The results herein show that this version of DIC is inappropriate. This study introduces five variants of DIC as a model selection index for MLIRT models with dichotomous outcomes. Considering a multilevel IRT model with three levels, five forms of DIC are formed: first‐level conditional DIC computed from the measurement model only, which is the index given by many software packages such as WinBUGS; second‐level marginalized DIC and second‐level joint DIC computed from the second‐level model; and top‐level marginalized DIC and top‐level joint DIC computed from the entire model. We evaluate the performance of the five model selection indices via simulation studies. The manipulated factors include the number of groups, the number of second‐level covariates, the number of top‐level covariates, and the types of measurement models (one‐parameter vs. two‐parameter). Considering the computational viability and interpretability, the second‐level joint DIC is recommended for MLIRT models under our simulated conditions.     

Monte Carlo Based Null Distribution for an Alternative Goodness-of-Fit Test Statistic in IRT Models

Assessing the correspondence between model predictions and observed data is a recommended procedure for justifying the application of an IRT model. However, with shorter tests, current goodness-of-fit procedures that assume precise point estimates of ability, are inappropriate. The present paper describes a goodness-of-fit statistic that considers the imprecision with which ability is estimated and involves constructing item fit tables based on each examinee's posterior distribution of ability, given the likelihood of their response pattern and an assumed marginal ability distribution. However, the posterior expectations that are computed are dependent and the distribution of the goodness-of-fit statistic is unknown. The present paper also describes a Monte Carlo resampling procedure that can be used to assess the significance of the fit statistic and compares this method with a previously used method. The results indicate that the method described herein is an effective and reasonably simple procedure for assessing the validity of applying IRT models when ability estimates are imprecise.     

A Comparison of Approaches for Estimating Covariate Effects in Nonparametric Multilevel Latent Class Models

The inclusion of covariates improves the prediction of class memberships in latent class analysis (LCA). Several methods for examining covariate effects have been developed over the past decade; however, researchers have limited to the comparisons of the performance among these methods in cases of the single-level LCA. The present study investigated the performance of three different methods for examining covariate effects in a multilevel setting. We conducted a simulation to compare the performance of the three methods when level-1 and level-2 covariates were simultaneously incorporated into the nonparametric multilevel latent class model to predict latent class membership at each level. The simulation results revealed that the bias-adjusted three-step maximum likelihood method performed equally well as the one-step method when the sample sizes were sufficiently large and the latent classes were distinct from each other. However, the unadjusted three-step method significantly underestimated the level-1 covariate effect in most conditions.

     

A Multilevel Mixture IRT Framework for Modeling Response Times as Predictors or Indicators of Response Engagement in IRT Models

Disengaged item responses pose a threat to the validity of the results provided by large-scale assessments. Several procedures for identifying disengaged responses on the basis of observed response times have been suggested, and item response theory (IRT) models for response engagement have been proposed. We outline that response time-based procedures for classifying response engagement and IRT models for response engagement are based on common ideas, and we propose the distinction between independent and dependent latent class IRT models. In all IRT models considered, response engagement is represented by an item-level latent class variable, but the models assume that response times either reflect or predict engagement. We summarize existing IRT models that belong to each group and extend them to increase their flexibility. Furthermore, we propose a flexible multilevel mixture IRT framework in which all IRT models can be estimated by means of marginal maximum likelihood. The framework is based on the widespread Mplus software, thereby making the procedure accessible to a broad audience. The procedures are illustrated on the basis of publicly available large-scale data. Our results show that the different IRT models for response engagement provided slightly different adjustments of item parameters of individuals’ proficiency estimates relative to a conventional IRT model.     

Development of Automated Scoring Algorithms for Complex Performance Assessments: A Comparison of Two Approaches

Performance assessments are typically scored by having experts rate individual performances. The cost associated with using expert raters may represent a serious limitation in many large-scale testing programs. The use of raters may also introduce an additional source of error into the assessment. These limitations have motivated development of automated scoring systems for performance assessments. Preliminary research has shown these systems to have application across a variety of tasks ranging from simple mathematics to architectural problem solving. This study extends research on automated scoring by comparing alternative automated systems for scoring a computer simulation test of physicians'patient management skills; one system uses regression-derived weights for components of the performance, the other uses complex rules to map performances into score levels. The procedures are evaluated by comparing the resulting scores to expert ratings of the same performances.