ltm : An R Package for Latent Variable Modeling and Item Response Theory Analyses

Journal of statistical software (Impact Factor: 3.8). 11/2006; 17(5). DOI: 10.18637/jss.v017.i05


The R package ltm has been developed for the analysis of multivariate dichotomous and polytomous data using latent variable models, under the Item Response Theory ap-proach. For dichotomous data the Rasch, the Two-Parameter Logistic, and Birnbaum's Three-Parameter models have been implemented, whereas for polytomous data Seme-jima's Graded Response model is available. Parameter estimates are obtained under marginal maximum likelihood using the Gauss-Hermite quadrature rule. The capabilities and features of the package are illustrated using two real data examples.

  • Source
    • "Questionnaire items of externalizing problems were analyzed with graded response models in IRT (Samejima, 1969) using the ltm package (Rizopoulos, 2006) in R 3.0 (R "
    [Show abstract] [Hide abstract]
    ABSTRACT: The present study applied item response theory to identify an efficient set of items of the Achenbach Externalizing scale from the Child Behavior Checklist (CBCL; 33 items) and Teacher's Report Form (TRF; 35 items) that were sensitive to clinical-range scores. Mothers and teachers rated children's externalizing problems annually from ages 5 to 13 years in 2 independent samples (Ns = 585 and 1,199). Item properties for each rater across ages 5-8 and 9-13 were examined with item response theory. We identified 10 mother- and teacher-reported items from both samples based on the items' measurement precision for subclinical and clinical levels of externalizing problems: externalizing problems that involve meanness to others, destroying others' things, fighting, lying and cheating, attacking people, screaming, swearing/obscene language, temper tantrums, threatening people, and being loud. Scores on the scales using these items had strong reliability and psychometric properties, capturing nearly as much information as the full Externalizing scale for classifying clinical levels of externalizing problems. Scores on the scale with the 10 CBCL items had moderate accuracy, equivalent to the full Externalizing scale, in classifying diagnoses of conduct disorder based on a research diagnostic interview. Of course, comprehensive clinical assessment would consider additional items, dimensions of behavior, and sources of information, too, but it appears that the behaviors tapped by this select set of items may be core to externalizing psychopathology in children. (PsycINFO Database Record (c) 2015 APA, all rights reserved).
    Full-text · Article · Aug 2015 · Psychological Assessment
  • Source
    • "The fixed amount forced the decision makers to allocate the points to the investment categories they believed to be the most ben - eficial contingent upon their decision scenarios . To derive sensible scores from the present scales ' ordinal items , a gener - alised partial credit model ( Muraki 1992 ) was estimated using the package ' ltm ' in R ( Rizopoulos 2006 ) . Managers ' scores were computed using expected a posteriori ( EAP ) estimates and subsequently rescaled to an interval from 0 to 10 to ease interpretation ( for the transformation , extreme values were determined by EAP estimates of observations scoring either in the lowest or highest categories for all items ) . "
    [Show abstract] [Hide abstract]
    ABSTRACT: Green investments are crucial mechanisms for translating green operation strategies into managerial action. We examine the impact of external pressures on green investment patterns in terms of their scope, type and time horizon across 251 German and US managers. A scenario-based experiment was conducted using a 2 × 2 × 2 factorial design in which managers were assigned to high and low consumer, community, and resource treatment groups before being asked to make green investment decisions. Our hypotheses are developed based on resource advantage theory and tested in a number of regression models. The results demonstrate that German and US managers respond differently to external pressures in their green investment decisions. Regarding the scope of green investments, German and US managers invest differently if end consumer pressure increases and partially differently if resource scarcity increases, but they act in a similar way if community pressure increases. Moreover, we detected specific variations in the type and time horizon of green investments across US and German managers. The theoretical and practical implications of these findings for green operations management research and for firms operating in multinational settings are explained.
    Full-text · Article · Jun 2015 · International Journal of Production Research
    • "For instance, I relied on user-written code for interpretation in Stata (Ho, 2014), but Mplus has already added several traditional IRT graphics to their built-in commands. Other software packages are also adding macros and built-in commands for implementing IRT models using various estimation approaches, and readers are encouraged to check their favorite software for the latest developments (e.g., in SAS, the experimental PROC IRT [SAS Institute, 2013], as well as macros for PROC GLIMMIX [Black & Butler, 2012] and PROC NLMIXED [Sheu, Chen, Su, & Wang, 2005]; in SPSS, extension commands based on R [IBM, 2014]; in R, various user-written codes, including Chalmers, 2012, and Rizopoulos, 2006). "
    [Show abstract] [Hide abstract]
    ABSTRACT: This article provides family scientists with an understanding of contemporary measurement perspectives and the ways in which item response theory (IRT) can be used to develop measures with desired evidence of precision and validity for research uses. The article offers a nontechnical introduction to some key features of IRT, including its orientation toward locating items along an underlying dimension and toward estimating precision of measurement for persons with different levels of that same construct. It also offers a didactic example of how the approach can be used to refine conceptualization and operationalization of constructs in the family sciences, using data from the National Longitudinal Survey of Youth 1979 (n = 2,732). Three basic models are considered: (a) the Rasch and (b) two-parameter logistic models for dichotomous items and (c) the Rating Scale Model for multicategory items. Throughout, the author highlights the potential for researchers to elevate measurement to a level on par with theorizing and testing about relationships among constructs.
    No preview · Article · Feb 2015 · Journal of Marriage and Family
Show more