1,721,110 research outputs found
A classification of university courses based on students’ satisfaction: an application of a two-level mixture item response model
No full textOver the past years, Italian universities have come under increased pressure to be more competitive and attract more students, and students’ satisfaction has received increasing attention. Students’ opinions about a few aspects of academic life are sought by Italian universities in the form of a satisfaction feedback questionnaire. The aim of this paper is to classify university courses into homogeneous classes with respect to the level of students’ satisfaction through the use of a two-level mixture item response model. The data are drawn from the Italian questionnaire on students’ satisfaction administered at a Faculty of Political Sciences. The latent variables measured by the questionnaire are detected performing a model-based hierarchical clustering. Then, a special case of multilevel mixture factor model characterised by an item response parameterisation and discrete latent variables at all hierarchical levels is estimated. The study allowed us to ascertain (i) the latent dimensionality of students’ satisfac- tion with higher education courses; (ii) the varied effect of first and second-level covariates on the satisfaction dimensions; and (iii) the different sources of strength/weaknesses of the best and worst courses
An approach for the assessment of the Health-related Quality of Life based on multidimensional latent class Rasch models
No full text
Two-Tier Latent Class IRT Models in R
No full textIn analyzing data deriving from the administration of a questionnaire to a group of individu als, Item Response Theory (IRT) models provide a flexible framework to account for several aspects involved in the response process, such as the existence of multiple latent traits. In this paper, we focus on a class of semi-parametric multidimensional IRT models, in which these traits are represented through one or more discrete latent variables; these models allow us to cluster individuals into homo geneous latent classes and, at the same time, to properly study item characteristics. In particular, we follow a within-item multidimensional formulation similar to that adopted in the two-tier models, with each item measuring one or two latent traits. The proposed class of models may be estimated through the package MLCIRTwithin, whose functioning is illustrated in this paper with examples based on data about quality-of-life measurement and about the propensity to commit a crime
Multilevel IRT models for the university teaching evaluation
No full textIn this paper, a generalization of the two-parameter partial credit model (2PL-PCM) and of two special cases, the partial credit model (PCM) and the rating scale model (RSM), with a hierarchical data structure will be presented. Having shown how 2PL-PCM, as with other item response theory (IRT) models, may be read in terms of a generalized linear mixed model (GLMM) with two aggregation levels, a presentation will be given of an extension to the case of measuring the latent trait of individuals aggregated in groups. The use of this Multilevel IRT model will be illustrated via reference to the evaluation of university teaching by students following the courses. The aim is to generate a ranking of teaching on the basis of student satisfaction, so as to give teachers, and those responsible for organizing study courses, a background of information that takes the opinions of the direct target group for university teaching (that is, the students) into account, in the context of improving the teaching courses available
Joint assessment of the latent trait dimensionality and observed differential item functioning of students’ national tests
No full textStudents’ assessment tests are routinely validated through item response theory (IRT) models which assume unidimensionality and absence of observable differential item functioning (DIF). In this paper, we investigate if such assumptions hold for two national tests administered in Italy to lower secondary school students: the Language Test and the Mathematics Test. To this aim, we rely on an extended class of multidimensional latent class IRT models characterised by: (i) a two-parameter logistic parameterisation for the conditional probability of a correct response, (ii) latent traits represented through a random vector with a discrete distribution, and (iii) the inclusion of (uniform) DIF to account for students’ gender and geographical area. A classification of the items into unidimensional groups is also proposed and represented by a dendrogram, which is obtained from a hierarchical clustering algorithm. The results provide evidence for observable DIF effects for both tests. Besides, the assumption of unidimensionality is rejected for the Language Test, whereas it is reasonable for the Mathematics Test
Mixtures of equispaced Normal distributions and their use for testing symmetry in univariate data
No full text
Decisioni razionali per il governo dell’Università, un prerequisito essenziale: la teoria dell’utilità
No full text- …
