A Bayesian Beta-Mixture Model for Nonparametric IRT (BBM-IRT)
In: Arenson, E., and Karabatsos, G. (2018). A Bayesian beta-mixture model for nonparametric IRT (BBM-IRT). Journal of Modern Applied Statistical Methods, 1.
In: Arenson, E., and Karabatsos, G. (2018). A Bayesian beta-mixture model for nonparametric IRT (BBM-IRT). Journal of Modern Applied Statistical Methods, 1.
SSRN
In: Studies in educational evaluation: SEE, Band 35, Heft 2/3, S. 57-63
ISSN: 0191-491X
In: Studies in educational evaluation: SEE, Band 35, Heft 2-3
ISSN: 0191-491X
In: Studies in educational evaluation, Band 35, Heft 2-3, S. 57-63
ISSN: 0191-491X
In: Political analysis: PA ; the official journal of the Society for Political Methodology and the Political Methodology Section of the American Political Science Association, Band 26, Heft 4, S. 431-456
ISSN: 1476-4989
Data sets quantifying phenomena of social-scientific interest often use multiple experts to code latent concepts. While it remains standard practice to report the average score across experts, experts likely vary in both their expertise and their interpretation of question scales. As a result, the mean may be an inaccurate statistic. Item-response theory (IRT) models provide an intuitive method for taking these forms of expert disagreement into account when aggregating ordinal ratings produced by experts, but they have rarely been applied to cross-national expert-coded panel data. We investigate the utility of IRT models for aggregating expert-coded data by comparing the performance of various IRT models to the standard practice of reporting average expert codes, using both data from the V-Dem data set and ecologically motivated simulated data. We find that IRT approaches outperform simple averages when experts vary in reliability and exhibit differential item functioning (DIF). IRT models are also generally robust even in the absence of simulated DIF or varying expert reliability. Our findings suggest that producers of cross-national data sets should adopt IRT techniques to aggregate expert-coded data measuring latent concepts.
In: Openbaar bestuur: tijdschrift voor beleid, organisatie en politiek, Band 6, Heft 1, S. 22
ISSN: 0925-7322
SSRN
Working paper
In: Australasian marketing journal: AMJ ; official journal of the Australia-New Zealand Marketing Academy (ANZMAC), Band 21, Heft 2, S. 111-118
Der vorliegende Artikel skizziert zunächst kurz die Zielsetzung des DFG-Projekts . Im darauf folgenden Abschnitt werden zwei wesentliche Charakteristika zur Unterscheidung von MIRT-Modellen sowie deren inhaltliche Implikationen fuhr die Modellierung von Kompetenzen dargestellt. Die zentrale Fragestellung des DFG-Projekts … ist, inwieweit sich sprachliche Kompetenzen, die mit demselben Aufgabenmaterial erfasst wurden, auf unterschiedliche Weise modellieren lassen und welche diagnostischen Aussagen aus den verschiedenen Auswertungen resultieren. Insbesondere interessiert, inwieweit komplexere mehrdimensionale Modelle gegenüber eindimensionalen Modellen (z.B. dem eindimensionalen Raschmodell) einen Zugewinn an diagnostischer Information liefern. (DIPF/Orig.)
BASE
In: Argumenta Oeconomica, Band 44, Heft 1, S. 207-226
Employment is at the heart of European Union (EU) policies as it is the basis for wealth creation. Knowing how satisfied EU residents are with their occupation is very important, since losing one's job may undermine one's life satisfaction and its overall meaning (European Commission 2015). According to the most recent Eurostat data (European Commission 2017), Poland reported an average job satisfaction well above the EU mean, ranked 8th (behind Denmark, Iceland, Austria, Finland, Norway, Switzerland and Sweden). Thus, it is interesting to present an analysis focused on the job satisfaction of workers in Poland – a country of emigration, with the highest percentage of temporary contracts in Europe (European Commission 2016). The main aim of our study is understanding how the different socio-economic features affect the groups of workers with similar job satisfaction levels in Poland. Most of the Polish job satisfaction studies are focused on selected professional groups, in selected regions of Poland. This article presents another, the latent variable models approach to the hetero-geneous data set for different subgroups of workers in all the regions of Poland. The combination of the two latent variable models enables to find homogeneous classes of individuals characterized by the similar latent ability levels, and at the same time, the item characteristics analysis (usually identified as discrimination indices and difficulty parameters) as well. Latent Class Item Response Theory (LC-IRT) models are more flexible in compa-rison with traditional formulations of Item Response Theory (IRT) models, often based on restrictive assumptions, such as normality of latent trait (explicitly introduced). Moreover, the authors also apply the extended latent variable models under the discrete assumption of the latent trait including individual socio-demographic features, such as age, sex, education, marital status or current financial situation. The article analyzes data collected as part of the International Social Survey Programme 2015 using R software. The results may help policymakers tailor their employment policies as well as to create and deliver services focused on special socio-economic groups of the Polish society.
In: EUVocS (EU-Vocabularium Savariense)
Small, Micro and Medium Enterprises (MSMEs) face quite complex problems in business development. One of the small businesses engaged in by the Sudagaran community in Yogyakarta is the production of tofu. Entrepreneurs know problems in the fields of production, environment, capital, and also marketing. The Service Team from the Department of Business Administration, FISIP UPN Veterans Yogyakarta collaborated with the Yogyakarta City Baznas and the Yogyakarta City Government to assist and assist SMEs. Empowerment programs that have been carried out have started from the implementation of production layouts, small business financial management, simple financial records, product innovation. packaging, and online-based marketing. Small entrepreneurs feel helped by the service program so that during the Covid-19 pandemic they have alternatives in product marketing.
BASE