Full-text resources of CEJSH and other databases are now available in the new Library of Science.
Visit https://bibliotekanauki.pl

Results found: 8

first rewind previous Page / 1 next fast forward last

Search results

Search:
in the keywords:  item response theory
help Sort By:

help Limit search:
first rewind previous Page / 1 next fast forward last
1

Comparison of Mantel–Haenszel test with IRT procedures for DIF detection and effect size estimation for dichotomous items

100%
Kondratek B., Grudniewska M.
Edukacja
|
2014
|
issue 5(130)
92–111
EN
The article compares two methods used to detect differential item functioning (DIF) of dichotomously scored items: a nonparametric solution based on the Mantel–Haenszel procedure (MH) and a parametric IRT approach with a likelihood ratio test. A Monte Carlo experiment was performed in order to evaluate performance of both statistics in various conditions of DIF uniformity. Results confirmed the theoretical prediction that the MH test has greater statistical power in detecting uniform DIF than the likelihood ratio test and less power than the LR test in cases of non-uniform DIF. Apart of examining statistical power of the test, specific measures of DIF effect size were compared: MH D–DIF and three measures of P–DIF expressed on the item easiness scale.
2
Content available remote

Adaptivní administrace NEO PI-R: výhody a omezení

100%
Jelínek M.
Československá psychologie (Czechoslovak Psychology)
|
2011
|
vol. 55
|
issue 1
69-81
EN
Adaptively administered tests with dichotomously scored items are already well described in the relevant literature and used in practice. The presented study analyses the possibilities of adaptive administration of test with polytomous items, which are commonly used in personality testing. Based on analysis of simulated adaptive administration of NEO PI-R, the limits and benefits of this approach are discussed. It was found that adaptive administration successfully and more effectively reconstructs the level of measured traits in comparison with full scale administration. On the other hand, significant problem consists in overexposure of several items with highest item discrimination power. Representative sample built for the purposes of Czech standardization of NEO PI-R was used (N = 2084).
3
Publication available in full text mode
Content available

Teoria reakcji na pozycję w podejściu modelowym w takso- nomii

100%
Genge E.
Econometrics. Ekonometria. Advances in Applied Data Analytics
|
2016
|
issue 1 (51)
9-19
EN
Item response theory is considered to be one of the two trends in methodological assessment of the reliability scale. In turn, latent class models can be viewed as a special case of model-based clustering, for heterogenous multivariate discrete data. We used the approach combining item response theory and latent class models to find groups of Polish households’ with similar saving ability levels. We analyzed data collected as part of the Polish Social Diagnosis using MultiLCIRT package of R.
4
Publication available in full text mode
Content available

Development of short form questionnaires for the assessment of work capacity in cardiovascular rehabilitation patients

88%
Haschke A., Abberger B., Wirtz M., Bengel J., Baumeister H.
International Journal of Occupational Medicine and Environmental Health
|
2013
|
vol. 26
|
issue 5
742-750
EN
Objectives: Prevention of job loss is an essential objective of cardiovascular rehabilitation. However, comprehensive and economic diagnostic instruments on work limitations are missing. The present study describes development of short form questionnaires from 2 domains of the WCIB-Cardio item banks for the assessment of work capacity in cardiovascular rehabilitation patients. Materials and Methods: 283 cardiovascular rehabilitation patients were recruited from 14 German rehabilitation clinics. Based on the WCIB-Cardio with the domains of cognitive and physical work capacity, we developed a short form for both domains. Item selection criteria were content coverage, content appropriateness, internal consistency reliability (≥ 0.8). We used correlation of person location scores of the short forms with person location scores of the full item banks to examine the extent of measurement precision. Results: For each domain of the WCIB-Cardio a short form was developed (cognitive work capacity - 14 items; physical work capacity 7 - items). In both domains psychometric properties were good (person separation index: cognitive work capacity - 0.80; physical work capacity - 0.80). Correlation measures of the short form with the full item banks showed a high accordance of person locations for both domains (cognitive work capacity: r = 0.97; physical work capacity: r = 0.95). Conclusions: The calibrated instrument WCIB-Cardio provides the possibility to develop short form questionnaires with high psychometric quality. These short forms make it possible to monitor patient's work capacity in cardiovascular rehabilitation settings in a more economical way.
5

Test Mantel–Haenszel oraz modelowanie IRT jako narzędzia wykrywania DIF i opisu jego wielkości na przykładzie zadań ocenianych dychotomicznie

75%
Kondratek B., Grudniewska M.
Edukacja
|
2013
|
issue 2(122)
34–55
PL
Artykuł porównuje dwie metody wykorzystywane do identyfikacji zróżnicowanego funkcjonowania zadań (DIF) ocenianych dychotomicznie: nieparametryczne rozwiązanie opierające się na statystyce Mantela–Haenszela (MH) oraz podejście bazujące na teście ilorazu funkcji wiarygodności. Porównanie przeprowadzono na gruncie teoretycznym i za pomocą symulacji. Wyniki symulacji potwierdziły przypuszczenie, że podejście opierające się na statystyce MH jest bardziej czułe na jednorodne efekty DIF, jednak traci moc, gdy wielkość DIF zmienia się w zależności od poziomu zmiennej ukrytej mierzonej testem. Oprócz mocy statystycznej analizowano również specyficzne miary wielkości efektu DIF stosowane w obu metodach: miarę MH D – DIF, wykorzystywaną standardowo przez Educational Testing Service do klasyfikacji wielkości DIF, oraz różne miary P – DIF określone na metryce łatwości zadania.
EN
The article compares two methods used to detect differential item functioning (DIF) of dichotomously scored items: a nonparametric solution based on the Mantel–Haenszel procedure (MH) and a parametric IRT approach with a likelihood ratio test. A Monte Carlo experiment was performed in order to evaluate performance of both statistics in various conditions of DIF uniformity. Results confirmed the theoretical prediction that the MH test has greater statistical power in detecting uniform DIF than the likelihood ratio test and less power than the LR test in cases of non-uniform DIF. Apart of examining statistical power of the test, specific measures of DIF effect size were compared: MH D–DIF and three measures of P–DIF expressed on the item easiness scale.
6

Modele analizy efektu oceniającego w pomiarze edukacyjnym

71%
Kulon F.
Edukacja
|
2014
|
issue 3(128)
95–111
PL
Artykuł powstał w wyniku poszukiwań optymalnego modelu analizy w ramach prowadzonych badań porównywalności oceniania i efektu egzaminatora w zakresie egzaminu maturalnego z języka polskiego i matematyki. W części pierwszej przedstawiono krótko teorię dotyczącą zagadnienia efektu oceniającego (rater effect), odnosząc je do obszaru pomiaru edukacyjnego w Polsce, w którym otrzymało ono nazwę efektu egzaminatora. Skupiono się na zagadnieniu od strony pomiarowej i nie rozważano psychologicznych podstaw oceniania. W drugiej części artykułu przedstawiono wybrane modele analizy tego efektu i wskazano, który model pozwala na oszacowanie największej liczby różnych aspektów efektu egzaminatora. Opisane zostały również symulacje sprawdzające przydatność modelu HRM-SDT do analizy danych z polskiego egzaminu maturalnego.
EN
The article is the result of a search for an optimal model of data analysis in a study on scoring comparability and rater effect in upper secondary school leaving examination in Polish language and mathematics. The first part briefly outlines the theory on rater effect. Mainly the measurement aspect is presented, the psychological bases of the scoring process are not discussed. In the second part selected models of rater effect analysis are described. Amongst them, the hierarchical rater model with signal detection theory is considered as covering the broadest range of different types of rater effect. This model was used in simulations to check its usefulness for data analysis of upper secondary school leaving examination in Poland.
7
Content available remote

Morality stereotype content scale (MSCS) : Rasch analysis and evidence of validity

63%
Sayans-Jiménez P., Cuadrado Guirado I., Blanc Molina A., Ordóñez-Carrasco J., Rojas Tejada A.
Československá psychologie (Czechoslovak Psychology)
|
2018
|
vol. 62
|
issue 4
366-381
EN
The aim of this work was to develop a new scale, the Morality Stereotype Content Scale (MSCS), and to analyze its psychometric properties. The MSCS contains both positive and negative morality- specific attributes. MSCS utility stands out for its concision and its translation ease. The study was developed with the collaboration of a sample of 550 Spaniards (keeping with the composition, in age and gender, of the Spanish population). Stereotypes toward the Spanish Roma people were assessed. Polytomous Rasch model (rating scale model) was employed. The results showed MSCS good psychometric properties that support its use for monitoring of intergroup perceptions and the assessment of the effectiveness of such interventions aimed to reduce negative intergroup stereotypes, particularly in the Spanish and European context.
CS
Cílem bylo vyvinout novou škálu, Škálu obsahů mravních stereotypů (MSCS) a analyzovat její psychometrické vlastnosti. Škála obsahuje jak pozitivní, tak negativní morálně specifické atributy. Užitečnost škály je dána její koncizností a snadností jejího překladu. Validizační studie byla realizována s 550 Španěly při zachování poměrů složení španělské populace (zastoupení věku a pohlaví). Byly zjišťovány stereotypy vůči španělským Romům. Byl použit polytomický Raschův model (model posuzovací škály). Výsledky ukázaly dobré psychometrické vlastnosti škály, které ji doporučují pro monitorování meziskupinového vnímání a hodnocení efektivity intervencí zaměřených na redukci negativních meziskupinových stereotypů, zvláště ve španělském a evropském kontextu.
8

Paradygmaty marketingu a modelowanie zmiennych ukrytych – porównanie modeli pomiarowych skali WSAW

63%
Sagan A.
Zeszyty Naukowe Uniwersytetu Ekonomicznego w Krakowie
|
2013
|
issue 909
17-27
PL
Celem artykułu jest przedstawienie specyfiki modeli pomiarowych ze zmiennymi ukrytymi w ramach trzech podstawowych paradygmatów marketingu: poznawczego, związanego z transakcyjnym nurtem (paradygmatem) marketingu, behawioralnego, odnoszącego się do predykcyjnego nurtu CRM, oraz relacyjnego paradygmatu sieciowego. W ewolucji modelowania strukturalnego i jego zastosowań w badaniach marketingowych scharakteryzowano trzy podstawowe klasy modeli odnoszących się do wskazanych paradygmatów i rozwijanych w ostatnich dwóch dekadach: model czynnika wspólnego, model składowej ze wskaźnikami formatywnymi oraz model losowych wskaźników.
EN
The purpose of this article is to present the evolution of measurement models with latent variables in the framework of three basic marketing paradigms: cognitive, associated with “transactional” marketing, behavioural, which applies to predictive CRM, and the relational and network paradigm in marketing. In the evolution of structural modelling and its applications in marketing research, three basic classes of measurement models applying to the above paradigms have been developed over the last two decades: the common factor model, the component model with formative indicators and the random loadings model.
first rewind previous Page / 1 next fast forward last
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.