Detection of Differential Item Functioning with difNLR Package
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985807%3A_____%2F18%3A00492638" target="_blank" >RIV/67985807:_____/18:00492638 - isvavai.cz</a>
Výsledek na webu
—
DOI - Digital Object Identifier
—
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Detection of Differential Item Functioning with difNLR Package
Popis výsledku v původním jazyce
ZÁKLADNÍ ÚDAJE: IMPS 2018: Abstracts Book - Talks. New York City: Columbia University, 2018. s. 42-42. [IMPS 2018: International Meeting of Psychometric Society. 09.07.2018-13.07.2018, New York City]. GRANT CEP: GA ČR GJ15-15856Y. ABSTRAKT: The R package difNLR (Drabinová, Martinková & Zvára, 2018) has been developed for detection of Differential Item Functioning (DIF), based on extensions of logistic regression model. These include guessing and non‐attention parameters which can differ for different groups. For dichotomous data, eleven predefined models have been implemented, however, user can constraint some parameters to be the same for different groups and hence create wide range of models that can be seen as proxies for item response theory models. The difNLR package offers various methods for estimation of parameters and DIF detection procedure. It also covers procedures in DIF identification such as item purification or corrections for multiple comparisons. Moreover, simulation studies suggest good properties even in smaller samples (Drabinová & Martinková, 2017), and thus the family of models offered by the difNLR library seems to be promising in DIF detection.
Název v anglickém jazyce
Detection of Differential Item Functioning with difNLR Package
Popis výsledku anglicky
ZÁKLADNÍ ÚDAJE: IMPS 2018: Abstracts Book - Talks. New York City: Columbia University, 2018. s. 42-42. [IMPS 2018: International Meeting of Psychometric Society. 09.07.2018-13.07.2018, New York City]. GRANT CEP: GA ČR GJ15-15856Y. ABSTRAKT: The R package difNLR (Drabinová, Martinková & Zvára, 2018) has been developed for detection of Differential Item Functioning (DIF), based on extensions of logistic regression model. These include guessing and non‐attention parameters which can differ for different groups. For dichotomous data, eleven predefined models have been implemented, however, user can constraint some parameters to be the same for different groups and hence create wide range of models that can be seen as proxies for item response theory models. The difNLR package offers various methods for estimation of parameters and DIF detection procedure. It also covers procedures in DIF identification such as item purification or corrections for multiple comparisons. Moreover, simulation studies suggest good properties even in smaller samples (Drabinová & Martinková, 2017), and thus the family of models offered by the difNLR library seems to be promising in DIF detection.
Klasifikace
Druh
O - Ostatní výsledky
CEP obor
—
OECD FORD obor
10103 - Statistics and probability
Návaznosti výsledku
Projekt
<a href="/cs/project/GJ15-15856Y" target="_blank" >GJ15-15856Y: Odhad psychometrických vlastností jako součást vývoje přijímacích testů</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Ostatní
Rok uplatnění
2018
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů