Evaluation of Model Selection Strategies for Cross-Level Two-Way Differential Item Functioning Analysis

Patarapichayatham, Chalie; Kamata, Akihito; Kanjanawasee, Sirichai

doi:10.1177/0013164411409743

Cited by 11 publications

(6 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One approach, for example, suggested for analyzing categorical dependent variables (e.g., item responses) with multilevel structure is the Hierarchical Generalized Linear Model (HGLM; Raudenbush & Simpson, 1999). HGLM is empirically supported for DIF detection (Cheong, 2006;Patarapichayathan, Kamata, & Kanjanawasee, 2012;Williams & Beretvas, 2006). In addition to the HGLM approach, a multilevel version of logistic regression has also been suggested for DIF detection in the multilevel context , as has the MIMIC model (Finch & French, 2011).…”

Section: Multilevel Data Structurementioning

confidence: 99%

Extensions of Mantel–Haenszel for Multilevel DIF Detection

French

Finch

2013

Educational and Psychological Measurement

View full text Add to dashboard Cite

Multilevel data structures are ubiquitous in the assessment of differential item functioning (DIF), particularly in large-scale testing programs. There are a handful of DIF procures for researchers to select from that appropriately account for multilevel data structures. However, little, if any, work has been completed to extend a popular DIF method to this case. Thus, the primary goal of this study was to introduce and investigate the effectiveness of several new options for DIF assessment in the presence of multilevel data with the Mantel–Haenszel (MH) procedure, a popular, flexible, and effective tool for DIF detection. The performance of these new methods was compared with the standard MH technique through a simulation study, where data were simulated in a multilevel framework, corresponding to examinees nested in schools, for example. The standard MH test for DIF detection was employed, along with several multilevel extensions of MH. Results demonstrated that these multilevel tests proved to be preferable to standard MH in a wide variety of cases where multilevel data were present, particularly when the intraclass correlation was relatively large. Implications of this study for practice and future research are discussed.

show abstract

Section: Multilevel Data Structurementioning

confidence: 99%

Extensions of Mantel–Haenszel for Multilevel DIF Detection

French

Finch

2013

Educational and Psychological Measurement

View full text Add to dashboard Cite

show abstract

“…Correlation Analysis (Patarapichayatham et al, 2012), (C. M. Decision Trees (Sheard & Hurst, 2003), (Weibelzahl et al, 2007), (Ranjan & Khalil, 2008), (Pechenizkiy et al, 2008), (Guo & Zhang, 2009), (Y. H. Wang et al, 2009), (Hämäläinen & Vinni, n.d.), (Y. Zhang et al, 2010), (Kovačić, 2010), (Bhargava et al, 2010), (Romero Morales, Cristóbal;Márquez Vera, Carlos;Ventura Soto, 2012), (Baradwaj & Pal, 2012b), (Hung et al, 2012), (Şen et al, 2012), (Agarwal et al, 2012), (Priya, 2013), (Lin et al, 2013), (Cristóbal Romero, Espejo, et al, 2013), (Priyam et al, 2013), (Blagojevic, 2013), (Badr et al, 2014), (Mugla, 2014), (Yukselturk & Education, 2014), (Saxena, 2015), (Patidar et al, 2015), (Thomas, 2015), (Shahiri & Husain, 2015), (Shukor et al, 2015) Regression Trees (Hu et al, 2014) Markov Chains (Stamper & Barnes, 2009) (Hämäläinen & Vinni, n.d.), (Y.…”

Section: Applied Techniquementioning

confidence: 99%

Data mining techniques applied in educational environments: Literature Review

Manjarres

Sandoval

Suárez

2018

DER

View full text Add to dashboard Cite

Data Mining for Education is an emerging discipline which seeks to develop methods to explore large amounts of data from educational settings, in order to understand students better. In recent years there have been various works related with this specialty and have used multiple data mining techniques derived from this to address different educational problems. The aim of this publication is to present a review of the works in which data mining techniques were used to solve specific problems of education and to do a classification of these, associated with diverse scenarios in which it has been applied.Resumen: La minería de datos para la educación es una disciplina emergente la cual busca el desarrollo de métodos que permitan explorar grandes cantidades de datos provenientes de entornos educativos con el fin de entender mejor a los estudiantes. En los últimos años se han realizado diversos trabajos relacionados con esta disciplina y se han utilizado múltiples técnicas de minería de datos para abordar diferentes problemáticas educativas. El objetivo de esta publicación es presentar una revisión de los trabajos en los cuales se han utilizado técnicas de minerías de datos para solucionar problemáticas particulares de la educación y realizar una clasificación de éstas asociadas a los diversos escenarios en los que se han aplicado.

show abstract

“…In nonparametric method there is no estimation of model parameters, whereas parametric method involves estimation of model parameters (Teresi & Fleishman, 2007). Both, the parametric and non-parametric needs a matching criterion (Patarapichayatham, Kamata, & Kanjanawasee, 2012;Scarpati, Wells, Lewis, & Jirka, 2011). In summary, the authors summarize it in Table 1.…”

Section: Differential Item Functioning (Dif)mentioning

confidence: 99%

Does Multidimensionality Cause DIF?

Ridho

2018

ANIMA IPJ

View full text Add to dashboard Cite

The differential item functioning (DIF) of an item that initially assumed unidimensional is frequently attributed to the assumption of multidimensionality. Therefore, it is important to test the assumption that multidimensionality causes an item to be functionally different between the disadvantaged group (i.e. focal group) and the benefited group (i.e. reference group) on the Aptitude Potential Test for New Student Selection in State Islamic University (AP SPMB-PTAIN). This study aims to: (a) explore and confirm the internal structure of AP SPMB-PTAIN; (b) identify items containing DIF based on the types of school the candidates attended (Madrasah Aliyah/MA, that is, secondary education managed by the Ministry of Religious Affairs, or regular high school/SMA); and (c) evaluate the multidimensionality effects on DIF. The data analyses (n = 10,000) showed that: (1) the internal structure of AP SPMB-PTAIN is semi-complex multidimensional; (2) 15 items contain DIF UIRT (12 items benefited high school graduates while three items benefited MA graduates); five items contain DIF MRT that benefited high school graduates; and (3) the multidimensionality difference between the focal and reference group did not appear to correspond to DIF.

show abstract

Evaluation of Model Selection Strategies for Cross-Level Two-Way Differential Item Functioning Analysis

Cited by 11 publications

References 5 publications

Extensions of Mantel–Haenszel for Multilevel DIF Detection

Extensions of Mantel–Haenszel for Multilevel DIF Detection

Data mining techniques applied in educational environments: Literature Review

Does Multidimensionality Cause DIF?

Contact Info

Product

Resources

About