The MIMIC Model as a Method for Detecting DIF: Comparison With Mantel-Haenszel, SIBTEST, and the IRT Likelihood Ratio

By Holmes Finch. 2005

250 citations recorded by the source corpus.

Subfields: Applied Psychology, Computer Networks and Communications, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/7f8815419277ad80b742989d1a6af1def4d64833

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Applied Psychology

In Computer Networks and Communications

In Management Science and Operations Research

Citations in the corpus

MIMIC Methods for Assessing Differential Item Functioning in Polytomous Items (2010, methodology): “Multiple indicators–multiple causes (MIMIC) confirmatory factor analysis has been used to assess differential item functioning (DIF) in dichotomous items (Finch, 2005; Navas-Ara & Gómez-Benito, 2002; Shih & Wang, 2009; Wang, Shih, & Yang, 2009).”
Testing for Nonuniform Differential Item Functioning With Multiple Indicator Multiple Cause Models (2011, methodology): “This is consistent with the way MIMIC models are described for testing DIF in extant literature (Chen & Anthony, 2003; Christensen et al., 1999; Finch, 2005; Fleishman, Spector, & Altman, 2002; Gelin, 2005; Grayson, Mackinnon, Jorm, Creasey, & Broe, 2000; Hagtvet & Sipos, 2004; MacIntosh & Hashim,…”
Measurement Invariance Testing with Many Groups: A Comparison of Five Approaches (2017, background): “Researchers can examine the source of intercept noninvariance by regressing the observed variables or items on the covariates (e.g., Finch, 2005).”
The DIF-Free-Then-DIF Strategy for the Assessment of Differential Item Functioning (2012, methodology): “This method has been implemented in Winsteps (Linacre, 2003) and IRTLRDIF (Thissen, 2001) as well as confirmatory factor analysis–based DIF detection methods (Finch, 2005; Stark, Chernyshenko, & Drasgow, 2006).”
Testing Measurement Invariance: A Comparison of Multiple-Group Categorical CFA and IRT (2011, methodology): “Although the strategies to identify the invariant referents were also introduced (French & Finch, 2008), in the reality of research, the unbiased reference variables are not likely to be apparent. For this reason, the backward procedure has an advantage in testing measurement invariance over the for”
What’s the DIF? Why Differential Item Functioning Analyses Are an Important Part of Instrument Development and Validation (2011, methodology): “Moreover, simulation studies have shown that using CFA to test for DIF results in extremely high Type I error rates when the three parameter logistic model underlies the data and a test is composed of only 20 items (Finch, 2005).”
A Synthesis of 15 Years of Research on DIF in Language Testing: Methodological Advances, Challenges, and Recommendations (2007, result): “Yet research suggests that the utilization of a single method is undesirable and might represent an oversimplification due to a method effect, because different statistical methods typically yield conflicting results, at least for some items (eg, Ferne, Choi, & Rupp, 2005 ; Greer, 2004 ; Holmes, ”
Differential Item Functioning Detection Using the Multiple Indicators, Multiple Causes Method with a Pure Short Anchor (2009, methodology): “In recent years, MIMIC methods have been used to detect DIF (Finch, 2005; Fleishman, Spector, & Altman, 2002; Gallo, Anthony, & Muthén, 1994; Glöckner-Rist & Hoitjink, 2003; Levine et al., 2003; MacIntosh & Hashim, 2003; Muthén, 1985, 1988; Muthén, Kao, & Burstein, 1991; Oort, 1998).”
Measurement Invariance and Differential Item Functioning in Latent Class Analysis With Stepwise Multiple Indicator Multiple Cause Modeling (2017, methodology): “The Type I error has been shown to be inflated with this approach (Finch, 2005) although there are effective adjustments to α levels or p values to control the Type I error inflation (J. Kim & Oshima, 2013).”
Comparing DIF Across Math and Reading/Language Arts Tests for Students Receiving a Read-Aloud Accommodation (2006, background): “It is important to note that a recent study showed IRTLRDIF methods to result in higher Type-I error when the test is contaminated with several DIF items (Finch, 2005).”
Testing Measurement Invariance Using MIMIC (2012, methodology): “The previous simulation studies consistently presented high Type I error rates in detecting noninvariant variables with MIMIC modeling (Finch, 2005; Navas-Ara & Gomez-Benito, 2002; Oort, 1998; Wang et al., 2009).”
Empirical Selection of Anchors for Tests of Differential Item Functioning (2009, background): “These problems may account for the more frequently reported problem of inflated Type I error or false discovery rate (Finch, 2005; Meade & Lautenschlager, 2004; Stark, Chernyshenko, & Drasgow, 2006; Wang, 2004; Wang & Yeh, 2003).”