Assessing the impact of uniform and nonuniform differential item functioning items on Rasch measure: the polytomous case

Golia, Silvia

doi:10.1007/s00180-014-0542-x

Cited by 4 publications

(14 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Two types of DIF can be identified and denoted as uniform and nonuniform [ 22 ]. Uniform-DIF is the simplest type of DIF where the probability of selecting the specific category of item is greater (or lesser) for one group than the other in all levels of the latent construct uniformly.…”

Section: Methodsmentioning

confidence: 99%

The Effect of Small Sample Size on Measurement Equivalence of Psychometric Questionnaires in MIMIC Model: A Simulation Study

Jamali

Ayatollahi

Jafari

2017

BioMed Research International

View full text Add to dashboard Cite

Evaluating measurement equivalence (also known as differential item functioning (DIF)) is an important part of the process of validating psychometric questionnaires. This study aimed at evaluating the multiple indicators multiple causes (MIMIC) model for DIF detection when latent construct distribution is nonnormal and the focal group sample size is small. In this simulation-based study, Type I error rates and power of MIMIC model for detecting uniform-DIF were investigated under different combinations of reference to focal group sample size ratio, magnitude of the uniform-DIF effect, scale length, the number of response categories, and latent trait distribution. Moderate and high skewness in the latent trait distribution led to a decrease of 0.33% and 0.47% power of MIMIC model for detecting uniform-DIF, respectively. The findings indicated that, by increasing the scale length, the number of response categories and magnitude DIF improved the power of MIMIC model, by 3.47%, 4.83%, and 20.35%, respectively; it also decreased Type I error of MIMIC approach by 2.81%, 5.66%, and 0.04%, respectively. This study revealed that power of MIMIC model was at an acceptable level when latent trait distributions were skewed. However, empirical Type I error rate was slightly greater than nominal significance level. Consequently, the MIMIC was recommended for detection of uniform-DIF when latent construct distribution is nonnormal and the focal group sample size is small.

show abstract

Section: Methodsmentioning

confidence: 99%

The Effect of Small Sample Size on Measurement Equivalence of Psychometric Questionnaires in MIMIC Model: A Simulation Study

Jamali

Ayatollahi

Jafari

2017

BioMed Research International

View full text Add to dashboard Cite

show abstract

“….) or unknown groups (latent classes) (Golia, 2015;Cho et al, 2016;Davis et al, 2020;Kleppang et al, 2020) for detecting so-called "biased" items and assessing their impact on the latent trait parameters' estimates. Although longitudinal IRT or RMT models could also be used in the same way to investigate RS or lack of longitudinal invariance depending on covariates, longitudinal invariance is rarely investigated with IRT or RMT models and covariates are seldom included in such models.…”

Section: Introductionmentioning

confidence: 99%

“…However, an additional issue arises as patients might have a different perception of the measured PRO depending on their group membership at a specific time point, a phenomenon known as differential item functioning (DIF). Integrating covariates in cross-sectional IRT or RMT models used as latent regression models is quite common to investigate DIF between known (e.g., gender, age…) or unknown groups (latent classes) ( Golia, 2015 ; Cho et al, 2016 ; Davis et al, 2020 ; Kleppang et al, 2020 ) for detecting so-called “biased” items and assessing their impact on the latent trait parameters’ estimates. Although longitudinal IRT or RMT models could also be used in the same way to investigate RS or lack of longitudinal invariance depending on covariates, longitudinal invariance is rarely investigated with IRT or RMT models and covariates are seldom included in such models.…”

Section: Introductionmentioning

confidence: 99%

How to Investigate the Effects of Groups on Changes in Longitudinal Patient-Reported Outcomes and Response Shift Using Rasch Models

et al. 2020

View full text Add to dashboard Cite

In order to investigate patients’ experience of healthcare, repeated assessments of patient-reported outcomes (PRO) are increasingly performed in observational studies and clinical trials. Changes in PRO can however be difficult to interpret in longitudinal settings as patients’ perception of the concept being measured may change over time, leading to response shift (longitudinal measurement non-invariance) and possibly to erroneous interpretation of the observed changes in PRO. Several statistical methods for response shift analysis have been proposed, but they usually assume that response shift occurs in the same way in all individuals within the sample regardless of their characteristics. Many studies aim at comparing the longitudinal change of PRO into two groups of patients (treatment arm, different pathologies, …). The group variable could have an effect on PRO change but also on response shift effect and the perception of the questionnaire at baseline. In this paper, we propose to enhance the ROSALI algorithm based on Rasch Measurement Theory for the analysis of longitudinal PRO data to simultaneously investigate the effects of group on item functioning at the first measurement occasion, on response shift and on changes in PRO over time. ROSALI is subsequently applied to a longitudinal dataset on change in emotional functioning in patients with breast cancer or melanoma during the year following diagnosis. The use of ROSALI provides new insights in the analysis of longitudinal PRO data.

show abstract

“…Lee and Zhang (2017) have found differences in estimations of ability when the ratio of items with DIF increased. Golia (2015) examined how the ability estimations would change in instrument that belonged the polytomously scored items with DIF. If the test belonged more than one items with DIF, there was a significant bias in estimations of ability.…”

Section: Discussionmentioning

confidence: 99%

“…Buna göre tek biçimli DMF, aynı yetenek düzeyindeki iki grubun bir maddeye doğru yanıt verme olasılıklarının tüm yetenek düzeyleri için sabit bir değer olması durumunda meydana gelir. Buna karşın tek biçimli olmayan DMF ise aynı yetenekteki iki grubun maddeye doğru yanıt verme olasılıkları farklı yetenek düzeylerinde tutarsız olduğu durumda oluşur (Camilli ve Shepard, 1994;Penfield ve Lam, 2000;Zumbo, 1999 Alanyazında maddelerin ikili puanlandığı (Lee ve Zhang, 2017;Li ve Zumbo, 2009;Roznowski ve Reith, 1999;Rupp ve Zumbo, 2003Wells, Subkoviak ve Serlin, 2002) ve çoklu puanlandığı (Golia, 2010(Golia, , 2015Tennant ve Pallant, 2007) durumlarda testten madde çıkarılmasının teste ilişkin istatistikleri nasıl değiştiğine dair çalışmalar bulunmaktadır. Bu çalışmaların bir kısmı madde parametreleri değişmezliği kapsamında bu durumu incelerken (Roznowski ve Reith 1999;Rupp ve Zumbo, 2003Well, Subkoviak ve Serlin, 2002), bazıları ise ilgili durumu bu çalışmada olduğu gibi DMF kapsamında parametre değişmezliği olarak ele almıştır (Golia, 2010(Golia, , 2015Lee ve Zhang, 2017;Li ve Zumbo, 2009;Tennant ve Pallant, 2007 DMF'li maddelerin çıkarıldığı testlerin bireylerin yetenek kestirimine olan etkilerinin araştırılmasında bu çalışma kapsamında tek biçimli, tek biçimli olmayan, hem tek biçimli hem tek biçimli olmayan DMF türünde %10 ve %20 DMF'li madde barındıran koşullarda farklı yöntemler karşılaştırılmıştır.…”

Section: Journal Of Measurement and Evaluation In Education Andunclassified

Performances Based on Ability Estimation of the Methods of Detecting Differential Item Functioning: A Simulation Study

Uysal

Ertuna

Ertaş

et al. 2019

Eğitimde Ve Psikolojide Ölçme Ve Değerlendirme Dergisi

View full text Add to dashboard Cite

The aim of the study is to examine differential item functioning (DIF) detection methods-the simultaneous item bias test (SIBTEST), Item Response Theory likelihood ratio (IRT-LR), Lord chi square (χ2), and Raju area measures-based on ability estimates when purifying items with DIF from the test, considering conditions of ratio of the items with DIF, effect size of DIF, and type of DIF. This study is a simulation study and 50 replications were conducted for each condition. In order to compare DIF detection methods, error (RMSD) and coefficient of concordance (Pearson's correlation coefficient) were calculated according to estimated and initial abilities for the reference group. As a result of the study, the lowest error and the highest concordance were seen in the case of 10% uniform DIF in the test and the method of IRT-LR, considering all other conditions. Moreover, for the method of SIBTEST and IRT-LR in all conditions, it was found that the error obtained by purifying items with C level DIF is lower than the error obtained by purifying items with both B and C level DIF. Similarly, for the method of SIBTEST and IRT-LR in all conditions, it was seen that the concordance coefficient found by purifying C level DIF is higher than the coefficient by purifying items with both B and C level DIF.

show abstract

Assessing the impact of uniform and nonuniform differential item functioning items on Rasch measure: the polytomous case

Cited by 4 publications

References 15 publications

The Effect of Small Sample Size on Measurement Equivalence of Psychometric Questionnaires in MIMIC Model: A Simulation Study

The Effect of Small Sample Size on Measurement Equivalence of Psychometric Questionnaires in MIMIC Model: A Simulation Study

How to Investigate the Effects of Groups on Changes in Longitudinal Patient-Reported Outcomes and Response Shift Using Rasch Models

Performances Based on Ability Estimation of the Methods of Detecting Differential Item Functioning: A Simulation Study

Contact Info

Product

Resources

About