Three Generations of DIF Analyses: Considering Where It Has Been, Where It Is Now, and Where It Is Going

Using the Johnson‐Neyman Procedure to Detect Item Bias in Personality Tests

“…Advancing item bias methods and making them available to a wider audience will not only yield more valid group comparisons and better refined measurement instruments, but also enhance theory development through understanding of why specific items are biased across groups. Therefore, as suggested by Zumbo (2007), new generation item bias methods should be made accessible to researchers who are not measurement specialists. The current chapter is an attempt in this direction.…”

“…The significance of this problem has also led to development of various statistical techniques to detect biased items (for reviews, see Reynolds, 2000;Zumbo, 2007). In the current chapter, I first review one of these techniques known as the analysis of variance (ANOVA) procedure (van de Vijver and Leung, 1997), which has been widely used in personality research (e.g., Caprara, Barbaranelli, Bermúdez, Maslach, & Ruch, 2000;Ramírez-Esparza, Gosling, Benet-Martínez, Potter, & Pennebaker, 2006;Vecchione, Alessandri, & Barbaranelli, 2012).…”

MIMIC, SIBTEST, Lojistik Regresyon ve Mantel-Haenszel Yöntemleriyle Gerçekleştirilen DMF ve Yanlılık Çalışması

“…Bu durumlarda performanslar arasında gözlenen farklılığa madde etkisi denir (Clauser ve Mazor, 1998). Madde etkisi, maddeler tarafından ölçülen yetenek bakımından gruplar arasında gerçek bir farklılık olması nedeniyle DMF'nin ortaya çıktığı durumları anlatmak için kullanılır (Zumbo, 2007). DMF, bahsedildiği üzere gruplar arasındaki gerçek farklılıktan değil de madde özellikleri veya test koşullarından kaynaklanıyorsa bu durum yanlılık olarak ifade edilir.…”

Propensity Score Matching Helps to Understand Sources of DIF and Mathematics Performance Differences of Indonesian, Turkish, Australian, and Dutch Students in PISA

“…DIF occurs and threatens the comparability when students with the same ability level on the underlying construct but coming from different groups (e.g., females and males) show dissimilar mean scores on an item (Van de Vijver & Leung, 1997;Zumbo, 2007). DIF analyses are used for many purposes such as fairness and equity in testing, dealing with a possible threat to internal validity, investigating the comparability of translated and/or adapted measures, trying to understand item response processes, and investigating (lack of) invariance (Zumbo, 2007).…”