Open AccessComparison sensitivity of the differential item function (DIF) detection method
Author(s) -
Ulfa Khoeruroh,
Heri Retnawati
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1511/1/012042
Subject(s) - differential item functioning , statistics , test (biology) , differential (mechanical device) , sensitivity (control systems) , meaning (existential) , psychology , mathematics , function (biology) , item response theory , econometrics , psychometrics , engineering , biology , aerospace engineering , evolutionary biology , paleontology , electronic engineering , psychotherapist
Differential Item Functioning (DIF) or often referred to as bias is a term that is not good, has a racial meaning, suppresses or is too fanatical about the object to be measured. Bias in a test is a test condition that is unfair, inconsistent, contaminated by factors outside the aspects to be tested and misuse of the test. The purpose of this study is to find out which method is the most sensitive in detecting the presence of DIF on items. The methodology used is to study literature from books, journals and study the results of research that has been done. There are many methods that can be used to detect DIF. These methods are the Mantel-Haenszel Method, the Chi-Squared Method of the Lord, the Probability Comparison Test Method, Item Characteristic Curve (ICC) and Extent Method According to Raju.