Onderzoek Naar Oorzaken van 'Differential Item Functioning' in de Eindtoets Basisonderwijs

Tamara van Schilt-Mol | Faculteit Communicatie en Cultuur, Universiteit van Tilburg

Ton Vallen | Faculteit Communicatie en Cultuur, Universiteit van Tilburg

Henny Uiterwijk | Cito, Arnhem

Previous research has shown that the Dutch 'Final Test of Primary Education' contains a number of unintentionally and therefore unwanted, difficult test items, leading to Differential Item Functioning (DIF) for immigrant minority students whose parents' dominant language is Turkish or Arab/Berber. Two statistical procedures were used to identify DIF-items in the Final Test of 1997. Subsequently, five experiments were conducted to detect causes of DIF, revealing a number of hypotheses concerning possible linguistic, cultural, and textual sources. These hypotheses were used to manipulate original DIF-items into intentionally DIF-free items. The article discusses three possible sources of DIF: (1) the use of fixed (misleading) answer-options and (2) of misleading illustrations (both in the disadvantage of the minority students), and (3) the fact that questions concerning past tense often lead to DIF (in their advantage).

Article language: Dutch

Published online: 24 March 2014
https://doi.org/10.1075/ttwia.74.13sch