Do the Multiple Choices of the Canadian English Language Proficiency Index Program–General Test (CELPIP-G) Reading Items Mean the Same to Different Sexes?

• In Event: Exemplary Practices in Test Validity Research and Evaluation

Abstract

The Canadian government uses the CELPIP-G reading test scores to make high-stakes decisions for screening immigration eligibility. This paper examined whether the options of the multiple-choice questions in one of the test forms were free from measurement bias between sexes. To this end, differential options functioning (DOF) based on multinomial logistic regression (Park & Wu, in press) was conducted to test measurement invariance at the option level, i.e., whether the probabilities of choosing the options (vs. the keyed option) were different between equally proficient males and females. The results showed that only 1.9% of the options exhibited moderate-sized DOF and 1.2% exhibited large-sized DOF. Also, testlets that required comprehension of more complex texts were more likely to exhibit DOF.

Authors

• Amery D. Wu, University of British Columbia

• Minjeong Park, The University of British Columbia