Date of Award

2000

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Education

First Advisor

Abbas Tashakkori

Abstract

The present study investigated the interaction of cognitive style and assessment format (multiple-choice (MC) and performance-based assessments (PBA)) in second language proficiency. The participants consisted of 258 eighth-grade students. The second language achievement of the students was assessed by the Eighth Grade Proficiency/Credit Exam: French I. The cognitive style of the students was assessed by the Group Embedded Figures Test. The study also examined study habits and attitudes of field-dependent (FD) and field-independent (FI) students toward the two assessment formats. The study utilized a sequential mixed-model design, consisting of both quantitative and qualitative methods. Phase I consisted of a quantitative study investigating performance difference between FD and FI students on different test formats. Phase II consisted of a qualitative study to investigate assessment preferences and study habits of FD and FI students. Results revealed that cognitive style had a statistically significant effect on student performance whereas the effects of gender, ethnicity and socio-economic status of the students were not significant. A two-factor split-plot analysis revealed a significant interaction of cognitive style and test format. FI students outperformed FD students in the MC, but no indication of such difference was observed for the PBA. Furthermore, FD students scored better on the PBA than they did on the MC. The FI students scored better on the MC. Overall, the study indicated that compared to multiple-choice format, the performance-based assessment of second language proficiency is less impacted by student attributes such as cognitive style, gender, ethnicity, and socio-economic status. Qualitative interviews with students and their teachers revealed that there are differences between study habits of FD and FI students, but no difference was observed in terms of their assessment preferences. Lincoln and Guba's (1985) constant comparative method was utilized for the analysis of the result.

ISBN

9780599681958

Pages

238

Share

COinS