Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Browse By Descriptor
Search Tips
Annual Meeting Theme
Exhibitors
About Philadelphia
About AERA
Personal Schedule
Sign In
X (Twitter)
The results from assessments are frequently used for making claims about the effectiveness of instruction at the classroom, school, district and national level. The validity of such claims critically depends on the degree to which they are meaningful for different student populations whose learning contexts may differ. The context of learning is of particular relevance in assessments of reading for students from different language backgrounds. Development of reading for students whose home language differs from that of their schooling and assessment may be different than those students whose home language is the same as for schooling and assessment. For example, development in reading is expected to be supported by engaging in reading materials at home in the same language as schooling. Other literacy resources and practices are also expected to affect students’ attitudes toward reading and ultimately affect development in reading. Thus, without an understanding of students’ contexts for learning, the interpretation of assessment scores in reading is next to impossible.
Validity inquiry requires the consideration of context, which includes the interpretation of assessment results with respect to the experience of the individual (Moss, 1992). Moss notes that the inclusion of context has become increasingly salient in the discussion of generalizability of scores. Specific to language and assessment, Solano-Flores & Trumbull (2003) also emphasized the importance of context. In particular, these authors called attention to the learning contexts of students from diverse linguistic backgrounds, and how contextual variation may affect the inferences from assessments. More specifically, they point out that students may respond to assessment items in distinct ways that are tied to the students’ linguistic or cultural backgrounds.
This research focused on examining differences in learning contexts of Canadian students who attend schools in English and (a) speak English at home, (b) speak French at home, or (c) speak another language at home. The three student groups were compared based on their (a) attitudes toward school, (b) diversity of reading, (c) enjoyment of reading, (d) library use, (e) use of memorization strategies, (f) online reading habits, (g) student-teacher relations, and (h) attitudes towards computers. These variables were chosen because factors such as attitudes toward school and reading, learning strategies, and typical reading habits all contribute to the context of learning and literacy. This paper investigated the question of whether students of different linguistic backgrounds differ on factors that are related to learning contexts. Comparisons were based on data from PISA 2009 reading literacy assessments for Canadian students.
Based on results from analysis of variance and chi-square difference tests, there were differences in reading habits and learning strategies for students who speak another language at home compared to those who speak one of Canada’s two official languages at home. These differences are further described in the paper, in addition to possible implications of these findings. The research contributes to methods used in examining the validity of assessments for ELLs. It also yields insights that can inform interpretation of scores for ELL and non-ELL students and differences in their academic achievement.
Juliette Lyons-Thomas, Regents Research Fund
Kadriye Ercikan, The University of British Columbia
Wolff-Michael Roth, University of Victoria
Dallie A. Sandilands, EMP Educational Measurement Professionals
Shawna Goodrich, The University of British Columbia