Language, identity and the achievement gap: comparing experiences of African-American students in a French immersion and a regular education context

Abstract

The Black/White Achievement gap has been a persistent problem in education. Previous research attributed this gap to students' culture (Jenks & Phillips, 1998; Ogbu, 1995a.b) or teachers' expectancy (Rist, 1970). Post-colonial literature suggests that this research itself is oppressive, and that learning is negotiating the "spaces" between students and teachers (Ellsworth, 1997); creating a hybrid "mestiza" space (Anzaldúa, 1987). The openness of immersion to diversity, and its subsequent educational benefits for African-American students (Caldas & Boudreaux, 1999) conforms to this post-colonial perspective. This mixed-methodology study examined both academic achievement and the experiences of Louisiana fourth grade students/teachers in both the regular education and the French immersion contexts. The quantitative phase compared these students' LEAP test scores. The qualitative phase was a cross-case comparison of four classrooms--an extreme class (90% of the school population in poverty) and a typical class (African-Americans of average academic achievement) in each context. Quantitative findings were that while there was a bridging of the achievement gap between the LEAP math scores of African-American immersion students and those of white students in regular education, the gap remained amongst immersion students. The qualitative phase found the regular education classroom was found to be a more fixed and assimilating context than immersion. Further, immersion students had higher collective self-esteems and a more positive view of schooling. Though Typical Immersion appeared to create a hybrid third space, the regular education context in which immersion programs were situated appeared to negatively influence these programs.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 10 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.