BackHome
Refresh: OffView Personal Schedule
Paper Summary

Do Reformed and Traditional Teaching Differently Influence Caucasian and Hispanic American Student Mathematics Achievements? Findings Using Data From TIMSS (Trends in International Mathematics and Science Study) 2007

Sat, April 14, 12:25 to 1:55pm, Vancouver Convention Centre, Floor: First Level, East Ballroom C

Abstract

Hispanic American students have been lagging behind their Caucasian peers in mathematics achievements over the years (National Center for Education Statistics, 2010). Their achievement gap undermines the U.S. economic competitiveness in today’s world that relies on the mathematics problem-solving skills of its workforce and U.S. political ideal as a democratic society where its citizens with different racial backgrounds can pursue social mobility and public good (Labaree, 1997) using well-developed mathematics knowledge. The teaching standards have been established to help teachers teach mathematics problem-solving by engaging students in communicating and justifying their own ideas, which the traditional teaching with its focus students on retaining facts, practicing procedures, and following routines is seen ineffective in developing (Romberg, 1992). This study examines whether and to what extent the two instructional approaches influence differently the achievements of the two racial groups in overall, problem-solving, and basic mathematics skills and knowledge.

This study is situated in two theoretical debates. The first debate exists between the argument that teaching students problem solving is more important than helping them retaining and applying basic mathematics concepts and rules (Romberg, 1992) and the contention that the well-retained mathematics concepts and fluent use of procedures acquired through traditional teaching are crucial for developing student mathematics problem solving skills (Wu, 1999). The second occurs between the implicit assumption that reformed teaching can be effective for students across racial and cultural groups (National Council of Teachers of Mathematics, 2000) and the argument of culturally relevant teaching that different racial groups need various instructional approaches to improve their learning due to their different learning needs, styles, and habits (Ladson-Billings, 1995, 1997).

This study used the U.S. data sets from TIMSS 2007 because of its national representative student sample (Joncas, 2008). The dependent variables were eighth grade students’ scores in overall, problem-solving, and basic mathematics skills and knowledge and the independent variables were two composite variables related to the reformed or traditional instructions based on teacher surveys. A two-level hierarchical linear modeling was used for the analysis to accommodate the clustered structure of TIMSS data (Raudenbush & Bryk, 2002).

Our analysis showed that the intra-class correlation coefficients from the baseline models ranged from .33 to .48 for all the three scores of the two groups, all ps < .001. In the full model, the between-classroom variance in the intercepts of three scores explained by Level 2 variables ranged from 5% to 9%. The traditional teaching variable was significantly related to all three scores of the two groups, all ps < .01, while the reformed teaching variable was not significantly related to all three achievement score of the two groups, all ps > .05.

These findings challenge the assumption supporting the reformed teaching (Romberg, 1992) while supporting the assumption underlying the traditional teaching (Wu, 1999). They also challenge the assumption of culturally relevant pedagogy (Ladson-Billings, 1995) since both teaching approaches had similar effects on the three achievement scores for both groups although each group had different learning needs, styles, and habits (Ogbu & Simons, 1994).

Authors