South Africa’s education system is still deep in the throes of reform under its third Minister of Education since 1994. Poor communities, in particular those of rural Africans, bear the brunt of the past inequalities. The challenge was to explore the extent of the ‘‘gap’’ in students’ scores by comparing the advantaged and disadvantaged communities in this context. The TIMSS-Repeat 1999 data were explored and 3 categories of students were ultimately identi?ed: advantaged, semi-advantaged, and disadvantaged groups. Partial least squares analysis was applied to explore the science performance but very few factors were found that consistently predicted performance across and within these groups. However, one dominant factor emerged in these models and that was the students’ performance in the locally developed English test that provided a measure of students’ pro?ciency in English, the language in which more than 70% of the students wrote the science tests. Students who had a higher score on the English test also performed better in the science test, despite their backgrounds.