Description: Australia is primarily a country of immigrants. Immigration began with the establishment of a British penal colony at Sydney Cove in 1788, and many free settlers joined the convicts who were transported to serve their sentences in Australia, particularly during the gold rushes of the 1850s. After the end of World War II, the Australian Government entered into agreements with a number of countries to provide free and assisted immigration to Australia, and Australia has continued to provide a home for a variety of people fleeing the aftermath of war or simply seeking a different life. Migrants relocating to Australia, particularly refugees, tend to move into areas into which others from their community have previously settled, often in areas where housing is cheap. As a result, schools in these areas often have a high proportion of students with a language background other than English and also often have a high proportion of students from a disadvantaged background. Intuitively, having a high proportion of students in a school with a language background other than English is problematic for teachers and the school. The TIMSS 2003 International Mathematics Report (Mullis, Martin, Gonzalez & Chrostowski, 1994) supports this, finding that in most countries students from homes where the language of the test is always or almost always spoken had a higher average achievement than those who spoke it less frequently (p. 127). In Australia, however, students who spoke English only sometimes at home achieved the highest scores on the TIMSS mathematics test. The Australian TIMSS analysis (Thomson & Fleming, 2004) also found that Year 8 students born in a non-English-speaking country, whose home language was not English, achieved at a significantly higher level than students whose home language was English, regardless of where they were born (p. 62). Other research examining subject choice by Australian-born students and students from migrant families has found that students from migrant families are more likely to be enrolled in higher mathematics and physical science courses in senior secondary school, supporting the popular belief that students from an LBOTE background are more likely to enrol in such subjects because they lead more often to higher education and the professions (Fullarton, Walker, Ainley & Hillman, 2003). Having a high proportion of students in a school from a disadvantaged background is also problematic for teachers and the schools, and the TIMSS 2002/03 data, both internationally and for Australia in particular, support this. Initial data analysis on the Australian TIMSS 2002/03 Year 8 data found that for schools with a low level of disadvantaged students, having a high proportion of students with a language background other than English (LBOTE) was reflected in a higher average score in mathematics (around half a standard deviation) than for those schools with lower proportions of LBOTE students. For those schools with a high proportion of disadvantaged students the proportion of LBOTE students barely had an effect. These findings illustrate one of the challenges in drawing conclusions from complex data sets such as TIMSS and this paper will explore in greater depth the relationship between language background and socioeconomic background in Australia.