The "mushrooming" of private schools for low-income families has been widely noted in the literature; however, very little is known about the quality of these schools. This research explored the relative quality of private unaided (recognised and unrecognised) and government schools in low-income areas of Hyderabad, India. A preliminary census to locate unrecognised private schools--not on official lists--was conducted. Data were collected on achievement and background variables for 3,910 pupils from a stratified random sample of schools. Using multilevel modelling shows that pupils in private unrecognised and recognised schools, when controlled for age, pupil's IQ, and class average IQ, achieve higher scores in mathematics and English than equivalent pupils in government schools. There is no significant difference between private and government schools in pupil achievement in Urdu. The achievement advantage for private schools did not arise because of greater resources available, at least in terms of per pupil teacher salaries. (Contains 16 tables and 8 figures.)