Is parental involvement a relevant factor in explaining academic performance in the most disadvantaged socioeconomic contexts? This article examines the effect of parental involvement on the Language and Mathematics performance of third grade students attending low-SES schools in Santiago de Chile. Multilevel linear regression models are used (individuals and school) along with variables related to academic performance. Results show that parents' educational expectations and perceived self-efficacy have a positive and significant effect on academic performance in Language and Mathematics. Parental involvement practices at home only affect performance in Language, although this finding is not conclusive. Estimates revealed that involvement practices at school are not statistically significant. The implications of these results are discussed considering their context: a country with the highest levels of family income inequality and one of the most segregated school choice models in the world.