Background:
The contradictory findings about the role of the socioeconomic status (SES) in accounting for substance use in adolescence can be explained by the lack of consensus on how best to conceptualize and measure SES. Based on Coleman's social theory, this study examines the association of family and school SES with tobacco and e-cigarette use.

Methods:
Data came from a cross-sectional, school-based survey of middle-school early adolescents attending 33 schools in three Argentinian cities (n=3,110). Family SES was evaluated using: 1) family affluence scale (FAS) for material capital, 2) parent education for human capital, and 3) nuclear family and parent-adolescent communication for social capital. A school SES index was computed aggregating FAS and parent education from students. Multilevel logistic regression models regressed current tobacco use (in the past 30 days) and ever e-cigarette use on family and school-level SES, controlling for sex and age.

Results:
17.7% students were current tobacco smokers and 7.7% have ever used e-cigarette. At family level, material capital was unassociated with tobacco smoking, but positively associated with e-cigarette use; human capital was unassociated with both tobacco smoking and e-cigarette use; and social capital was inversely associated with both tobacco smoking and e-cigarette use. After adjusting for family SES, students from schools with higher SES had lower likelihood of tobacco smoking (OR=0.25, 95%CI 0.09-0.74) and higher likelihood of e-cigarette use (OR=4.36, 95%CI 1.25-15.23) than those from schools with lower SES.

Conclusions:
Family capital measures are differentially related to student substance use, pointing out that SES indicators are not interchangeable. School SES has an independent effect on substance use above and beyond that of family SES. These findings emphasize the need for SES conceptualization and consistent selection of indicators in order to study the effect of SES on health research and its role in identifying students at-risk of substance use.