EMPIRICAL MODELS FOR LONGITUDINAL POST-NATAL GROWTH

SuperLearning to define and predict composite outcomes

To develop a data-driven method for combining multivariate outcome measures (e.g., achievement scores) into a composite score in situations where investigators lack scientific grounding for the use of other composite scores.

OUTCOMES

Native language, learned language, and math competency test scores.

Predictors (other than time)

Health care access; use of preventive health care; child:adult ratio; child dependency ratio; crowding index; urban score; total family income; socioeconomic status; sanitation; access to clean water; mother's age,height, years of education, marital status, age at first child, and parity; father's age and years of education; weight-for-age z-score and HAZ (0,6,12,18, and 24 months); maternal smoking during pregnancy, child's sex and gestational age at birth

Explore Further