Modeling Parent Decisionmaking: Employment and Child Care

• Activity Number: 658
• Type: Contributed
• Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Caroline Carlin*+ and Elizabeth Davis
• Companies: University of Minnesota and University of Minnesota
• Keywords: selection ; choice modeling ; simultaneous equations ; Bayesian ; child care ; employment

Abstract

Decisions that parents make with regard to non-parental care for their children are intertwined with (usually maternal) employment decisions. In addition, cost of a potential care option is observed only when the care option is selected. This project uses recent, nationally representative, longitudinal data (Early Childhood Learning Survey - Birth Cohort) and Bayesian econometric methods to explore this network of decisions. Our model captures separately the decision to work for pay (none, part-time or full-time), the choice of child care setting (relative, non-relative, center, Head Start, parental), and the option-specific price of child care. The price model explicitly considers (for each type of care) the decision to use free care (e.g., grandma) or fee-based care, the use of public subsidies, and ultimately predicts the subsidized or unsubsidized cost of care. Together, this combined employment and child care choice model also uses innovative methods to allow the outcome of one decision to influence the other, capturing error terms from latent variables to use as a regressors in subsequent equations, thus controlling for selection and feedback in the system of equations.