Panel Paper: The Mpact Initiative: Using Behavioral Tools to Improve Children's Early Math Skills

Saturday, April 13, 2019
Continuing Education Building - Room 2030 (University of California, Irvine)

*Names in bold indicate Presenter

Ariel Kalil, Susan E. Mayer and William Delgado, University of Chicago


Disadvantaged parents spend less time and especially less time in educational activities with their children compared to their more advantaged counterparts. Low parental time in enriching activities has adverse consequences for children’s future attainments. One potential explanation for the lack of parental engagement is present bias. Present-oriented parents over-value immediate payoffs to their time in, e.g., leisure or non-parental activities, and they under-value future payoffs to their time investment in children. This leads to procrastination and underinvestment in their children.

We designed the Math for Parent and Children Together (MPACT) intervention to estimate the impacts on children’s early math skills of a low-cost, light-touch treatment intended to overcome present bias in parental investment. We test (i) whether providing information and materials in the form of a math activity booklet increases test scores, (ii) whether a treatment designed to overcome present bias and increase parents’ use of the materials affects child’s math skills beyond providing the information and materials alone, and (iii) whether child’s skills are influenced by their peers.

MPACT is a 12-week intervention currently in the field and has recruited more than 1,400 children from low-income families enrolled in Head Start programs in Chicago. Randomization occurs in two stages. In the first stage, classrooms are assigned to the untreated and treated classroom groups. In the second stage, children in untreated classrooms are assigned to the control group only, and children in treated classrooms are randomly assigned to the control, MKit, and MKit + present bias treatment groups. Control families receive a placebo treatment (a reading book), families in the MKit group are provided a booklet with math activities for parents and children to do together, and those in the MKit + present bias group get the same booklet and in addition receive multiple text messages per week intended to overcome present bias. The primary outcome is the child’s math skills measured with the Woodcock Johnson IV Applied Problems and the Preschool Early Numeracy Scales. These outcomes are measured at baseline and immediately following the 12-week intervention.

The data collection is being conducted in three rounds. We have data from parent surveys, teacher surveys, parents’ scores on time-preference tasks, and child assessments. We will finalize the third and last round in early March and will present results at the APPAM Regional Student Conference.