Session Summary
Share...

Direct link:

The New Kids on the Methods Block: Using New Methods in Existing Data to Get New Answers

Wed, April 7, 10:00 to 11:30am EDT (10:00 to 11:30am EDT), Virtual

Session Type: Paper Symposium

Abstract

The rapid expansion of quantitative methods in the social sciences can provide child development researchers with opportunities to address new questions, or to get new answers, in existing datasets. At the same time, child development research often requires that these methods are adapted to represent the phenomena we are interested in, being either guided by the idiosyncrasies of human development, of developmental theory, or of a need for theories of measurement.

In this symposium, three papers will be presented that expand on methodology for causal inference from other fields of social science, by adapting them to special circumstances relevant for child development research. The first paper uses synthetic controls to model the effects of a school-based intervention (City Connects) when assumptions for traditional quasi-experimental designs are violated. The second paper exploits a well-known dataset in our field, the Family Life Project, using a triple-differencing model to estimate the effect of a tobacco-tax (hence reduced smoking in the home) on child outcomes. The third paper shows how structural equation modeling can be used to combine psychometric measurement models with quasi-experimental approaches primarily used in economics, like instrumental variable-, difference-in-difference-models. The discussant is a labor economist well versed in the field of child development, who will address the importance of expanding novel methodologies in this field.

Sub Unit

Chair

Discussant

Individual Presentations