Abstract

The leading theories on parenting in the United States suggest parenting approaches vary by socioeconomic status, with middle/upper-class parents practicing “concerted cultivation”—marked by parents’ intensive efforts to foster their children’s development—and working-class parents engaging in the “accomplishment of natural growth”—with children given more freedom to manage their own time. This project will measure parenting approaches using computational social science methods applied to two data sources: an existing survey in which respondents gave parenting advice about a hypothetical scenario, and new data from

Amazon Mechanical Turk where respondents will respond to multiple scenarios in order to examine the durability of parenting approaches across contexts. We will analyze both datasets with a recently developed topic modeling approach that is appropriate for short texts and clusters respondents (stLDA-C), and we will examine how topics and clusters vary across social class, race/ethnicity, and age. Preliminary results suggest greater heterogeneity in parenting styles than can be described by the dominant binary categorization, particularly with multiple nuanced versions of “concerted cultivation.” We expect to refine these findings on parenting approaches, understand if they vary by sociodemographic characteristics, and demonstrate how computational social science can be used to study long standing “traditional” social science questions.

Principal Investigators