Center on Children and Families

FamilyScape

The Family Formation Module for the Social Genome Project, FamilyScape, is a cutting-edge policy simulation model of pregnancy and family formation. It was created by CCF researchers Adam Thomas, Isabel Sawhill and Emily Monea in collaboration with Metascape. The model uses an agent-based computational modeling approach to simulate the key antecedents of pregnancy (i.e., sexual activity, contraceptive use, and female fecundity) and many of its most important outcomes (e.g., childbearing within and outside of marriage, children’s chances of being born into poverty, and abortion).

The model is populated with individuals (or "agents") whose demographic characteristics are consistent with the characteristics of the members of a "real-world," nationally-representative dataset. These individuals are heterogeneous, which is to say that each agent is assigned a set of demographic and behavioral characteristics that help to govern the decisions that he or she will make over the course of the simulation: Who do the individuals hang out with? Whom will they choose to have sex with and will they use protection? If two individuals choose to have sex, what is the likelihood that the woman gets pregnant? If she gets pregnant, what are the odds that she will choose to have an abortion or to give birth? The model tracks these outcomes on an annual basis, and they can be used to produce aggregate estimates of phenomena of interest, such as rates of pregnancy, childbearing, and abortion among teens, unmarried women, married couples, and so forth.

The model is validated by comparing these aggregate outcomes to their real-world counterparts. Especially among the unmarried population, FamilyScape generally does a good job of replicating real-world phenomena of interest. FamilyScape can be used to conduct policy simulations by changing the features of the model that govern dynamics such as sexual behavior and contraceptive use under the assumption that a given policy has a particular effect on behaviors of this sort.

Thus far, FamilyScape researchers have written several papers documenting the model's architecture and detailing the results of a set of benefit-cost policy simulations. We have also presented the model and our simulation results to audiences of policymakers, policy researchers, modelers, and other experts and practitioners. The model has also been used as one of the starting points for the Social Genome Project. In addition to conducting new policy simulations with FamilyScape, we will continue to refine and enrich the model in the coming months and years within the framework of the Social Genome Project.

FamilyScape Research, Commentary, and Related Resources