Department of Statistics

Case study: Making better policy decisions about children

Does good-quality pre-school education overcome disadvantage in early life?

The Statistical Consulting Centre's Jessica McLay is helping one of her clients, the Centre for Methods and Policy Application in the Social Sciences (COMPASS), build a computer-based simulation tool that will help policy-makers consider these sorts of questions.

Visit the Centre for Methods and Policy Application in the Social Sciences (COMPASS) website

The idea, says project leader Peter Davis, is that decision-makers will be as well-informed as possible when responding to issues concerning children and young people.

The five-year project, called Modelling the Early Life Course (MELC) is founded on a representation of a birth cohort using data from the Christchurch Health and Development Study, which has been tracking 1,265 children born in 1977 and has gathered 50 million pieces of information.

Jessica's job is to distil facts about that cohort into statistical models that quantify what factors might determine certain outcomes in a child's life, among those indices housing quality, health-service use and educational attainment. In the future, by slicing and dicing data in a particular way, tool users will be able to test assumptions and what-if scenarios.

Jessica is using the software programmes R, SAS and Excel to build the model, and says she enjoys the work for its complexity, writing code to do the simulation and the challenge of overcoming 'how-to' problems. "In this project, I've also had the opportunity to learn new statistical modelling methods."

MELC is funded by the Ministry of Science and Innovation.