Spring 2014 - 63783 - PA397C - Advanced Empirical Methods for Policy Analysis

Making Sense of Survey Data

This course will teach students to rigorously analyze survey data and write research briefs to present the results. Often researchers, policy analysts, and practitioners rely on survey data to learn about a population or program; yet too often the results are not analyzed rigorously to provide evidence of generalizability or impact. This course will teach students how to:

develop a survey instrument based on a solid theoretical framework,
sample a population to ensure representativeness and generalizability,
construct a testable research question that is theoretically motivated,
use Stata to analyze cross-sectional and longitudinal survey data,
develop variables, including indices, scales, categorical, linear, and binomial variables,
deal with missing data,
test significance of results,
conduct multivariate analyses and interpret coefficients,
use mediating and moderating variables appropriately,
understand the threats to validity,
write a research brief, and
present results orally.

The course is geared toward students who prefer an applied method of learning advanced empirical methods. Students will be provided access to three Child and Family Research Partnership (CFRP) datasets (described below) and required to use at least two of them over the course of the semester. Students will be expected to use Stata; limited guidance will be provided by the professor, but students will be expected to invest the time necessary to conduct their analyses.

In addition to other assignment, students will be assigned one research question developed by the professor and expected to analyze the data and present the results in a research brief. Students will also be expected to develop their own research question and analyze the data and present the results in writing and orally. Although the substantive area for the course will focus on families, the skills that students learn in the course are applicable to all substantive areas.