This thesis involves three essays that explore the theory and application of econometric analysis to labour market interventions. One essay is methodological, and two essays are applications. The first essay contributes to the literature on inference with data sets containing within-cluster correlation. The essay highlights a problem with current practices when the number of clusters is 11 or fewer. Current practices can result in p-values that are not point identified but are instead p-value intervals. The chapter provides Monte Carlo evidence to support a proposed solution to this problem. The second essay analyzes a labour market intervention within Canada--the Youth Hires program--which aimed to reduce youth unemployment. We find evidence that the program was able to increase employment among the targeted group. However, the impacts are only present for males, and we find evidence of displacement effects amongst the non-targeted group. The third essay examines a set of Graduate Retention Programs that several Canadian provinces offer. These programs are aimed at mitigating future skill shortages. Once the solution proposed in the first essay is applied, I find little evidence of the effectiveness of these programs in attracting or retaining recent graduates. / Thesis (Ph.D, Economics) -- Queen's University, 2013-07-05 15:56:33.805

The present study provides an example of implementing the difference in differences (DD) estimator for a two-group, pretest-posttest design with K-12 educational intervention data. The goal is to explore the basis for causal inference via Rubin's potential outcomes framework. The DD method is introduced to educational researchers, as it is seldom implemented in educational research. DD analytic methods' mathematical formulae and assumptions are explored to understand the opportunity and the challenges of using the DD estimator for causal inference in educational research. For this example, the teacher intervention effect is estimated with multi-cohort student outcome data. First, the DD method is used to detect the average treatment effect (ATE) with linear regression as a baseline model. Second, the analysis is repeated using linear regression with cluster robust standard errors. Finally, a linear mixed effects analysis is provided with a random intercept model. Resulting standard errors, parameter estimates, and inferential statistics are compared among these three analyses to explore the best holistic analytic method for this context.