Experimental data

Abstract: A general inductive approach for analysis of qualitative evaluation data is described. The purposes for using an inductive approach are to (a) condense raw textual data into a brief, summary format; (b) establish clear links between the evaluation or research objectives and the summary findings derived from the raw data; and (c) develop a framework of the underlying structure of expe- riences or processes that are evident in the raw data. The general inductive approach provides an easily used and systematic set of procedures for analyzing qualitative data that can produce reliable and valid findings. Although the general inductive approach is not as strong as some other analytic strategies for theory or model development, it does provide a simple, straightforward approach for deriving findings in the context of focused evaluation questions. Many evaluators are likely to find using a general inductive approach less complicated than using other approaches to qualitative data analysis.

Abstract: The last two decades have seen a notable growth in the use of qualitative methods for applied social policy research. Qualitative research is now used to explore and understand a diversity of social and public policy issues, either as an independent research strategy or in combination with some form of statistical inquiry. The wider use of qualitative methods has come about for a number of reasons but is underpinned by the persistent requirement in social policy fields to understand complex behaviours, needs, systems and cultures.

Abstract: Semiparametric methods are developed to estimate the bias that arises from using nonexperimental comparison groups to evaluate social programs and to test the identifying assumptions that justify matching, selection models, and the method of difference-in-differences. Using data from an experiment on a prototypical social program and data from nonexperimental comparison groups, we reject the assumptions justifying matching and our extensions of it. The evidence supports the selection bias model and the assumptions that justify a semiparametric version of the method of difference-in-differences. We extend our analysis to consider applications of the methods to ordinary observational data.