On Strategy for Methodological Analysis

Publisher Website
Google Scholar
Abstract
Manski (1993) recommends a strategy for methodological analysis built on the notion that the ''top-down'' mathematical study of identification, not the more conventional ''bottom-up'' study of statistical models and inference, is the fundamental problem in the social sciences. He illustrates with identification analyses of four key methodological issues: (I) extrapolation (or prediction), (2) selection bias (or inference with missing data), (3) modeling relationships with endogenous social factors as predictors (hierarchical models for cross-level inferences), and (4) modeling the link between attitudes (or intentions) and behavior. His approach relies on nonparametric methods that give bounds on the expectations, probabilities, quantiles, or contrasts that would be estimated from sample data. We recommend Manski's article to all social researchers and methodologists. But we believe that statistical issues broadly conceived are just as pressing as the identification problem and cannot be separated from it.

This publication has 0 references indexed in Scilit: