New Survey Methods: Tools to Dig for Gold

Surveys are widely used by scholars, companies, and public policymakers to generate invaluable insights. Despite the popularity of surveys, there are several biases that can affect the validity of self-reported data. In his inaugural address,

Martijn de Jong discusses how new survey methods can help to extract valid information from surveys. Several examples are presented that showcase the relevance of better research design and careful statistical modeling of the response process. In addition, De Jong addresses some commonly held perceptions about the ability to make causal inferences with survey data.

From the report:
Let’s go back in history a bit to see why survey researchers make allusions to causality. In an Econometrica paper from the 1970s, Goldberger defines a structural equation as one representing a causal relationship, as opposed to a relationship that simply captures statistical associations (Goldberger 1972). The article’s conclusion contains a fascinating sentence: 

“economic, sociological, psychological, and political theory all have something to say about the causal links…” (p. 999).........


Download (pdf): New Survey Methods: Tools to Dig for Gold

Source: RePub