Linking National Surveys, Administrative Records and Mass Media Content: Methodological Issues of Constructing the Harmonized Data-File.

This paper discusses the challenges of construction of the harmonized data-file that links data from three sources: national surveys, administrative records, and the media. The basis of the data-file comes from 22 well-known international survey projects containing questions on protest behavior. It consist of 1721 national surveys covering 132 countries. The data from administrative country-level records on population size, ethnic fractionalization, GDP and other characteristics, as well as media content (e.g. Reuters’ event data on protest) are incorporated into the integrated data-file. From the methodological point of view, there are a number of challenges to overcome for reaching the aim of the project: building the integrated data-file. In the presentation we concentrate on proposed ways of linking data for multi-level analyses, with countries and years as macro-levels. We discuss data quality on both the micro- and macro-levels, and some aspects of secondary data usage of survey and non-survey data together. The logic of data linkage and data processing procedures are of general nature and can be applied to other comparative projects. The paper is a part of the project “Democratic Values and Protest Behavior: Data Harmonization, Measurement Comparability, and Multi-Level Modeling in Cross-National Perspective”, financed by the Polish National Science Centre (2012/06/M/HS6/00322), located at the Polish Academy of Sciences and The Ohio University.