The Harmonization Project

by Irina Tomescu-Dubrow and Kazimierz M. Slomczynski, Polish Academy of Sciences and CONSIRT

The Democratic Values and Protest Behavior: Data Harmonization, Measurement Comparability, and Multi-Level Modeling study is financed by the (Polish) National Centre of Science and supported by The Ohio State University. CONSIRT hosts the project in Poland. While there are a number of survey data harmonization projects that have informed our own, each with their own acronyms (Dubrow and Tomescu-Dubrow 2014), we have come to call this large-scale research, simply, the Harmonization Project.

To test this model we need data at both the individual- and the country-level that vary over time and across space. The Harmonization Project sets out to create comparable measurements of political protest, social values, and demographics via ex-post harmonization of variables from international survey projects and append them with macro-level variables from external sources such as the World Bank, OSCE, UN agencies, Transparency International, and others.

Substantively, the project engages with the relationship between democracy and protest behavior in comparative, cross-national perspective. Political protest can be of various types, such as participation in demonstrations, signing petitions, or contacting politicians. Drawing on extant research (Benson and Rochon 2004; Kriesi 2004; Dubrow, Slomczynski and Tomescu-Dubrow 2008; Dalton Sickle and Weldon 2009; Marien, Hooghe and Quintelier 2010; Vrablikova 2013), we develop a two-level model where protest (individual-level) is explained by a set of theoretically-informed characteristics of people and countries in which they live (country-level), and cross-level interactions.

Table 1. Selected International Survey Projects.

Note: In this table EVS and WVS are joined in one row because they share one data file. The total number of cases  refers to all cases in source data files.

We selected 22 well-known international survey projects – listed in Table 1 – that span almost 50 years (1966-2013) and a total of 142 countries or territories.¹ In all survey projects, the units of observations are individuals. We took into account only projects designed primarily for academic use and with coverage of at least three countries. The data from selected projects are in the public domain, either in social science data archives or projects’ own webpages that are open to scholars. Documentation of these projects is in English. Surveys contain political (e.g. protest), demographic (e.g. gender and age) and social stratification (e.g. education) items, but vary somewhat in their content and form.

From the selected projects, we pooled 81 data files, with 89 waves, into a relational database. It is a database containing 1726 national samples for which interviews were conducted in all waves (project*wave*country). All these surveys cover a total of almost 2.3 million respondents.² The platform for data files of national surveys is organized such that in the future any variable could be extracted and moved to the virtual integrated dataset (see Powałko 2014, and in this Newsletter).

We identified relevant original (source) variables that appear in at least five of the survey waves. Using various data processing procedures we produce, in the database, common (target) variables according to a unified measurement scheme. This scheme is well grounded in the past important discussions on ex-post harmonization (Gunther 2003; Minkel 2004; Ehling, Rendtel, et al. 2006; Granda and Blasczyk 2010; Granda, Wolf and Hadron 2010).

We select two types of source variables for harmonization: technical variables, provided by survey administrators, and variables of substantive interest. The list of variables is not closed, thanks to the flexible set-up of the programming environment we are using.

The Harmonization Project is work in progress. As it unfolds, it prompts us to reconsider how existing survey data can best be used in the harmonization framework by including controls of various quality aspects of existing surveys and harmonization procedures. We construct quality controls of the general survey documentation, the specific data description, and original data in the computer files. In addition, we apply quality control to specific harmonization procedures that could influence validity and reliability of the target variables. We suggest that quality-control variables for each of these aspects be included in substantive analyses (see Slomczynski and Tomescu-Dubrow in this Newsletter). Their relevance has to be empirically assessed.

Notes

• We refer to the selected projects as well-known on the basis of publication records and the impact that they have on the social science disciplines. For practical reasons, we stopped adding new data in the second quarter of 2014.
• Because of the thematic coverage criterion, we include only survey waves that contain relevant questions on protest behavior and/or democratic values; thus, not all waves of ISSP, EB and CNEP are in our data.

Irina Tomescu-Dubrow is Associate Professor at the Institute of Philosophy and Sociology, Polish Academy of Sciences. She is Program Manager at Cross National Studies: Interdisciplinary Research and Teaching Program (CONSIRT), of the Polish Academy of Sciences and The Ohio State University.

Kazimierz M. Slomczynski directs CONSIRT. He also directs the Polish Panel Study 1988-2013 (POLPAN), a unique panel survey on the transformation of the Polish social structure.

References

Benson, Micelle and Thomas R. Rochon. 2004. “Interpersonal Trust and the Magnitude of Protest: A Micro and Macro Level Approach.” Comparative Political Studies 3 (4): 435-457.

Dalton, Russel J., Alix van Sickle, and Steven Weldon. 2009. “The Individual-Institutional Nexus of Protest Behaviour.” British Journal of Political Science 40(1): 51-73.

Dubrow, Joshua Kjerulf., Kazimierz M. Slomczynski, and Irina Tomescu-Dubrow. 2008. “Effects of Democracy and Inequality on Soft Political-Protest in Europe: Exploring the European Social Survey Data.” International Journal of Sociology 38(3): 36 – 51.

Dubrow, Joshua Kjerulf and Irina Tomescu-Dubrow. 2014. “A History of Cross-national Survey Data Harmonization Projects in the Social Sciences: Emergence of an Interdisciplinary Methodological Field.” CONSIRT Working Paper Series 1 (CONSIRT Lab: Methodology of Survey Data Harmonization) at consirt.osu.edu.

Ehling, Manfred and Ulrich Rendtel, et al. 2006. Synopsis. Research Results of Chintex – Summary and Conclusions: www.destatis.de/DE/Methoden/Methodenpapiere/Chintex/ResearchResults/Downloads/Synopsis.html [Last accessed November 28, 2014].

Gelman, Andrew and Jennifer Hill. 2007. Data Analysis Using Regression and Multilevel/Hierarchical Models. New York: Cambridge University Press.

Granda Peter and Emily Blasczyk. 2010 Cross-Cultural Survey Guidelines. XIII. Data Harmonization. http://ccsg.isr.umich.edu/harmonization.cfm [Last accessed November 24, 2014].

Granda Peter, Christof Wolf, and Reto Hadorn. 2010. “Harmonizing Survey Data.” Pp. 315-332 in Janet A. Harkness, Michael Braun, Brad Edwards, Timothy P. Johnson, Lars Lyberg, Peter Ph. Mohler, Beth-Ellen Pennell, and Tom W. Smith (eds.), Methods in Multinational, Multicultural and Multiregional Contexts. Hoboken, NJ: John Wiley & Sons.

Gunther, Ronald. 2003. “Report on Compiled Information of the Change from Input Harmonization to Ex-post Harmonization in National Samples of the European Community Household Panel – Implications on Data Quality.“ Wiesbaden: Statistisches Bundesamt – Working Paper 19.

Kriesi, Hanspeter. 2004. “Political Context and Opportunity.” Pp. 67-90 in David A. Snow, Sarah A. Soule and Hanspeter Kriesi (eds.) Blackwell Companion to Social Movements. Malden:  Blackwell.

Marien, Sofie, Marc Hooghe, and Ellen Quintelier. 2010. “Inequalities in Non-Institutionalized Forms of Political Participation: A Multi-level Analysis of 25 Countries.” Political Studies 58(1): 187-213.

Minkel, Hartmut. 2004. “Report on Data Conversion Methodology.” CHINTEX Working Paper  20.

Powałko, Przemek. 2014. “Working with Big Data: Experiences with the Cross-National Survey Data Harmonization Project.” CONSIRT Working Papers Series (CONSIRT Labs: Methodology of Survey Data Harmonization) at consirt.osu.edu.

Preacher, Kristopher J., Zhen Zhang, and Michael J. Zyphur. 2010. “Alternative Methods for Assessing Mediation in Multilevel Data: The Advantages of Multilevel SEM”. Structural Equation Modeling: A Multidisciplinary Journal 18(2): 161-182.

Schoene, Matthew and Marta Kołczyńska. 2014. “Survey Data Harmonization and the Quality of Data Documentation in Cross-National Surveys.” CONSIRT Working Papers Series (CONSIRT Labs: Methodology of Survey Data Harmonization) at consirt.osu.edu.

Vrablikova, Katerina. 2013. “How Context Matters? Mobilization, Political Opportunity Structure and Nonelection Political Participation in Old and New Democracies.” Comparative Political Studies 47 (5), on line first.