Social-Structural Differences in Identification with Europe

Abstract

Most research on European identity in political science and sociology either applies a theoretical and/or normative framework or concentrates on the magnitude of identification. This study goes beyond this by exploring the socio-structural conditions of European identity. Instead of bluntly asking how many identify with Europe, we look at social stratification as a potential cause of identification with Europe. Using Eurobarometer data from more than 30 countries, we report considerable differences in identification with Europe according to almost every social divide. Characteristics such as gender, age, education, area of living and social class play important roles in this respect. Our research further shows different divides in European identity across different European countries. This emphasizes that in order to better understand European identity in the future we should examine its background rather than the magnitude and allocate resources to the cross-national study of European identity.

Key Words:

• European identity
• Eurobarometer
• social stratification
• comparative quantitative analysis

Notes

A search for ‘European identity’ or ‘identification with Europe’ alone results in 216 publications.

Alternatively, the search was specified to ‘identification with Europe’.

Alternative searches for ‘urban’ and ‘rural’ instead of ‘region’ yield three and one hits respectively.

To mark the huge neglect, we run a search for ‘social class’ alone. This yields 6518 published works in Sociological Abstracts!

The search was undertaken on 20 February 2007.

Reported percentages are based on weighted data. The weight is the country design weight, adjusting for differences between the sample composition and population statistics.

Children in Europe enter school education at different age, most likely between five and seven years of age. The age of seven has thus been specified as the school entry age. A descriptive analysis shows that a very small minority of respondents (14 individuals) say they have finished full-time education at the age of seven or under. These few cases are treated as outliers (suspicious information) and thus excluded from subsequent analysis.

Because we were not sure how to treat this occupational position due to potential large differences between business such as shops or craftsmen, we did not include this category into another social class. Subsequent analysis, however, suggests treating self-employed/owners similar to the intermediate class.

From the original categories in Eurobarometer 64.2, we categorized the proxy of social class as follows:

• (a) Salariate (professional/managerial): professional; business proprietors, owner (full or partner) of a company; employed professional, general management, director or top management; middle management, other management.

  (b) Self-employed/owners: Owner of a shop, craftsmen, other self-employed person.

  (c) Intermediate class: employed position mainly at a desk; employed position not at a desk but travelling; employed position not at a desk but in a service job; supervisor.

  (d) Working class: skilled manual worker; other (unskilled) manual worker, servant.

  (e) Agriculture: farmer; fisherman.

  (f) Never in paid work: none of these occupational positions at present or in the past.

This is because the presented parameters in Model 3 are now only valid for Germany (the reference country), whereas they showed the overall effect in Model 2 covering all 30 countries.

The effect sizes of social structure variables for Germany are the main effects presented in Model 3 in Table 2.

Because gender had a different impact only in France, where men were much more likely to identify with Europe than women compared to other countries, a diagrammatical presentation has been skipped for gender interactions.

National differences concerning the effect of living in large towns or cities are presented in the figure. National differences concerning living in small or middle-sized towns are not present but included in the model. The latter are less articulate.