Social Surveys as a Data Source for Country-Level Indicators |
|
| Session Organiser | Dr Marta Kolczynska (Institute of Philosophy and Sociology, Polish Academy of Sciences) |
| Time | Friday 19th July, 09:00 - 10:30 |
| Room | D24 |
Several fundamental social science concepts, such as social cohesion, social capital, solidarity, inequality, welfare, or legitimacy, are properties of groups and not of individuals. Some of these societal characteristics can be operationalized as aggregates of individual characteristics, estimated with survey data, and compared across countries and over time. For example, if social cohesion corresponds to the level of agreement in values, attitudes, and practices among individuals, survey data offer opportunities of describing the extent of this agreement or disagreement directly by comparing pairs of respondents. Still, such attempts at constructing country-level measures beyond simple means or proportions are rare. One exception is economic inequality, for which there exist several measures constructed via the aggregation of individual-level data, such as the Gini index, Theil index, percentile ratios, etc., and their properties are well understood.
This session invites methodological and substantive papers that use theory-informed country-level measures of social phenomena derived from survey data. Welcome contributions include, but are not limited to, papers that (a) make a case for a particular way of measuring a well-defined concept at the group (societal) level using survey data, (b) discuss different ways of measuring a group (societal) level characteristic using survey data and data from other sources, (c) use one or more measures of societal characteristics derived from survey data as dependent or independent variables in substantive models.
Keywords: survey data aggregation, country-level characteristics, comparative analyses
Mr David Buil-Gil (University of Manchester) - Presenting Author
Dr Angelo Moretti (University of Manchester)
Worry about crime is not homogeneously distributed across space. There are certain countries where people are more worried about crime than other countries. However, worry is also known to be unequally distributed across the regions in each country. In order to map worry about crime across countries, cross-national surveys are the most important source of information. These are often designed to record representative samples at a state level, and smaller geographical units (i.e. regions) are unplanned areas and suffer from small and unrepresentative samples. This research aims to produce reliable small area estimates of dysfunctional worry about crime at a regional level in Europe based on the European Social Survey 5 data, and to examine the macro-level predictors of worry about crime. Model-based small area estimation techniques allow producing precise and accurate estimates of parameters of interest in unplanned areas and domains; yet they are underutilised in the study of perceptions and emotions towards crime. A small area estimation approach to worry about crime allows a precise and accurate inter-regional comparison of its distribution across the European regions. Furthermore, model-based small area estimation is helpful to obtain information about significant area-level predictors of the variable of interest. We make use of the Spatial Empirical Best Linear Unbiased Predictor (SEBLUP), which borrows strength both from related and neighbouring areas. Our estimates show that Eastern and Southern European regions are the areas with highest proportions of citizens worried about crime. More specifically, Greek, Slovakian, Estonian, Lithuanian and Bulgarian regions have particularly high proportions of citizens worried, as well as certain Portuguese, Spanish and Southern-French areas. At the other end, most regions in Central Europe and Scandinavia show the lowest estimates of worry, especially Dutch, Norwegian, Swedish, Danish and Finish regions. Unemployment is the best predictor of worry about crime.
Dr Oriane Sarrasin (University of Lausanne) - Presenting Author
Professor Eva G. T. Green (University of Lausanne)
Dr Jasper Van Assche (Ghent University)
Conservative vs. progressive ideological climates (measured with aggregated survey ideology items or referenda results) are known to influence individuals’ social and political attitudes beyond the impact of personal-level factors. For instance, individuals with liberal values have been found to express higher levels of anti-immigrant prejudice in conservative countries, regions or districts than individuals with similar values living in more inclusive contexts. Research has so far operationalized normative climates as the average ideologies in a given place, giving no indication of the extent to which they are shared. This is surprising, since cohesive and widely shared beliefs have a powerful impact on individuals’ attitudes and behaviours. To examine in which contexts heterogeneous vs. homogeneous ideological climates emerge, the present study relied on data from the 2013 International Social Survey Programme to examine to which extent individuals living in 188 regions of 21 European countries share a similar conception of who is a “true” member of the nation. With this goal in mind, indicators of regional heterogeneity of ethnic (e.g., having national ancestry) and civic (e.g., respecting national institutions and laws) conceptions of nationhood were built. Multilevel analyses revealed that a greater heterogeneity in endorsement of ethnic criteria was found in countries with more inclusive policies (i.e., which had an inclusive citizenship regime or which granted more rights to immigrants). Civic criteria were deemed in average more important, which explains the lower heterogeneity in endorsement and the lack of impact of national policies. Overall, these findings suggest that inclusive settings leave more room for individuals to express their personal worldviews resulting in heterogeneous ideological climates.
Dr Steven Marshall (Welsh Government) - Presenting Author
In 2015 the Welsh Government put in place a landmark piece of legislation: The Well-being of Future Generations (Wales) Act. It is aimed at improving the wellbeing of Wales by ensuring that public bodies think more about the long-term and adopt new ways of working.
The Act required Welsh Ministers to set national indicators to help measure the progress. The indicators need to reflect the inter-related dependencies across different domain of wellbeing (seven wellbeing goals are defined in the Act). Social surveys are ideal for considering such inter-relations.
Up until 2015, the Welsh Government and its sponsored bodies conducted five large-scale social surveys: the National Survey for Wales, Welsh Health Survey, Active Adults Survey, Arts in Wales Survey and the Welsh Outdoor Recreation Survey. A review of the way social surveys were carried out in Wales concluded that a more sustainable approach was to bring together the five surveys into a single new National Survey, and this approach has been in place since 2016. The survey uses a random sample, face-to-face approach and has around 12000 responses each year.
The new approach to social surveys in Wales came in at the ideal time to be used as a major source for the new Wellbeing Indicators and the National Survey for Wales provides the data for 15 of the 46 indicators. It also allows a number of subjective measures to be included in the indicator set. The unusually wide range of topics included in the survey and the size of the dataset allows detailed statistical analysis such as regression and latent class analysis to be carried out. A range of outputs have been produced using these techniques with the aim of helping public bodies understand how their activities could impact on the national indicators.