The Economic Crisis and Social Cohesion in Spanish Society

Funded by the Center for Sociological Investigation, 2014-2016

Principal Investigator: Miranda J. Lubbers

It is well known that the economic crisis does not have a homogeneous effect in Spanish society: it affects some parts of the population more heavily than others. However, this is most likely not only true for people personally, but also for their social environments. In other words, some people may know quite a few others who are affected by the crisis (experiencing unemployment, poverty or eviction) and others may know none. With this study, we intend to estimate the extent to which the consequences of the economic crisis in Spain are clustered in the personal networks of individuals, and whether the over- or underrepresentation of these consequences in individuals' networks is related to individuals´ own socio-economic characteristics. Furthermore, we hypothesize that the clustering of consequences of the crisis in personal environments affects individuals´ trust in public institutions, interpersonal trust, and the availability of informal social support. The latter might question the extent to which networks can compensate for economic hardship. 

Our research questions are:

1. To what extent can we observe an accumulative effect of the consequences of the economic crisis (e.g., unemployment, migration) in the social environments of individuals? In other words, do individuals vary in the extent to which their network members have experienced the negative consequences of the crisis? We distinguish between the larger social environment (the acquaintanceship network) and more intimate subsets of family and friends.

2. To what degree is the clustering of these consequences in the personal environments related to the socio-economic characteristics of the respondents? For example, do people with low incomes know more people who are affected by the crisis? And do young, highly educated people know more others who have migrated to other countries in order to find a job?

3. To what degree is the clustering of these consequences in the personal environments related to the trust individuals have in their compatriots and in public institutions and to the availability of social support? Does this clustering explain additional variance in these variables, on top of the variance explained by socio-economic characteristics?

In order to answer these questions, we have designed a module for the Special Barometer that was administered by the Center of Sociological Investigation (CIS) to a random sample of approximately 2,500 Spaniards in December 2014 and January 2015.

The measures that were used to estimate the clustering effect are based on the innovative Scale-Up method originally proposed by Killworth, Bernard, and McCarty (Killworth et al., 1998, 2003, 2006; McCarty et al., 2001, Bernard et al., 2010) and further developed by Zheng et al. (2006), McCormick et al. (2010), DiPrete et al. (2011) and Feehan & Salganik (2016). The Scale-Up Method was designed to estimate the size of hard-to-count populations in representative random samples, based on estimates of the size of personal networks and the size of specific subpopulations in the personal networks. However, as Zheng et al. (2006), McCormick et al. (2010) and Diprete et al. (2011) argued, the information can also be used to estimate to what extent subpopulations are over- or underrepresented in the social environments of subpopulations. For this study, we have designed an instrument for the Spanish society based on these earlier studies.


Lubbers, M. J., Molina, J. L., & Valenzuela, H. (2019). When networks speak volumes: Variation in the size of broader acquaintanceship networksSocial Networks, 59, 55-69.

Abstract: Personal network researchers have extensively studied the characteristics and effects of individuals’ closest relationships, but they have paid much less attention to broader acquaintanceship networks, despite evidence that weak ties can also provide social support. In this paper we focus on one aspect of these networks: acquaintanceship volume. We estimate its distributional parameters for a large, representative sample of the general population of Spain, explore its variation across social groups as well as its implications for social support availability. We designed a survey instrument based on the Network Scale-Up Method and implemented it in a national survey in Spain. Our results suggest that Spaniards have approximately 530 acquaintances, with a large inter-individual variation, comparable to the estimates reported for the American population. Acquaintanceship volume vary with gender, age, education, and income. These differences are partially related to the unequal participation of social groups in voluntary associations, confirming the civic value of such associations, and in employment. Even with similar core network size, acquaintanceship volume increases the likelihood of having adequate social support available, suggesting that broader acquaintanceship networks also structure individual outcomes.     

[Second paper currently under review]

Open Data

The data, codebook and technical files are openly accessible via the Center for Sociological Research, Study 3036 "Social cohesion and Trust"




Campus d'excel·lència internacional U A B