Diverse Workplaces and Interethnic Friendship Formation-A Multilevel Comparison across 21 OECD Countries

Bibliographic Data

ID3609922
AuthorsAndrej Kokkonen (0000-0001-8127-0800, corresponding author), Peter Esaiasson (0000-0001-8572-5462), Mikael Gilljam
Year2015
Volume41
Issue2
Pages284-305
Publication date2015-01-28
Peer ReviewedYes
Open AccessNo
TypeARTICLE
VenueNew Community (JOURNAL)
Journal identifiersISSN: 0047-9586 • E-ISSN: 1469-9451
PublisherInforma UK Limited (PUBLISHER • GB)
DOI10.1080/1369183x.2014.902300
OpenAlexW2001145733
LanguageEN
Citations received27
References cited66

We study how workplace diversity affects how native-born populations form interethnic friendship relations. Drawing on network elasticity research, the relationship between functional and cognitive interdependencies, and intergroup contact theory, we argue that diverse workplaces—because they restrict individuals' opportunities to act on tendencies towards homophily more than other social units—have a particularly strong potential for integration. We test our argument by analysing data from the first round of the European Social Survey with multilevel modelling. In line with expectations, we find that individuals who work at diverse workplaces are more likely than individuals who work at homogenous workplaces to have immigrant friends; that workplace diversity is more efficacious than neighbourhood diversity in generating interethnic friendship relations; and that the association between workplace diversity and the likelihood of having immigrant friends is stronger among individuals who have a low level of education, feel economically vulnerable, and who live in countries characterized by relative resource scarcity.Keywords: IntegrationWorkplacesFriendshipEthnic RelationsContact HypothesisConflict Hypothesis Notes[1] Previous research has shown that the effects of having 'some' and 'many' immigrant neighbours on having immigrant friends do not differ significantly (Semyonov and Glikman Citation2009). Our own estimates confirm this finding (results available upon request).[2] The OECD provides standardised and purchasing power parity-adjusted figures for the GDP per capita that differ somewhat from those provided by the World Bank. However, the correlations between the different measurements is above 0.95 and models using the alternative OECD figures provide identical results to those presented in the main text.[3] All analyses have been carried out with the xtmelogit command in Stata 12/SE.[4] In this and subsequent analyses, the significance of the interaction effect has been tested with the Stata command Inteff. It shows a significant interaction for a vast majority of the observations in the data (the mean Z values for the interactions with upper secondary and tertiary education are -3.96 and -4.11, respectively).[5] The interaction effect is statistically significant for a vast majority of the observations in the data (the mean Z value is 3.45).[6] The effect is statistically significant for all observations in the data (the mean Z value is -11.35)

Demographic economics · Economics · European Social Survey · Friendship · Homophily · Immigration · Interdependence · Interpersonal ties · Multilevel model · Per capita · Political science · Population · Social science · Sociology · Demography · Psychology · Social and Cultural Dynamics · Social and Intergroup Psychology · Social Capital and Networks · Social Psychology

Tools

Open DOISci-Hub