Discovering neighbors: the regional migration experience as a source of intergroup contact

ABSTRACT

Contact Theory suggests that interaction across ethnic lines can erode prejudices and lessen interethnic conflict. Hypotheses drawn from the theory have a mixed empirical record, however, with negative results commonly attributed to the presence or absence of conditions that mediate contact’s effects. Analyzing survey data collected in the Balkans, we test Contact Theory’s observable implications using a case that offers unique theoretical leverage: Albanian attitudes toward Greeks. It is an especially appropriate case due to the historical isolation of Albanians from their Greek neighbors, as well as the elite-level cultivation of anti-Greek prejudice that took place during that period. And it is an especially tough test for Contact Theory due to the presence of multiple circumstances thought to weaken (if not reverse) the estimated benefits from contact. Nevertheless, our results largely support Contact Theory. Albanian respondents who’ve personally met at least one Greek feel more warmly toward the group than those who have not, and that warmth tends to be greatest among Albanians who spent time living in Greece. The benefits of contact do not appear, however, among those who consider Greeks to be a national-security threat, supporting the belief that ideological or instrumental motives can neutralize contact’s ameliorative effects.

KEYWORDS:

• Contact Theory
• prejudice
• feeling thermometer
• Albanian diaspora
• Balkan migration

Acknowledgements

The authors thank Kujtim Bytyqi for assistance on a much earlier incarnation of this research. Members of UC-Berkeley’s Immigration Workshop, especially Taesoo Song, René D. Flores, and workshop organizer Irene Bloemraad, provided solid feedback during the revision process. We also thank the student interviewers, who often paid for their field expenses, namely Habibe Ademi, Rrezart Dema, Arton Demolli, Erenik Dujaka, Lekë Hoxha, Valmira Hoxha, Blerina Islami, Ardit Konjufca, Arbëresha Mehmetaj, Blerta Morina, Gazmend Obrazhda, Drenushë Osmani, Brigita Rexha, Gentiana Sahiti, Vildane Xhemajli and Blerina Zeqiri. All remaining weaknesses are ours.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We acknowledge that group boundaries help delimit variable expectations about human behavior, and that because those expectations are socially constructed, they can serve an integrative rather than an exclusionary function under some circumstances (Bloemraad et al., 2023, pp. 5–6). Fear of the unknown, while common, is not necessarily universal. Nevertheless, nationalities – tied as they are to historical country borders – will have high perceived continuity, accentuating the intergroup dynamics discussed here (Warner et al., 2016).

2 Note that negative stereotypes can persist despite an increase in warmth toward a group, and the reverse is true as well (Reny & Manzano, 2016). These are probabilistic connections, and highly stochastic. The generation of intergroup conflict, and also the mitigating effects of intergroup contact, depend in part on context (Bloemraad et al., 2023, p. 16).

3 As with Contact Theory, the Group Threat Approach has been falsified in multiple settings (Carsey, 1995), including in a series of articles by Stephen Voss (Lublin & Voss, 2000, 2002; Voss, 1996; Voss & Lublin, 2001; Voss & Miller, 2001). The outcome, as with Contact Theory, has been a research agenda focused on specifying when and where the perspective applies (Fossett & Kiecoult, 1989; Giles, 1977; Zingher & Thomas, 2014). A few scholars have offered what amounts to synthesis between the Contact and Group Threat theories (Baybeck, 2006; Forbes, 2004; Liu, 2001; Voss, 2000a; Voss & Wallace, 2024).

4 Whereas weights should be used for descriptive inference, no consensus exists as to whether they should be used for predictive inference, such as when running regression analysis (Kott, 2007; Winship & Radbill, 1994). We use probability weights to compute descriptive statistics but not for regression models.

5 Merely having met one individual, of course, is a fairly low threshold, and leaves up to the respondent the judgment of how sustained or how intimate the contact needed to be to qualify as an actual meeting. In this respect, the question parallels how survey researchers have captured contact with ‘homosexuals’ (Overby & Barth, 2002, p. 439). It’s asking whether the respondent has at least one real-life person available in memory to attach to the abstraction ‘the Greeks.’

6 If anything, therefore, the presence of a Greek minority ought to make our test of Contact Theory more conservative. Having the Greek minority ‘close but not touching’ could create a ‘siege mentality’ effect, serving to exacerbate rather than mitigate stereotyping (Voss, 2000a; Voss & Wallace, 2024).

7 These institutional-trust variables could have conditioned the effect of contact, because of how they resemble contingencies found in Contact Theory. Greek-Albanian efforts focused on the EU could represent the sort of common enterprise that enhances contact’s benefits, while trusting current leaders could mean the respondent is less mired in past anti-Greek socialization able to neutralize contact. We did check to see whether contact interacted with those two variables, but ultimately it did not (analysis not shown).