Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire

Abstract

Experimental studies on immigration attitudes have been conducted overwhelmingly in Western countries and have focused on immigrant admission and naturalization, neglecting deportation as a possible outcome. In a survey experiment in Côte d’Ivoire, where immigrants represent more than one-tenth of the population, we randomized attributes of hypothetical immigrants to determine which factors influenced respondents’ support for naturalization or deportation compared with staying in the country without citizenship. Support for naturalization was shaped by several expected economic and social attributes, while deportation preferences were influenced primarily by the immigrant’s legal status and level of savings. Cultural proximity produced mixed results, with respondents less likely to support the naturalization of immigrants from neighboring African countries but also less likely to deport immigrants with whom they shared a religious faith. Finally, respondents were more likely to support the naturalization of immigrants who planned to vote if granted citizenship, especially when they were of the same religion as the respondent, indicating a degree of electoral calculation in a context where voting patterns are associated with religious identities. Together, these findings suggest that citizen preferences for naturalization and deportation are influenced by somewhat different factors, a possibility that warrants further testing in other contexts.

Keywords

Côte d’Ivoire immigration attitudes naturalization deportation voting

Immigration is a contentious topic in many countries. Broadly speaking, attitudes are shaped by sociotropic concerns about the economic and cultural impact of immigrants (Hainmueller and Hopkins 2014), though there are mixed findings about the influence of specific factors at the macro and individual levels. Existing research on immigration attitudes has been based overwhelmingly on countries in Western Europe and North America, where public opinion surveys are conducted regularly. Yet, debates about immigration are just as spirited in developing countries, especially those with high levels of immigration.

More than half of international migrants from the Global South live in other developing countries (World Bank 2016). Within Africa, two-thirds of international migrants go to other countries within the region. Their top destinations are Côte d’Ivoire and South Africa, with each hosting more than two million African immigrants (Gonzalez-Garcia et al. 2016; World Bank 2016, 2018). Immigration has been a divisive issue in both countries, at times to the point of violence. Nevertheless, while African countries have received ample attention from regional specialists (e.g., Adida 2011, 2014; Bah 2010; Geschiere 2009; Landau 2011; Mitchell 2011; Neocosmos 2010; Nyamnjoh 2006; Whitaker 2015; Whitaker and Giersch 2015), they have been largely overlooked in broader cross-regional studies of immigration attitudes.

This article makes several important contributions. First, we examine attitudes toward immigrants in a developing country. Côte d’Ivoire hosts more than 2.4 million immigrants, roughly 11 percent of its population (World Bank 2016), and recently experienced a conflict driven by issues of migration and citizenship. By randomly selecting 250 respondents in six research sites in different parts of the country, we gain a broad understanding of the factors shaping attitudes toward immigrants in that context. Second, we employ experimental methods that only recently have started to be used to study immigration attitudes in non-Western settings (Harris et al. 2017). By randomizing immigrant attributes like national origin, education, and voting intentions, we test existing hypotheses and introduce new variables that have not been examined elsewhere. Third, in evaluating hypothetical immigrants, we allow respondents to choose among three options: naturalization, remain in the country without naturalization, or deportation. This provides a more complete range of possible outcomes and allows us to simultaneously test which factors influence preferences for each.

Our analysis shows that Ivorian respondents’ support for naturalization was influenced by several expected economic and social attributes, including the immigrant’s ability to contribute to the economy and the extent of their connections to the host country. Deportation preferences, on the contrary, were shaped most consistently by the immigrant’s legal status and level of savings to start a business. In testing prevailing assumptions about cultural proximity, we obtained mixed results. Immigrants from neighboring African countries were less likely to be selected for naturalization compared with those from outside the region, but a shared religious faith between host and immigrant also reduced support for deportation. Finally, in a novel test of a political attribute of immigrants, respondents were more likely to support the naturalization of immigrants who planned to vote if granted citizenship, particularly when those immigrants shared a religious faith with the respondent. In a context where voting patterns are associated with religious identities, this finding suggests a degree of electoral calculation shaping immigration preferences. Together, these results demonstrate that different immigrant attributes may have differing effects on citizen support for naturalization versus their support for deportation. More research is needed to test this possibility in other contexts and to examine the factors that influence support for deportation in particular.

Explaining Attitudes toward Immigrants

Existing literature has identified a range of variables that influence immigration attitudes. Some studies highlight macro-level economic, demographic, and political factors (Coenders and Scheepers 2008; Hopkins 2010; Money 1997; Quillian 1995; Rydgren 2003), while others emphasize individual-level determinants of immigration preferences (Burns and Gimpel 2000; Hainmueller and Hiscox 2007, 2010; Mayda 2006; O’Rourke and Sinnott 2006; Scheve and Slaughter 2001; Sides and Citrin 2007; Sniderman, Hagendoorn, and Prior 2004). A growing body of literature focuses on specific characteristics of immigrants themselves, separating the question of whom to admit from that of how many to admit (Hainmueller and Hopkins 2015).

With few exceptions (Adida 2011; Gordon and Maharaj 2015; Harris et al. 2017; Whitaker and Giersch 2015), previous quantitative studies have focused overwhelmingly on immigration attitudes in Western countries (Money 2010), for which survey data are readily available. Even recent experimental studies have been conducted mainly in North America or Europe. The majority of global migration takes place within the developing world, however, raising the question of whether findings from Western countries apply in such settings.

Another shortcoming of existing literature is its emphasis on immigrant admission. Few studies explore citizens’ attitudes toward immigrant naturalization¹ or deportation, despite the fact that policy debates in many countries often center on these questions. An important exception is work by Hainmueller and Hangartner (2013), who took advantage of a unique system in Switzerland that allowed citizens to vote on the citizenship applications of individual immigrants to analyze attributes influencing those outcomes. Drawing on a survey experiment instead of a natural one, we similarly examine how different factors influence citizens’ attitudes toward naturalization. We go further, however, by allowing respondents to choose among three options for each hypothetical immigrant (naturalization, remain in the country without naturalization, or deportation), thus including a broader range of possible outcomes.

Our first cluster of hypotheses explores economic influences on immigration attitudes. Citizens generally prefer immigrants who are more likely to contribute to the host country’s economy. Experimental studies have found more support for immigrants with higher levels of education, more job experience, employment contracts, greater host-country language proficiency, and more skilled professions (Hainmueller and Hangartner 2013; Hainmueller and Hiscox 2010; Hainmueller and Hopkins 2015; Harell et al. 2012; Wright, Levy, and Citrin 2016). There is evidence that labor market competition influences attitudes, with less skilled workers and unemployed individuals generally more opposed to immigration (Mayda 2006; O’Rourke and Sinnott 2006; Scheve and Slaughter 2001). But highly educated people are more likely to support immigrants of all skill levels, even ones that will compete with them for jobs, suggesting that education nurtures certain values regardless of the labor market (Hainmueller and Hiscox 2007, 2010). Consistent with these findings, we expect Ivorian respondents to be more likely to support the naturalization of immigrants who can make clear contributions to the economy and less likely to support their deportation. In our experiment, this would include immigrants who are employed, have substantial financial savings, have a university degree, or speak French.²

A second cluster of hypotheses centers around a prospective immigrant’s relationship to the host country. Past research suggests that citizens prefer immigrants with deeper connections in the host country; studies find significantly more support for immigrants who have relatives who are citizens, were born in the country themselves, or have lived there for a longer period (Hainmueller and Hangartner 2013; Wright, Levy, and Citrin 2016). Legal status also has been found to be important in the United States, with citizens less likely to support immigrants who entered the country illegally (Hainmueller and Hopkins 2015; Wright, Levy, and Citrin 2016). In our experiment, we thus expect respondents to favor naturalization of immigrants with deeper roots in Côte d’Ivoire, including those who have an Ivorian parent or who have lived in the country for longer periods, and to prefer immigrants who entered the country legally.

Our third cluster of hypotheses explores how cultural proximity between immigrants and hosts influences attitudes. Studies from Western countries have found that citizens are more likely to support immigrants with whom they share common cultural attributes (Brader, Valentino, and Suhay 2008; Ford 2011; Hainmueller and Hangartner 2013; Hellwig and Sinno 2017), including a preference for Christian immigrants (Wright, Levy, and Citrin 2016). Experimental research has tended to focus on a prospective immigrant’s country of origin, with some studies finding significantly less support for immigrants from more distant countries or regions (Hainmueller and Hangartner 2013; Harell et al. 2012) and others showing little to no effect (Hainmueller and Hopkins 2015; Wright, Levy, and Citrin 2016).³ However, research findings from West Africa raise questions about the generalizability of the cultural proximity hypothesis. Using unique survey data from Benin, Ghana, and Niger, Adida (2011, 2014) shows that religious and ethnic similarities exacerbate host-immigrant relations instead of improving them. She reasons that community leaders on both sides highlight group boundaries to protect their own areas of authority and economic advantage. The contradictory findings from Western and African contexts suggest a need to investigate this issue further.

In our experiment in Côte d’Ivoire, we address the issue of cultural proximity in two ways. Borrowing from previous experimental studies, we vary the country of origin of prospective immigrants. Attitudes are likely to be influenced by the complicated history of immigration to Côte d’Ivoire (see below) and the prevalence of immigrants from certain countries. As a result, we expect Ivorians to be more reluctant to support the naturalization of immigrants from neighboring African countries (Ghana, Mali, and Burkina Faso), which have been the source of significant migration over the years. They also may have skepticism about immigrants from France, the former colonial power, and Lebanon, which already has a large expatriate community in the country. Respondents are expected to be most supportive of the naturalization of immigrants from less common source countries (China and the United States), who tend to be associated implicitly with foreign investment and business opportunities.

Innovating on previous experimental research, we also examine how the combination of immigrant and respondent attributes affects attitudes. More specifically, we explore whether a “match” between the hypothetical immigrant and the respondent on certain variables affects support for naturalization or deportation. Given recent political dynamics in Côte d’Ivoire, we hypothesize that citizens will be more likely to advocate naturalization for immigrants with whom they share a religious identification (whether Christian or Muslim) and less likely to prefer deportation of those same immigrants. We also expect Ivorian citizens whose parents or grandparents came from a specific country of origin to be more likely to support the naturalization of immigrants from that country.

A fourth cluster of hypotheses tests the degree to which the perceived number of immigrants from a particular country and an immigrant’s intended settlement location affect support for naturalization or deportation. Some past research shows that anti-immigrant attitudes rise as the proportion of migrants increases (Coenders and Scheepers 2008; Hopkins 2010; Quillian 1995), while other work finds no relationship between the size of the immigrant population and citizens’ attitudes (Hjerm 2007; Scheve and Slaughter 2001; Sides and Citrin 2007). Our design allows us to test whether a respondent’s perception that there are “too many immigrants” from a particular country affects their preference for naturalization or deportation of an individual from that country. By randomizing the intended settlement location of hypothetical immigrants among our main collection sites, we are also able to test the possibility of a “not in my backyard” effect. Craig and Richeson (2014) found such an effect at the national level, but we are not aware of any experimental study that has tested this hypothesis for individual immigration cases at the subnational level.

Finally, in our fifth cluster of hypotheses, we make a unique contribution to the literature by testing a political attribute: the immigrant’s intention to vote in future elections if granted citizenship. To our knowledge, explicit immigrant-level political variables have not been examined previously, though other political factors have been shown to influence attitudes. Research from Western countries highlights the role of radical right parties in legitimizing underlying hostility (Art 2011; Rydgren 2003, 2005) and the influence of media frames on citizens’ attitudes (Brader, Valentino, and Suhay 2008; Hopkins 2011). Within Africa, case studies have emphasized the use of anti-immigrant rhetoric by political elites in their competition for power (Geschiere 2009; Landau 2010; Mitchell 2011; Neocosmos 2010; Whitaker 2005). Along these lines, cross-national research finds that opposition to immigration is higher in African countries that are more democratic and where the survey is conducted closer to a national election (Whitaker and Giersch 2015).

With various political factors at play, immigrants’ voting intentions could have significant bearing on respondents’ support for naturalization. The question is in what direction. On one hand, citizens often encourage voting and other forms of civic engagement among fellow citizens by rewarding participation and applying selective sanctions for noncompliance (Rosenstone and Hansen 1993). To the degree that voting is viewed as a desired behavior, respondents may prefer to naturalize immigrants who intend to vote in future elections. Voting has been seen as an indicator of immigrant integration (Cho 1999; Ramakrishnan and Espenshade 2001), and citizens tend to have more positive attitudes toward more integrated immigrants (Van Oudenhoven, Prins, and Buunk 1998). On the other hand, an immigrant who intends to vote after receiving citizenship could be perceived as more of a political threat. Immigrants who cannot vote should not constitute much of a political threat at all (Hopkins 2010, 2011). Once they are naturalized, however, each additional vote effectively reduces the value of existing votes, which could be frustrating in a country where votes have only recently come to play a real role in the selection of leaders (Whitaker and Giersch 2015).

Nevertheless, not all additional votes would logically be perceived as a political threat, especially in a context where identity-based voting is common. To the degree that an immigrant shares a politically salient cultural trait with a citizen and might therefore be expected to vote in a similar way, naturalizing the immigrant would make electoral sense in terms of helping the citizen’s preferred side win elections. Therefore, we expect support for naturalization to be influenced by an interaction between politically salient cultural attributes and the immigrant’s intention to vote, with respondents especially likely to support the naturalization of culturally similar immigrants who intend to vote. This interaction should only affect support for naturalization, however, and not support for deportation, since voting in Côte d’Ivoire is limited to citizens. To the degree that respondents may still interpret intending to vote as a positive indicator of an immigrant’s character and civic engagement, though, this attribute may still independently decrease a respondent’s likelihood of supporting deportation.

Migration and Citizenship in Côte d’Ivoire

After Côte d’Ivoire gained independence in 1960, the government of Félix Houphouët-Boigny expanded the colonial-era practice of recruiting people from relatively poor neighboring countries, including Burkina Faso, Guinea, Mali, and Niger, to increase cocoa and coffee production (Daddieh 2001). Migrant labor thus fueled the Ivoirian Miracle, which generated annual growth rates above 10 percent and established the country as a leading cocoa producer (Mitchell 2011). But under the surface of economic success, tensions grew. As the foreign-born population rose to as much as 25 percent of the total in the late 1970s, many residents felt that immigrants and migrants from other parts of the country were reaping “the lion’s share of the benefits of economic development of the southwestern region” (Boone and Kriger 2010, 186). Despite a recession in the 1980s, large-scale migration into Côte d’Ivoire continued.

With the onset of multiparty elections in 1990 and the death of Houphouët-Boigny in 1993, the country was thrust into a period of uncertainty. Faced with a shortage of commodity profits and foreign aid, previously used to buy political loyalty, Henri Konan Bédié and his ruling party colleagues embraced the concept of ivorité to generate support (Crook 1997; Whitaker 2015). By defining ivorité as southern and Christian, it excluded not only foreigners but also northern Ivorians who were predominantly Muslim (Daddieh 2001). Laws were changed to limit eligibility for the presidency to native-born citizens whose parents were also born in the country. This was seen as a blatant effort to disqualify Alassane Ouattara, a popular northern Muslim politician, from running for president based on the claim that one or both of his parents were from Burkina Faso (Whitaker 2005).

After winning a controversial election in 2000, Laurent Gbagbo refused to hold a new round of voting in which Ouattara could compete. This angered Ouattara’s supporters, leading to a violent rebellion in 2002. The ensuing civil war effectively split the country into a predominantly Christian south, led by Gbagbo’s government, and a predominantly Muslim north, led by a rebel alliance. When Gbagbo finally allowed elections in 2010, both he and Ouattara claimed victory, sparking renewed violence. With external intervention, Gbagbo was arrested in 2011 and sent to the International Criminal Court for trial. Ouattara, widely seen as the rightful winner of the election, assumed the presidency. In 2015, Ouattara won his second five-year term with strong support in the north and “visibly lower” turnout in the south (Coulibaly and Bavier 2015). Although violence has abated, many of the underlying citizenship questions at the root of the conflict have yet to be fully addressed. Côte d’Ivoire thus offers a good example of a developing country in which immigration policies have been highly politicized, though questions remain about the factors that shape individual citizens’ attitudes.

Research Design

We conducted a survey experiment in Côte d’Ivoire from May 30 to June 26, 2016. Respondents were presented with the profile of a hypothetical immigrant with twelve randomized attributes. The full list of attributes and a sample profile can be found in the supplemental material. All values for each attribute were equally likely to be selected, with the exception of parents’ citizenship; hypothetical immigrants had a 50 percent chance of having no Ivorian parents, 25 percent chance of having an Ivorian citizen father, and 25 percent chance of having an Ivorian citizen mother. The only forbidden combination was an immigrant from France who does not speak French.⁴ After viewing the hypothetical immigrant’s profile, respondents were asked in French the following question:

On the basis of the above information, would you like

the immigrant to be granted Ivorian citizenship?

the person to be allowed to stay in Côte d’Ivoire, but not granted Ivorian citizenship?

the person to be deported to their country of origin?

Importantly, respondents were not forced to select naturalization or deportation for any specific number of hypothetical immigrants, as in a forced choice design. Although forced choice designs are well-suited to simulate situations in which only one option can be selected (as when voting) or only one option is likely to be selected (as when choosing among similar products to buy), they are not well-suited to measure preferences concerning immigrants. Immigration cases are decided individually. Forcing respondents to select for naturalization (or deportation) an immigrant that they do not support, simply because that immigrant is viewed as relatively better (or worse) than another randomly paired immigrant, would not reflect real world preferences. In their comparison of different experimental designs, Hainmueller, Hangartner, and Yamamoto (2015) find that the forced choice design overestimated the effects of several immigrant attributes on support for naturalization in Switzerland.

Instead, our survey experiment utilized a “single-profile vignette design, where a single immigrant profile is presented . . ., and . . . respondents are asked to accept or reject the applicant” (Hainmueller, Hangartner, and Yamamoto 2015, 2396). In our vignettes, immigrant attributes appeared as a list of complete sentences, and respondents were asked to choose among three options: naturalize, stay in the country without naturalization, or deport. The order of the sentences describing immigrant attributes was randomized for each respondent to avoid any order bias within the aggregate data. Each respondent was asked to repeat the exercise for a total of six hypothetical immigrants. After doing so, respondents completed an exit survey asking about their gender, age range, education level, religion, and employment status (see supplemental material). The exit survey also asked respondents to indicate whether they have parents or grandparents who immigrated from any of the countries of origin and their perception of the existing number of immigrants from each country. Our research design thus allows us to test the influence on respondents’ support for naturalization or deportation of multiple immigrant characteristics, respondent characteristics, and interactions among them.

Research sites were selected in Korhogo, Ferkéssedougou, Bondoukou, and Sikolo in northern Côte d’Ivoire, and Abidjan and Grand-Bassam in the south. Within each site, neighborhoods were randomly chosen, and within each neighborhood, random number sheets were used to geographically space out requests for interviews and to randomly select individuals present at the time. These efforts were taken to increase the representative nature of the sample and to limit selection bias. In addition, our exit survey allows us to control statistically for any respondent attribute being over- or underrepresented. Overall, 136 men (54%) and 114 women (46%) participated in the survey, with each evaluating six hypothetical immigrants. Our response rate exceeded 95 percent and only one person stopped the survey without finishing it. Respondents evaluated 1,498 hypothetical immigrants, but some respondents did not report all of their personal information, resulting in 1,450 observations with full data reported. Roughly half (52%) of the respondents were Christian, most reported having earned a high school (44%) or college (10%) diploma, and the majority (59%) were employed. In terms of location, 45 percent were surveyed in the South of the country and the rest were in the North. The supplemental material includes a map with data sites indicated and a table of respondent descriptive statistics.

Our experiment randomly varied the hypothetical immigrant’s country of origin from among seven possible countries. Based on the locations of our data collection sites, we included three countries that neighbor Côte d’Ivoire to the north (e.g., Mali and Burkina Faso) and the east (e.g., Ghana). Demographic shifts associated with high levels of immigration from these countries, especially the northern two, has contributed to Côte d’Ivoire’s recent social and political unrest. To avoid having too many levels of one particular attribute, the western neighbors of Guinea and Liberia were not included. We also included three nonneighboring countries of origin. France is the former colonizer and continues to have strong—and often controversial—influence in Côte d’Ivoire (Simonen 2012). Lebanon was included because there is a small, yet economically important, Lebanese community in Côte d’Ivoire dating back to the 1920s. With Lebanese merchants owning roughly one-sixth of retail businesses, despite constituting less than 1 percent of the population, “some Ivorian officials . . . have sought to employ the Lebanese ‘menace’ as a scapegoat for their own economic policy failures” (Bierwirth 1999, 83). China was included because of its increased levels of engagement with and migration to Africa (Brautigam 2009) and its relative lack of cultural proximity. Finally, we included the United States, whose expatriate community in Côte d’Ivoire is small and which historically has deferred to France in its former African colonies.

Empirical Results

Of the 1,498 hypothetical immigrants assessed by our respondents, 46 percent were supported for naturalization, 18 percent were selected for deportation to their country of origin, and the remaining 36 percent were selected to remain in the country without being granted citizenship. Figure 1 displays the percent of immigrant profiles overall and with each randomly selected attribute for which respondents selected naturalization. It is important to be cautious about drawing conclusions from the raw numbers, which do not account for immigrant and respondent controls or test for interaction effects. Even so, the generally high rate of support for immigrant naturalization is noteworthy in a country like Côte d’Ivoire, where migration and citizenship have been at the heart of recent political conflicts.⁵ Despite a complicated history, Ivorians still seem open to the possibility of more immigration.

Figure 1.

Naturalization by immigrant characteristic.

To assess the relative influence of different variables, we turn to a series of multinomial logistic regressions. This approach is optimal considering our research design. Respondents selected one of three options for each hypothetical immigrant: naturalize, remain without naturalizing, or deport. Although there is some logic in viewing the dependent variable in an ordered way, it is likely, and indeed expected, that some variables would affect the likelihood of support for one action more than the other. Multinomial logistic regression with robust standard errors clustered by respondent thus is ideal because it can account for all three dependent variable options as well as respondent predilection toward either action while avoiding the introduction of a probability odds assumption. The multinomial logit assumes independence of irrelevant alternatives (IIA), or, in other words, that removing one choice would not change the decision between the two remaining choices. In our case, we assume IIA holds because if a respondent chooses to naturalize, for example, it is likely they would still have chosen naturalize even if one of the other options was not available.⁶

We begin with model 1 in Table 1, which presents the effects of various immigrant and respondent characteristics on the decision to naturalize or deport; the status quo option of remaining in the country without naturalization is the reference category. Of the immigrant characteristics included in model 1, only the manner of entry and level of savings are significant for both the likelihood of being recommended for naturalization and for deportation. More specifically, illegal entry decreases the relative risk ratio (or odds) of being supported for naturalization (compared with the base category of remaining in the country) by about 0.75 times and increases the odds of being selected for deportation by 2.04 times. Similarly, a one-unit increase in the level of savings, which corresponds in the experiment to having saved five million more CFA francs, is associated with a 1.15 times increase in the odds of being supported for naturalization and a 0.68 times decrease in the risk of being selected for deportation. Respondents thus prefer immigrants who entered the country legally and have savings that can contribute to the economy, and they consider both factors when choosing to naturalize or deport a hypothetical immigrant.

Table 1.

Statistical Results: Multinomial Logit Estimator.

	Model 1		Model 2		Model 3
	Naturalize	Deport	Naturalize	Deport	Naturalize	Deport
Mother is a citizen	1.052***	−0.366	1.077***	−0.350	1.065***	−0.359
Mother is a citizen	(0.168)	(0.223)	(0.169)	(0.223)	(0.169)	(0.223)
Father is a citizen	1.713***	−0.316	1.726***	−0.283	1.718***	−0.313
Father is a citizen	(0.178)	(0.225)	(0.178)	(0.222)	(0.179)	(0.225)
Speaks French	0.205	−0.107	0.216	−0.112	0.213	−0.103
Speaks French	(0.135)	(0.157)	(0.135)	(0.157)	(0.135)	(0.157)
Entered illegally	−0.294**	0.712***	−0.270**	0.730***	−0.273**	0.723***
Entered illegally	(0.134)	(0.186)	(0.134)	(0.187)	(0.134)	(0.186)
Employed	0.432***	−0.175	0.415***	−0.171	0.417***	−0.184
Employed	(0.139)	(0.164)	(0.139)	(0.166)	(0.139)	(0.166)
Female	0.203	0.086	0.191	0.079	0.196	0.083
Female	(0.125)	(0.154)	(0.126)	(0.152)	(0.125)	(0.153)
Years lived in country	0.267***	−0.085	0.272***	−0.073	0.268***	−0.084
Years lived in country	(0.086)	(0.091)	(0.086)	(0.090)	(0.086)	(0.091)
Level of education	0.185**	0.001	0.187**	0.014	0.187**	0.004
Level of education	(0.075)	(0.099)	(0.076)	(0.099)	(0.075)	(0.098)
Level of savings	0.142*	−0.381***	0.148**	−0.379***	0.147**	−0.381***
Level of savings	(0.075)	(0.094)	(0.075)	(0.094)	(0.075)	(0.094)
Christian	0.004	−0.312	0.015	−0.292	0.006	−0.310
Christian	(0.154)	(0.192)	(0.153)	(0.195)	(0.153)	(0.192)
Muslim	0.122	−0.201	0.128	−0.203	0.119	−0.200
Muslim	(0.166)	(0.201)	(0.167)	(0.202)	(0.166)	(0.201)
Employed (respondent)	0.060	0.0240	0.038	0.032	0.051	0.020
Employed (respondent)	(0.157)	(0.224)	(0.159)	(0.224)	(0.158)	(0.224)
Education level (respondent)	−0.106	−0.082	−0.114	−0.099	−0.110	−0.083
Education level (respondent)	(0.121)	(0.155)	(0.122)	(0.155)	(0.121)	(0.155)
Age (respondent)	−0.060	−0.349***	−0.058	−0.336***	−0.056	−0.347***
Age (respondent)	(0.091)	(0.117)	(0.091)	(0.117)	(0.090)	(0.117)
Female (respondent)	−0.585***	0.383*	−0.587***	0.383*	−0.588***	0.383*
Female (respondent)	(0.160)	(0.210)	(0.162)	(0.211)	(0.160)	(0.210)
Southern location (respondent)	0.337**	−0.709***	0.354**	−0.701***	0.346**	−0.706***
Southern location (respondent)	(0.164)	(0.218)	(0.163)	(0.218)	(0.163)	(0.218)
From Burkina Faso			0.037	−0.496*
From Burkina Faso			(0.226)	(0.291)
From Mali			0.225	−0.153
From Mali			(0.244)	(0.299)
From Lebanon			0.180	0.066
From Lebanon			(0.232)	(0.291)
From France			0.320	0.047
From France			(0.245)	(0.278)
From China			0.544**	−0.157
From China			(0.248)	(0.326)
Immigrant from United States			0.318	−0.263
Immigrant from United States			(0.251)	(0.307)
African immigrant					−0.253*	−0.135
African immigrant					(0.135)	(0.159)
Constant	−1.257***	0.051	−1.512***	0.124	−1.162***	0.102
Constant	(0.289)	(0.404)	(0.319)	(0.423)	(0.296)	(0.411)
Observations	1,450		1,450		1,450
Pseudo-R²	.120		.124		.121

Reference category is to let immigrant stay in country but not to naturalize. Robust standard errors in parentheses.

p < .10. **p < .05. ***p < .01.

Other immigrant characteristics significantly affect the likelihood of one action, but not the other. Having an Ivorian parent has a strong and positive effect on the likelihood of being supported for naturalization. This suggests that jus sanguinis (right of blood) is a factor in Côte d’Ivoire. Although the citizenship status of both the mother and father are strong and positive, the father’s citizenship status has a stronger effect than does the mother’s, suggesting widespread patrilineal sentiment. If the hypothetical immigrant’s mother is a citizen, the odds of being supported for naturalization increase by 2.86 times over remaining in the country; if the father is a citizen, the odds increase by 5.55 times. Nevertheless, while parents’ citizenship status significantly affects support for naturalization, it does not significantly affect support for deportation. Ivorians clearly concern themselves with jus sanguinis when it comes to granting citizenship more than when it comes to allowing someone to remain in the country.

Model 1 also yields three additional immigrant variables that affect the likelihood of a respondent supporting naturalization, but not deportation: number of years in the country, level of education, and employment status increased the odds of being supported for naturalization by 1.31, 1.20, and 1.54 times, respectively. Respondents were unlikely to recommend deportation based on any of these factors, however, suggesting a desire to have a steady supply of lower skilled migrant workers. Higher levels of education and better employment status indicate that the immigrant can contribute more to the economy, while a greater number of years in the country implies that the immigrant has developed deeper connections to Côte d’Ivoire.

Several respondent characteristics also affect the likelihood of recommending naturalization or deportation. Interestingly, female respondents were statistically less likely to support naturalization and more likely to recommend deportation. This is consistent with some cross-national studies that find greater opposition to immigration among women (Mayda 2006; O’Rourke and Sinnott 2006; Whitaker and Giersch 2015), and supports research showing that women are less supportive of globalization more broadly due to its competitive nature (Mansfield, Mutz, and Silver 2015). Even so, other empirical studies show that women in the United States and Europe are less opposed to immigration than men (Ceobanu and Escandell 2008; Fetzer 2000), suggesting a need for cross-regional research on this question.

Beyond gender, respondents in the southern research sites were more likely to support naturalization and less likely to recommend deportation than respondents in the north. This could reflect an urban-rural divide, as the southern collection sites are far more urbanized (Abidjan is the largest city in Côte d’Ivoire and Grand-Bassam is effectively a suburb). Past research shows that urban residents tend to be more supportive of immigration than rural residents (Espenshade and Hempstead 1996). Finally, older respondents were significantly less likely to recommend deportation, but were not more likely to support naturalization. This could suggest a more hands-off attitude among older people, though we are cautious to draw such a conclusion given the small number of older respondents in our sample.⁷

Based on the results in model 1, Figure 2 displays the marginal effect of each immigrant attribute on the likelihood of being recommended for naturalization or deportation. Dots indicate the point estimates and lines show the 90 percent confidence intervals for each attribute value; dots without lines indicate the reference category for each attribute. Thus, when the confidence interval line does not cross the center vertical line, there is a statistically significant difference between the given attribute value and the reference category. The figure confirms that respondents were significantly more likely to select naturalization for hypothetical immigrants who have parents who are Ivorian citizens, have lived in the country for twenty years, have a university diploma, have saved large sums to start a business, and are employed; respondents were significantly less likely to support naturalization for immigrants who did not enter the country at a legal port of entry. In the right side of the figure, these factors tend to have the opposite effect on support for deportation, although the effects are not always statistically significant, as discussed above. These findings support the notion that citizens generally prefer immigrants who can contribute to the economy, have deeper connections to the host country, and entered legally. With respect to our first two clusters of expectations, then, evidence from Côte d’Ivoire is largely consistent with past findings from Western countries.

Figure 2.

Marginal effects of immigrant characteristics.

Extending the baseline model to assess our cultural proximity hypotheses, model 2 adds the immigrant’s country of origin, leaving out Ghana as the reference category. Hypothetical immigrants from China were significantly more likely to be recommended for naturalization than those from Ghana, although there was no effect on their likelihood of deportation. In contrast, immigrants from Burkina Faso were less likely to be supported for deportation than immigrants from Ghana but no more likely to be supported for naturalization.

We expected support to be higher for hypothetical immigrants from non-African countries (China, United States, France, Lebanon) and lower for those from neighboring African countries (Ghana, Mali, Burkina Faso). Model 3 thus simplifies the country of origin variables into a single African variable, which takes a value of “1” if the hypothetical immigrant is from an African country and “0” otherwise. As expected, hypothetical immigrants from African countries were significantly less likely to be selected for naturalization, although there was no significant effect on support for deportation. The odds of being supported for naturalization decreased by 0.78 times for African immigrants compared with immigrants from outside the region. Given the parameters of our experiment, it is impossible to determine whether Ivorian respondents would have been more open to Africans from nonneighboring countries.

We next examine how shared attributes between the hypothetical immigrant and the respondent influence support for naturalization or deportation. Model 4 in Table 2 adds several “match” variables to our baseline model.⁸ Respondents were significantly less likely to support the deportation of immigrants with whom they shared a religious faith (religion match), but this did not affect support for naturalization. Substantively, immigrants of the same religion as the respondent were 0.65 times less likely to be selected for deportation. In contrast, respondents were no more likely to recommend naturalization or deportation for hypothetical immigrants with whom they shared a common ancestry (ancestry match). In the exit survey, most respondents indicated that there were already too many immigrants in Côte d’Ivoire from Burkina Faso (93%), Mali (87%), Lebanon (83%), Ghana (59%), and France (53%), while smaller portions thought there were too many immigrants from China (17%) and the United States (9%). Even so, a belief that there were already too many immigrants from a hypothetical immigrant’s country (too many match) did not significantly affect their preference concerning that individual. Furthermore, we found no evidence to support a “not in my backyard” effect when an immigrant was presented as intending to settle in the respondent’s geographic area (location match).

Table 2.

Statistical Results: Multinomial Logit Estimator.

	Model 4		Model 5
	Naturalize	Deport	Naturalize	Deport
Mother is a citizen	1.058***	−0.383*	1.049***	−0.394*
Mother is a citizen	(0.170)	(0.219)	(0.171)	(0.220)
Father is a citizen	1.712***	−0.338	1.716***	−0.331
Father is a citizen	(0.180)	(0.225)	(0.181)	(0.225)
Speaks French	0.202	−0.119	0.206	−0.109
Speaks French	(0.135)	(0.158)	(0.135)	(0.158)
Entered illegally	−0.295**	0.719***	−0.301**	0.721***
Entered illegally	(0.135)	(0.187)	(0.134)	(0.188)
Employed	0.434***	−0.167	0.445***	−0.166
Employed	(0.140)	(0.166)	(0.140)	(0.165)
Female	0.191	0.086	0.213*	0.087
Female	(0.126)	(0.157)	(0.128)	(0.156)
Years lived in country	0.260***	−0.096	0.263***	−0.099
Years lived in country	(0.086)	(0.092)	(0.086)	(0.091)
Level of education	0.187**	0.005	0.193***	0.004
Level of education	(0.076)	(0.100)	(0.075)	(0.099)
Level of savings	0.153**	−0.384***	0.159**	−0.386***
Level of savings	(0.075)	(0.094)	(0.074)	(0.093)
Employed (respondent)	0.087	−0.002	0.065	−0.010
Employed (respondent)	(0.157)	(0.224)	(0.160)	(0.222)
Education level (respondent)	−0.118	−0.075	−0.060	−0.046
Education level (respondent)	(0.120)	(0.158)	(0.119)	(0.164)
Age (respondent)	−0.070	−0.351***	−0.062	−0.346***
Age (respondent)	(0.091)	(0.120)	(0.092)	(0.117)
Female (respondent)	−0.577***	0.424**	−0.568***	0.440**
Female (respondent)	(0.162)	(0.212)	(0.161)	(0.213)
Southern location (respondent)	0.346**	−0.750***	0.405**	−0.719***
Southern location (respondent)	(0.167)	(0.221)	(0.168)	(0.222)
Religion match	0.178	−0.434**
Religion match	(0.148)	(0.202)
Ancestry match	0.374	0.0237
Ancestry match	(0.242)	(0.318)
Too many match	−0.111	−0.019
Too many match	(0.134)	(0.170)
Location match	−0.052	0.057
Location match	(0.167)	(0.235)
Christian	−0.081	−0.105	−0.261	−0.116
Christian	(0.167)	(0.211)	(0.213)	(0.258)
Christian (respondent)			−0.241	−0.057
Christian (respondent)			(0.378)	(0.435)
Christian × Christian (respondent)			0.547*	−0.391
Christian × Christian (respondent)			(0.299)	(0.373)
Muslim	0.048	−0.059	0.265	−0.023
Muslim	(0.174)	(0.204)	(0.224)	(0.235)
Muslim (respondent)			0.413	0.162
Muslim (respondent)			(0.400)	(0.433)
Muslim × Muslim (respondent)			−0.361	−0.488
Muslim × Muslim (respondent)			(0.327)	(0.400)
Constant	−1.230***	0.095	−1.409***	0.010
Constant	(0.302)	(0.426)	(0.481)	(0.517)
Observations	1,444		1,444
Pseudo-R²	.125		.126

Reference category is to let immigrant stay in country but not to naturalize. Robust standard errors in parentheses. Robust standard errors in parentheses.

p < .10. **p < .05. ***p < .01.

These results provide partial confirmation of the cultural proximity hypothesis that citizens will favor immigrants with whom they are more culturally similar, at least when it comes to religion, and challenge recent findings from West Africa that cultural proximity exacerbates tensions between hosts and immigrants (Adida 2011, 2014). Even so, to tease out the influence of a religious match and to see if it applies across different religious groups, model 5 replaces the match variables with interaction variables for the dominant religions of the hypothetical immigrants and respondents (Christian × Christian Respondent and Muslim ×Muslim Respondent). The results show that Christian respondents were significantly more likely to support the naturalization of Christian immigrants but were not less likely to support the deportation of fellow Christians; Muslim respondents were not more or less likely to support naturalization or deportation in either case. Further research is needed to determine why a common religion may be significant for Christians, but not Muslims, in Côte d’Ivoire.

Finally, we tested the influence of an immigrant’s intention to vote in elections after receiving citizenship. Although voting could be interpreted as a positive indicator of intended civic engagement and integration into the host country, it could also be perceived as a political threat. Starting with the earlier baseline, model 6 in Table 3 includes religion match and African, which were found to be significant above, and adds a variable (will vote in future) for the voting intentions of the hypothetical immigrant. The results show that immigrants who intend to vote in elections after receiving citizenship were 1.54 times more likely to be supported for naturalization over remaining in the county than those who did not intend to vote. This suggests that the net effect of an immigrant intending to vote is positive, with the indicator of civic engagement and integration into the host country outweighing any concerns about vote dilution.

Table 3.

Statistical Results: Multinomial Logit Estimator.

	Model 6		Model 7
	Naturalize	Deport	Naturalize	Deport
Mother is a citizen	1.075***	−0.381*	1.086***	−0.375*
Mother is a citizen	(0.171)	(0.221)	(0.171)	(0.221)
Father is a citizen	1.720***	−0.345	1.726***	−0.342
Father is a citizen	(0.180)	(0.226)	(0.180)	(0.226)
Speaks French	0.191	−0.098	0.200	−0.094
Speaks French	(0.134)	(0.159)	(0.133)	(0.159)
Entered illegally	−0.276**	0.728***	−0.283**	0.727***
Entered illegally	(0.136)	(0.188)	(0.136)	(0.188)
Employed	0.434***	−0.195	0.422***	−0.201
Employed	(0.141)	(0.170)	(0.141)	(0.170)
Female	0.202	0.077	0.206	0.078
Female	(0.127)	(0.155)	(0.127)	(0.156)
Years lived in country	0.258***	−0.097	0.261***	−0.096
Years lived in country	(0.086)	(0.092)	(0.086)	(0.092)
Level of education	0.191**	0.007	0.195***	0.008
Level of education	(0.075)	(0.099)	(0.076)	(0.099)
Level of savings	0.153**	−0.383***	0.161**	−0.380***
Level of savings	(0.075)	(0.094)	(0.075)	(0.094)
Christian	−0.078	−0.115	−0.070	−0.110
Christian	(0.167)	(0.212)	(0.166)	(0.212)
Muslim	0.025	−0.052	0.036	−0.049
Muslim	(0.173)	(0.203)	(0.171)	(0.204)
African immigrant	−0.257*	−0.153	−0.251*	−0.152
African immigrant	(0.138)	(0.162)	(0.138)	(0.162)
Employed (respondent)	0.055	−0.008	0.065	−0.003
Employed (respondent)	(0.158)	(0.223)	(0.158)	(0.223)
Education level (respondent)	−0.105	−0.073	−0.103	−0.072
Education level (respondent)	(0.121)	(0.156)	(0.122)	(0.156)
Age (respondent)	−0.055	−0.339***	−0.056	−0.338***
Age (respondent)	(0.089)	(0.117)	(0.089)	(0.117)
Female (respondent)	−0.605***	0.427**	−0.602***	0.428**
Female (respondent)	(0.162)	(0.211)	(0.162)	(0.210)
Southern location (respondent)	0.330**	−0.740***	0.329**	−0.740***
Southern location (respondent)	(0.166)	(0.219)	(0.166)	(0.219)
Will vote in future	0.430***	−0.270	0.257	−0.343*
Will vote in future	(0.137)	(0.172)	(0.165)	(0.204)
Religion match	0.144	−0.412**	−0.137	−0.518**
Religion match	(0.147)	(0.203)	(0.198)	(0.262)
Will Vote × Religion Match			0.518*	0.215
Will Vote × Religion Match			(0.276)	(0.374)
Constant	−1.371***	0.261	−1.318***	0.281
Constant	(0.306)	(0.416)	(0.308)	(0.417)
Observations	1,444		1,444
Pseudo-R²	.132		.133

Reference category is to let immigrant stay in country but not to naturalize. Robust standard errors in parentheses. Robust standard errors in parentheses.

p < .10. **p < .05. ***p < .01.

To the degree that an immigrant who intends to vote shares a politically salient cultural attribute with a respondent, they may be expected to vote in a similar way and thus may be viewed as a political ally. Model 7 thus includes an interaction between immigrant voting intentions and a religion match between immigrant and respondent (Intends to Vote × Religion Match). The results show that respondents were significantly more likely to support the naturalization of an immigrant who intended to vote in future elections when that immigrant shared their religious faith. More specifically, the odds of a hypothetical immigrant being selected for naturalization increased by 1.68 times when they intended to vote and were of the same religion as the respondent. Interestingly, once we account for this interaction effect, the independent effects of both intends to vote and religion match on support for naturalization become insignificant. With respect to support for deportation, however, the opposite occurs. The interaction between intends to vote and religion match has no significant effect, while each variable on its own independently and significantly decreases support for deportation. These findings highlight the degree to which citizens may make electoral calculations when considering preferences for naturalization versus deportation.

As discussed above, our design did not force respondents to choose between immigrants. Respondents thus could choose to naturalize (or deport) all of the profiles they reviewed. We considered this possibility and examined choices by respondent. Only three respondents chose to naturalize every hypothetical immigrant, and only one respondent chose to deport every time.⁹ The lack of “categorical” decision making in our data provides an interesting contrast with existing research (Wright, Levy, and Citrin 2016), which suggests that respondents make categorical decisions especially when assessing illegal immigration.

Given regional divisions within Côte d’Ivoire, we also divided the sample into respondents from the North and the South to see if estimates varied by region. We replicated models 1 and 2 with results available in the supplemental material. We find a few differences from model 1 in the South. Speaking French remains positive but is now significant, as is the estimate for a female immigrant. In contrast, level of education is positive but not significant for naturalization. In the North, respondents were significantly less likely to deport if the immigrant’s mother or father is a citizen. Illegal entry still has a negative effect on naturalization, but is not significant; it remains positive and significant for deportation. Finally, education in the North is positive and significant for naturalization. We also further tested the effect of an immigrant coming from Burkina Faso, which in Table 1 has a negative and significant effect on the decision to deport. This finding appears to be driven by respondents in the North, who have more historical and cultural connections with migrants from Burkina Faso. Northern respondents were significantly less likely to deport a hypothetical immigrant from Burkina Faso, while the estimate for respondents in the South is positive but insignificant.

Conclusion

The results presented in this paper make an important contribution to our understanding of how different citizen and immigrant attributes influence support for naturalization or deportation. Although these issues have received significant scholarly attention in North America and Europe, little experimental research has been conducted in developing countries where a large portion of global migration occurs. Based on a survey experiment in Côte d’Ivoire, where immigrants represent more than one-tenth of the total population, our research builds on findings from existing literature while also testing new hypotheses.

Using a unique research design that allows us to examine factors influencing support for naturalization and support for deportation, our findings from Côte d’Ivoire show that an immigrant’s ability to contribute to the economy and legal status affect support for either outcome, while deeper connections to the host country affect only support for naturalization. With deportation on the rise around the world, these results suggest that greater attention should be given to factors that influence support for one outcome but not the other. With respect to country of origin, we found significantly more support for the naturalization of immigrants from non-African countries, contradicting past findings from other regions. Even so, respondents were less likely to support deportation for hypothetical immigrants with whom they shared a religion, and Christians in particular were more likely to support the naturalization of fellow Christians, providing some evidence for theories of cultural proximity. Finally, we introduced a novel attribute that has not been tested elsewhere: an immigrant’s intention to vote if granted citizenship. Respondents were significantly more likely to support the naturalization of immigrants who planned to vote, especially those with whom they shared a religious faith, indicating a degree of electoral calculation. Overall, our findings suggest that citizen support for naturalization and support for deportation of immigrants are influenced by somewhat different factors, highlighting the need for more research on attitudes toward deportation in particular.

A key question is the degree to which our results are valid both externally and internally. In the African context, Côte d’Ivoire is similar to other emerging democracies such as South Africa, Kenya, and Nigeria, which also host more than one million African immigrants each (World Bank 2018) and have contentious debates surrounding immigration.¹⁰ Although none of those countries has experienced a civil war over issues of citizenship, they have seen lower levels of violence and discrimination. Existing research suggests that many of the same economic, cultural, and political factors influence attitudes in these countries, and opposition to immigration is often even higher than it was among our respondents in Côte d’Ivoire, especially in South Africa (Whitaker and Giersch 2015). Ultimately, more research is necessary to determine the extent to which our findings hold in other developing countries in Africa and beyond.

With respect to the degree to which respondents were reacting to our experimental treatments, there are reasons to believe that our results are internally valid. First, almost all respondents finished the surveys and each was given ample time to consider each immigrant profile. Follow-up questions indicated that respondents were able to quickly identify the qualities of each immigrant that influenced their decision. Second, because each respondent evaluated multiple profiles, we were able to account for any time-constant factors using a fixed-effects estimator. This approach controls for possible alternative influences, such as if the respondent had prior views on immigrants. Results using the fixed-effects logit estimator are consistent with the results presented here and are available in the supplemental material. Third, as noted, respondents generally did not select one response across all six profiles, indicating there is no testing effect in which one learns or chooses the preferred response.

In the end, our results are relevant to the formulation of immigration policies in Côte d’Ivoire and elsewhere. Our findings suggest that citizens are more likely to support policies that allow for the naturalization of immigrants with family connections to the country or who have been in the country for longer periods, those who have entered the country legally, and those who can contribute economically. This latter part fits with the movement to develop merit-based criteria for immigration policy. Of course, governments must weigh merit against demand, and many immigrants may not have the levels of education or savings that are seen favorably but still seek the chance for a better life. As policy makers seek to balance competing forces, electoral considerations will continue to affect debates about immigrants and immigration policy in both developed and developing countries.

Supplemental Material

cdw_supplementary_material – Supplemental material for Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire

Supplemental material, cdw_supplementary_material for Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire by Nathaniel Terence Cogley, John Andrew Doces and Beth Elise Whitaker in Political Research Quarterly

Supplemental Material

Replication_data – Supplemental material for Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire

Supplemental material, Replication_data for Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire by Nathaniel Terence Cogley, John Andrew Doces and Beth Elise Whitaker in Political Research Quarterly

Supplemental Material

Replication_PRQ_Final – Supplemental material for Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire

Supplemental material, Replication_PRQ_Final for Which Immigrants Should Be Naturalized? Which Should Be Deported? Evidence from a Survey Experiment in Côte d’Ivoire by Nathaniel Terence Cogley, John Andrew Doces and Beth Elise Whitaker in Political Research Quarterly

Footnotes

Acknowledgements

We would like to thank Donald P. Green, Dan Hopkins, Joel Middleton, and three anonymous reviewers for their valuable comments and support, and Donatien Kouakou Adou for his research assistance. We received useful feedback at Bucknell University’s economics department seminar series. Institutional review board (IRB) approval was obtained through Bucknell University. All errors remain our own.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Supplemental Materials

Supplemental materials and replication materials for this article are available with the manuscript on the Political Research Quarterly (PRQ) website.

References

Adida

Claire L.

2011. “Too Close for Comfort? Immigrant Exclusion in Africa.” Comparative Political Studies 44 (10): 1370–96.

Adida

Claire L.

2014. Immigrant Exclusion and Insecurity in Africa. New York: Cambridge University Press.

Art

David

. 2011. Inside the Radical Right: The Development of Anti-immigrant Parties in Western Europe. New York: Cambridge University Press.

Bah

Abu Bakarr

. 2010. “Democracy and Civil War: Citizenship and Peacemaking in Côte d’Ivoire.” African Affairs 109 (437): 597–615.

Bierwirth

Chris

. 1999. “The Lebanese Communities of Côte d’Ivoire.” African Affairs 98 (390): 79–99.

Boone

Catherine

Kriger

Norma

. 2010. “Multiparty Elections and Land Patronage: Zimbabwe and Côte d’Ivoire.” Commonwealth & Comparative Politics 48 (2): 173–202.

Brader

Ted

Valentino

Nicholas A.

Suhay

Elizabeth

. 2008. “What Triggers Public Opposition to Immigration? Anxiety, Group Cues, and Immigration Threat.” American Journal of Political Science 52 (4): 959–78.

Brautigam

Deborah

. 2009. The Dragon’s Gift: The Real Story of China in Africa. New York: Oxford University Press.

Burns

Peter

Gimpel

James G.

2000. “Economic Insecurity, Prejudicial Stereotypes, and Public Opinion on Immigration Policy.” Political Science Quarterly 115 (2): 201–25.

10.

Ceobanu

Alin M.

Escandell

Xavier

. 2008. “East is West? National Feelings and Anti-immigrant Sentiment in Europe.” Social Science Research 37 (4): 1147–70.

11.

Cho

Wendy K. Tam

. 1999. “Naturalization, Socialization, Participation: Immigrants and (Non-)Voting.” The Journal of Politics 61 (4): 1140–55.

12.

Coenders

Marcel

Scheepers

Peer

. 2008. “Changes in Resistance to the Social Integration of Foreigners in Germany 1980-2000: Individual and Contextual Determinants.” Journal of Ethnic and Migration Studies 34 (1): 1–26.

13.

Cosby

Arthur

Aanstoos

Kristen

Matta

Marissa

Porter

Jeremy

James

Wesley

. 2013. “Public Support for Hispanic Deportation in the United States: The Effects of Ethnic Prejudice and Perceptions of Economic Competition in a Period of Economic Distress.” Journal of Population Research 30 (1): 87–96.

14.

Coulibaly

Loucoumane

Bavier

Joe

. 2015. “Ivory Coast’s Ouattara Secures Second Term in Landslide Poll Win.” Reuters, October 28. https://www.reuters.com/article/us-ivorycoast-election/ivory-coasts-president-ouattara-poised-to-win-landslide-poll-victory-idUSKCN0SL2VI20151028.

15.

Craig

Maureen A.

Richeson

Jennifer A.

2014. “Not in My Backyard! Authoritarianism, Social Dominance Orientation, and Support for Strict Immigration Policies at Home and Abroad.” Political Psychology 35 (3): 417–29.

16.

Crook

Richard C.

1997. “Winning Coalitions and Ethno-regional Politics: The Failure of the Opposition in the 1990 and 1995 Elections in Côte d’Ivoire.” African Affairs 96 (383): 215–42.

17.

Daddieh

Cyril K.

2001. “Elections and Ethnic Violence in Côte d’Ivoire: The Unfinished Business of Succession and Democratic Transition.” African Issues 29 (1-2): 14–19.

18.

Espenshade

Thomas J.

Hempstead

Katherine

. 1996. “Contemporary American Attitudes toward U.S. Immigration.” International Migration Review 30 (2): 535–70.

19.

Fetzer

Joel S.

2000. “Economic Self-interest or Cultural Marginality? Anti-immigration Sentiment and Nativist Political Movements in France, Germany and the USA.” Journal of Ethnic and Migration Studies 26 (1): 5–23.

20.

Ford

Robert

. 2011. “Acceptable and Unacceptable Immigrants: How Opposition to Immigration in Britain Is Affected by Migrants’ Region of Origin.” Journal of Ethnic and Migration Studies 37 (7): 1017–37.

21.

Geschiere

Peter

. 2009. The Perils of Belonging: Autochthony, Citizenship, and Exclusion in Africa and Europe. Chicago: University of Chicago Press.

22.

Gonzalez-Garcia

Jesus

Hitaj

Ermal

Mlachila

Montfort

Viseth

Arina

Yenice

Mustafa

. 2016. Sub-Saharan African Migration: Patterns and Spillovers. Washington, DC: International Monetary Fund.

23.

Gordon

Steven

Maharaj

Brij

. 2015. “Neighbourhood-level Social Capital and Anti-immigrant Prejudice in an African Context: An Individual-level Analysis of Attitudes towards Immigrants in South Africa.” Commonwealth & Comparative Politics 53 (2): 197–219.

24.

Hainmueller

Jens

Hangartner

Dominik

. 2013. “Who Gets a Swiss Passport? A Natural Experiment in Immigrant Discrimination.” American Political Science Review 107 (1): 159–87.

25.

Hainmueller

Jens

Hangartner

Dominik

Yamamoto

Teppei

. 2015. “Validating Vignette and Conjoint Survey Experiments against Real-World Behavior.” Proceedings of the National Academy of Sciences 112 (8): 2395–2400.

26.

Hainmueller

Jens

Hiscox

Michael J.

2007. “Educated Preferences: Explaining attitudes toward Immigration in Europe.” International Organization 61 (2): 399–442.

27.

Hainmueller

Jens

Hiscox

Michael J.

2010. “Attitudes toward Highly Skilled and Low-skilled Immigration: Evidence from a Survey Experiment.” American Political Science Review 104 (1): 61–84.

28.

Hainmueller

Jens

Hopkins

Daniel J.

2014. “Public Attitudes toward Immigration.” Annual Review of Political Science 17 (1): 225–49.

29.

Hainmueller

Jens

Hopkins

Daniel J.

2015. “The Hidden American Immigration Consensus: A Conjoint Analysis of Attitudes toward Immigrants.” American Journal of Political Science 59 (3): 529–48.

30.

Harell

Allison

Soroka

Stuart

Iyengar

Shanto

Valentino

Nicholas

. 2012. “The Impact of Economic and Cultural Cues on Support for Immigration in Canada and the United States.” Canadian Journal of Political Science 45 (3): 499–530.

31.

Harris

Adam S.

Findley

Michael G.

Nielson

Daniel L.

Noyes

Kennard L.

2017. “The Economic Roots of Anti-immigrant Prejudice in the Global South: Evidence from South Africa.” Political Research Quarterly 71 (1): 228–41.

32.

Hellwig

Timothy

Sinno

Abdulkader

. 2017. “Different Groups, Different Threats: Public Attitudes towards Immigrants.” Journal of Ethnic and Migration Studies 43 (3): 339–58.

33.

Hjerm

Mikael

. 2007. “Do Numbers Really Count? Group Threat Theory Revisited.” Journal of Ethnic and Migration Studies 33 (8): 1253–75.

34.

Hopkins

Daniel J.

2010. “Politicized Places: Explaining Where and When Immigrants Provoke Local Opposition.” American Political Science Review 104 (1): 40–60.

35.

Hopkins

Daniel J.

2011. “National Debates, Local Responses: The Origins of Local Concern about Immigration in Britain and the United States.” British Journal of Political Science 41 (3): 499–524.

36.

Hopkins

Daniel J.

2015. “The Upside of Accents: Language, Inter-group Difference, and Attitudes toward Immigration.” British Journal of Political Science 45 (3): 531–57.

37.

Landau

Loren B.

2010. “Loving the Alien? Citizenship, Law, and the Future in South Africa’s Demonic Society.” African Affairs 109 (435): 213–30.

38.

Landau

Loren B.

, ed. 2011. Exorcising the Demons within: Xenophobia, Violence and Statecraft in Contemporary South Africa. Johannesburg: Wits University Press.

39.

Long

Scott J.

Freese

Jeremy

. 2014. Regression Models for Categorical Dependent Variables in Stata. 3rd ed. College Station: Stata Press.

40.

Mansfield

Edward D.

Mutz

Diana C.

Silver

Laura R.

2015. “Men, Women, Trade, and Free Markets.” International Studies Quarterly 59 (2): 303–15.

41.

Mayda

Anna Maria

. 2006. “Who Is against Immigration? A Cross-Country Investigation of Individual Attitudes toward Immigrants.” Review of Economics and Statistics 88 (3): 510–30.

42.

Mitchell

Matthew I.

2011. “Insights from the Cocoa Regions in Côte d’Ivoire and Ghana: Rethinking the Migration-conflict Nexus.” African Studies Review 54 (2): 123–24.

43.

Money

Jeannette

. 1997. “No Vacancy: The Political Geography of Immigration Control in Advanced Industrial Countries.” International Organization 51 (4): 685–720.

44.

Money

Jeannette

. 2010. Comparative Immigration Policy. The International Studies Association Compendium Project. http://www.isacompss.com/.

45.

Neocosmos

Michael

. 2010. From “Foreign Natives” to “Native Foreigners”: Explaining Xenophobia in Post-apartheid South Africa. Dakar, Senegal: Council for the Development of Social Science Research in Africa.

46.

Nyamnjoh

Francis B.

2006. Insiders and Outsiders: Citizenship and Xenophobia in Contemporary Southern Africa. London: Zed Books.

47.

O’Rourke

Kevin H.

Sinnott

Richard

. 2006. “The Determinants of Individual Attitudes towards Immigration.” European Journal of Political Economy 22 (4): 838–61.

48.

Quillian

Lincoln

. 1995. “Prejudice as a Response to Perceived Group Threat: Population Composition and Anti-immigrant and Racial Prejudice in Europe.” American Sociological Review 60 (4): 586–611.

49.

Ramakrishnan

S. Karthick

Espenshade

Thomas J.

2001. “Immigrant Incorporation and Political Participation in the United States.” International Migration Review 35 (3): 870–909.

50.

Rosenstone

Steven J.

Hansen

John Mark

. 1993. Mobilization, Participation, and Democracy in America. New York: Macmillan.

51.

Rydgren

Jens

. 2003. “Meso-level Reasons for Racism and Xenophobia: Some Converging and Diverging Effects of Radical Right Populism in France and Sweden.” European Journal of Social Theory 6 (1): 45–68.

52.

Rydgren

Jens

. 2005. “Is Extreme Right-wing Populism Contagious? Explaining the Emergence of a New Party Family.” European Journal of Political Research 44 (3): 413–37.

53.

Scheve

Kenneth F.

Slaughter

Matthew J.

2001. “Labor Market Competition and Individual Preferences over Immigration Policy.” Review of Economics and Statistics 83 (1): 133–45.

54.

Sides

John

Citrin

Jack

. 2007. “European Opinion about Immigration: The Role of Identities, Interests and Information.” British Journal of Political Science 37 (3): 477–504.

55.

Simonen

Katariina

. 2012. “Qui S’excuse S’accuse . . . An Analysis of French Justifications for Intervening in Côte d’Ivoire.” International Peacekeeping 19 (3): 363–76.

56.

Sniderman

Paul M.

Hagendoorn

Louk

Prior

Markus

. 2004. “Predisposing Factors and Situational Triggers: Exclusionary Reactions to Immigrant Minorities.” American Political Science Review 98 (1): 35–49.

57.

Van Oudenhoven

Jan Pieter

Prins

Karin S.

Buunk

Bram P.

1998. “Attitudes of Minority and Majority Members towards Adaptation of Immigrants.” European Journal of Social Psychology 28 (6): 995–1013.

58.

Whitaker

Beth Elise

. 2005. “Citizens and Foreigners: Democratization and the Politics of Exclusion in Africa.” African Studies Review 48 (1): 109–26.

59.

Whitaker

Beth Elise

. 2015. “Playing the Immigration Card: The Politics of Exclusion in Côte d’Ivoire and Ghana.” Commonwealth & Comparative Politics 53 (3): 274–93.

60.

Whitaker

Beth Elise

Giersch

Jason

. 2015. “Political Competition and Attitudes towards Immigration in Africa.” Journal of Ethnic and Migration Studies 41 (1): 1536–57.

61.

World Bank. 2016. Migration and Remittances Factbook 2016. 3rd ed. Washington, DC: International Bank for Reconstruction and Development.

62.

World Bank. 2018. Bilateral Migration Matrix 2018. World Bank Migration and Remittances Data. http://www.worldbank.org/en/topic/migrationremittancesdiasporaissues/brief/migration-remittances-data.

63.

Wright

Matthew

Levy

Morris

Citrin

Jack

. 2016. “Public Attitudes toward Immigration Policy across the Legal/Illegal Divide: The Role of Categorical and Attribute-based Decision-making.” Political Behavior 38 (1): 229–53.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.90 MB

0.26 MB