RESEARCH ARTICLE – JCTS Volume 8, Number 1, 2020
By Emre Erdoğan & Pınar Uyan Semerci
Decades of globalization and related global economic crises have had repercussions throughout the semi-peripheries of the developing world. Adana in Turkey’s Çukurova region was among the leading industrial cities after the first wave of globalization and tensions existed between agricultural landlords, bureaucrats, early industrialists, and peasants. The massive inflow of Syrians to Turkey after the Syrian Civil War produced social problems, such as xenophobia and exclusion that compounded pre-existing issues linked to high youth unemployment and a seasonal agricultural labour force. This paper draws on research conducted in Adana in 2016 to show that locals see Syrians as the cause of key economic problems. A review of anti-immigrant literature reveals the factors that lead to negative and positive perceptions of Syrians among the populace, and a range of attitudes towards Syrian immigrants are identified, namely threat perceptions, positive perceptions, and varied views on their rights. The research evaluates independent demographic variables, immigrant status, and links to internal Kurdish migration and considers contact as a mediating variable.
Keywords: Anti-Immigrant Attitude; Perception; Threat; Contact; Rights; Turkey; Syrians
Scapegoating, or putting the blame on “others”, is a common reactionary mechanism in tense situations. The problems we face are attributed to “others”, those who are not “like us”. This mechanism is particularly often used to justify anti-immigrant attitudes, especially when an economic or security crisis occurs in a country. Anti-immigrant attitudes have been most frequently studied in the “developed” Western world, in so-called receiving countries which provide opportunities for immigrants and where negative attitudes can prevail among the pre-existing population. However, developments over the past few decades have challenged this simple view of migration. Migration is presently viewed in diverse ways and most countries in the world both send and receive migrants. Many migrants do not end up in the Global West or North, and ‘‘East to East’’ and ‘‘South to South’’ migration is possible. It is therefore important to study perceptions of foreigners within various different countries in order to understand commonalities and differences. In this article, we will elaborate on the case of Turkey which currently hosts the largest number of refugees worldwide. We focus on how locals in the province of Adana perceive Syrians and the factors that lead to positive and negative perceptions about them.
In summarizing the existing literature, we first need to note the factors that shape individuals’ attitudes towards migrants. The ways in which migrants are perceived at the individual level seems to be affected by personal demographic features such as age, gender, socio-economic conditions, employment status, class, and housing conditions Perceptions at the individual level are related to trust and to collective or personal ideas about whether or not migrants represent a threat; they are also influenced by structural or contextual variables in a country, which can include economic conditions, trade structures, and institutional characteristics, such as relatively high rates of resident migrants, immigrants, and asylum seekers; high levels of unemployment; and relatively low levels of GDP allocated to social welfare.
In order to explain anti-immigrant attitudes, most studies base their arguments on the kinds of “threat” that individuals ascribe to migrants. Many studies argue –mainly on the basis of one or two variables that are not always explicit- that individual conditions, perceptions, the group dimension, and contextual factors create different degrees of threat perception, which lead to varying levels of anti-immigrant sentiment. Theoretical explanations that do not explicitly name the perceived threats that lead to anti-immigrant attitudes often still implicitly assume that anti-immigrant attitudes are actually an outcome of feeling threatened. In fact, all individual and contextual factors that increase the level of anti-immigrant attitudes create a feeling of threat to the physical, economic, and symbolic status of the pre-existing population.
Integrated threat theory, which is one of the most well-known socio-psychological explanations for how anti-immigrant attitudes form, suggests that these attitudes arise when a population perceives a combination of realistic economic threats and symbolic threats, which are based on perceived differences such as values, religion, and culture. Since 9/11, however, various arguments based on security and safety have also come to the fore to explain anti-immigrant feeling. A rhetoric of existential threat – read as physical threat – is used to remove migrants from the context of “normal politics” and force them into the “security” realm. The increasing number of terrorist acts that have threatened everyday life in various cosmopolitan cities, such as Paris, Istanbul, London, and Brussels, has seen this rhetoric repeatedly employed in the media.
However, one of the most frequently stated explanations for anti-immigrant attitudes is economic competition. According to this view, increasing international labour flows create insecurity and feelings of threat, which are operationalized at both the individual level, through people’s employment status and access to high- or low-skilled labour, as well as at the contextual level where it figures, for example, in public rhetoric and statistics around employment. While rational choice theory explains anti-immigrant attitudes on the basis of realistic/economic threats, group threat theory argues that, in the struggle for scarce resources, members of the majority group will show negative attitudes toward “outgroups”. Contact has also been identified as a factor that determines anti-immigrant attitudes; however, contact can work in countervailing ways: it can be positive if backed by true acquaintance and negative if acquaintance is casual. People are less threatened by true acquaintances, but feel more threatened and are made less secure by frequent casual contact.
This study focuses on research conducted in Adana, Turkey, to show how perceptions of Syrians are shaped in a specific context. In particular, it identifies factors that determine negative and positive perceptions about this group, and it also explores views on the approval or denial of migrant rights. Turkey hosts more than three million people, the largest number of refugees hosted by any country worldwide. Work to shape people’s perceptions on the issue is crucial and the existing literature on anti-immigrant attitudes provides a good starting point.
The case of Adana
Adana, located on the south coast of Turkey, is Turkey’s sixth largest city with a population of 2,165,595. According to the government’s calculations, Adana is ranked 16th for socioeconomic development, which places it in the second tier of development. Adana has been one of the country’s major industrial areas since the middle of the 19th century, when cotton production began in response to a decline in cotton imports from America during its long civil war. During the earlier period of the Turkish Republic, Adana attracted a significant amount of infrastructural investment, for projects that, for example, built large dams and irrigation systems that contributed to the productivity of the agricultural sector. Moreover, the region’s geographical advantages attracted a significant amount of foreign and domestic investment in industries that were based on agricultural products such as cotton. For a couple of decades, Adana was synonymous with the rising Anatolian bourgeoisie and it was the fourth most industrial province in Turkey.
However, this picture changed over time. The average share of manufacturing employment in the TR62 Region (where Adana and Mersin are located) was measured at 6.2 per cent between 1983 and 1985 and declined to 4.2 per cent between 1998 and 2000. By 2009, this number was 2.45 per cent, and according to the last figures this trend of deindustrialization continues. Today, the service sector accounts for more than half of all employment (53 per cent), whereas agriculture and industry are responsible for about 23 per cent each. The industrial labour force is mainly employed in textiles (17 per cent) and food production (16 per cent). However, when industrial enterprises are classified by sector, it emerges that while one fifth are involved in food production, nine per cent in plastics, and eight per cent in metal production, just seven per cent are involved in the textile industry. These figures show that textile production involves relatively low levels of labour compared to food production. Moreover, we know that 86 per cent of enterprises in Adana are micro-enterprises or small businesses employing fewer than 10 people, and these figures illustrate that the majority of employment is contained in household enterprises, which deliver 30 per cent of Adana’s current employment. A further 40 per cent of employees are working in the 765 enterprises – two per cent of the overall total of businesses – that employ more than 49 employees. We can conclude that Adana has the characteristics of a dual economy. The majority of enterprises are small, owned by households, and work to satisfy domestic demands, while a small percentage of enterprises are relatively big and integrated with the national and global economies.
The deindustrialization of the region, and the emergence of a dual economy, has been reflected in increased rates of unemployment. According to the most recent available statistics, the unemployment rate in the region is 10.7 per cent, slightly higher than the national average of 9.9 per cent, although there are significant differences across age and gender categories.
The unemployment rate among women is relatively high in each age bracket and it reaches 29.1 per cent between the ages of 20 and 24; almost one fifth of women between the ages of 25 and 34 are unemployed. This gender gap is also visible in terms of labour force participation. As these figures represent a significant unemployment problem across the RR62 region, we can assume that the situation is no better in Adana province. The official statistics do not provide provincial figures for the last five years, but previous figures show that unemployment has always been higher in Adana than in Mersin and it reached 26 per cent in the aftermath of the financial crisis of 2009.
Adana has been an attractive destination for immigrants since the mid-19th century, and rapid industrialization in the 1950s attracted people to its workforce from neighbouring provinces. During the 1990s, the city hosted a significant number of Internally Displaced Persons from the southeastern region, and during the 2000s people living in rural areas came to the city as a result of declining living conditions. Over the last decade, Adana has become a province with net emigration, and those leaving the province are relatively more qualified than its immigrants, in terms of education and other factors. This situation seems to be linked with the deindustrialization of the province and its declining attractiveness for potential white-collar workers.
As well as experiencing this demographic shift, Adana also hosts a significant number of seasonal workers who are employed in agricultural production. Although the official figures estimate this number as lower than 8000, it is significantly higher according to local observers, and this workforce is far from being “temporary” since the majority of these workers spend almost three quarters of the year in the Çukurova region. Although these workers are not directly included in Adana’s labour market, their presence in the agricultural sector contributes to the intensification of job competition for locals.
In addition to deindustrialization, decreasing job opportunities, and the declining attractiveness of the province, the presence of Syrians in Adana makes the situation very complex. According to official statistics, Adana hosts 168,187 Syrians who are designated as having temporary protection status. This group makes up 7.6 per cent of the total population and, comparatively, Adana has the sixth largest Syrian population of any province in Turkey. However, the local government claims the population is significantly higher, at 200,000, and this difference may be a result of the existence of unregistered Syrians in the province. As a result of their financial problems, Syrians have to live in the province’s most vulnerable neighbourhoods, in undesirable housing conditions, with what are, in relative terms, very high rental payments.
A limited number of surveys show that only a small portion of Syrians can find a decent, regular job. Unemployment rates among Syrians are very high and they tend to work in jobs for which they are not trained, such as porterage, construction, and textile work. They are also employed in the agricultural sector, where they are regarded as offering an alternative to seasonal workers. Syrian women are generally excluded from the job market and deal with housework. According to experts, child labour is a common practice among Syrian families. As in other provinces, the wages of Syrians are significantly lower, at 43 TL a day, than the official minimum wage, which is double that sum, and they endure the worst working conditions. They lack job security and are not covered by the social security system. Although economic integration is stated as a necessity, the increased visibility of Arabic signboards and the growing presence of Syrians in the current labour market also leads to tension between refugees and local people. The Syrian workforce, living with the constant fear of unemployment, is a “new precariat” and its members experience low wages, the seizure of their salaries, labour exploitation, and poor workplace conditions. The findings of our research must be understood in this context because Adana, with its agricultural sector and current high unemployment rate, is among the cities in which the formation of this “new precariat” is evident.
Our findings are part of a larger study that analyses Turkey’s labour market, and research was conducted in Adana in February 2017. The study has employed qualitative and quantitative data collection methods. In the qualitative stage, 10 focus groups were conducted in Adana with employed women and men, unemployed women and men seeking job opportunities, and employers. Each group was composed of eight to 10 participants. All participants were recruited through the use a database set up according to strictly defined criteria. The moderation guide prepared by our team focused on understanding how Adana’s residents perceive the labour market, their previous employment/unemployment experiences, and Syrians. Focus group discussions, which were recorded with the permission of participants, were fully transcribed and analysed through the use of NVivo software.
In the quantitative stage, a face-to-face survey was conducted in 75 neighbourhoods of Adana (seven were self-representing units and 68 neighbourhoods were selected randomly according to the method of probability proportional to size). In each neighbourhood, eight interviews were conducted in four streets, selected from a Kish table, and in each street two houses were randomly selected according to a systematic walk principle. Participants were selected by using the nearest birthday method. The total number of participants was 1,013.
The questionnaire was prepared by our team and included demographic questions, as well as questions about the employment status of the respondent, work search channels, and perceptions about the labour market in Adana. We also asked questions about each participant’s level of contact with Syrians living in Adana, their perceptions of them, and other related issues. Interviews were controlled by field supervisors and phone calls. Data was analysed using STATA 15 software.
Although our focus group discussions were oriented towards collecting information about participants’ relationships with the labour market, discussions about Syrians dominated the agenda in all groups. In all focus groups, participants mentioned Syrians before the issue was raised by the moderator, and almost all participants told us a story that scapegoated Syrians.
Before we report their negative perceptions, which were mainly framed in terms of various forms of threat, we note that two positive statements were made. One participant noted that some Syrians can be hard workers: “Not majority of them, but there are some. For example, that man, like a fire, works in the exterior masonry, composite, both welding and composite, when he works as if there were three workers”. Another showed compassion for the experiences Syrians had had before they arrived in Adana: “However, there is another thing, ok you were born in Syria, you’re a Syrian kid. You see nothing but war. How would you grow up? I think they are not to blame for the existence of those big powers being there”. These two statements represented the only examples of positive perceptions. The rest of the responses can be classified as reflecting negative perceptions related to job scarcity and unfair competition, crime rate increases, and perceived lower moral standards.
Job scarcity and unfair competition
Almost all participants agreed on the lack of job opportunities in the town and stated that Syrians were to blame for job scarcity and unfair competition because their presence had lowered wages. This view is in line with the economic threat argument expressed in existing research literature. One participant talked in terms of expulsion: “I do not want to enter politics, but I believe that the Syrians should be returned to their country when problems are solved. So I do not approve that they will be working here so much. When there are so many unemployed men here, even Syrian women are working”. Another talked in terms of poorer employment prospects for locals:
They cannot find a job right now, Syrian workers work very cheap. I’m not talking about business establishments, I’m talking about employees as workers. As he says now, his friend is going to get 100 liras per day and he is willing to get 30 liras per day. The employer prefers him and he does not do his insurance either. In this regard, laborers working at a minimum wage level, fewer lower classes became unemployed in Adana. It was very influential to it.
Responsibility for this situation was also attributed to the government. One participant argued that “here the government’s policy made everything completely complicated. On top of that, the Syrians became salt, pepper. Or Adana is really rich under normal conditions”. Syrians are perceived to be particularly responsible for declining wages because they accept work with a very low wage rate, especially in the agricultural sector or in construction, for example. One participant painted a scene in which Syrian households absorb wages available to Adana’s workers:
Something like this is normally 60 liras of orange harvest, 70 liras for the Turks. But when Syrian can work for 30 liras, 35 liras […] what people do not understand is that the Syrians come with 10 people, 10 people come from a house, so this 30 lira becomes 300 liras actualized. They even bring their tiny kids, that’s the worst.
Another noted that this kind of process “has significant bad effects. The man who works in the building has lost his wages. There are Syrians working for 20 liras. They work without insurance. Due to these reasons, men (Turkish) are not preferred in construction work”.
Although participants accepted that, from the employer’s perspective, hiring a Syrian is an advantage because of the “cheap labour”, they insisted it was better to favour locals, since the local man “is one of us”:
“From the point of view of working, some of our friends recruiting them, I say to them, when we have our own people, do not give the job from the outsiders. That man will die from hunger? But the other one, the man is one of us”.
“It is not good to distinguish between Kurds and Turks, but the rights holders must not let them to work. We do not want to, for example, we say they certainly will not come”.
“No, I don’t have thus it can be. Maybe because I think differently. Talented Turkish youth living in Turkey for me have the priority. If they both have the same talent; I will prefer Turk. I want to help Syrians, but emotionally I will prefer Turks”.
Crime and betrayal
Participants believe that employers who work with Syrians will be more profitable in the short term, but will have regrets later as Syrians will “betray” them in the long term. One explained that he had not used Syrian labour:
I did not use it at all. Even if I am a day-to-day person, I have heard so many of my friends who have been betrayed. Stealing and stuff. The guys are coming to find out two things and try to open something for themselves. It’s cheaper because it’s Syrian, but I said to everybody they will be trouble for us.
Thus, one of the repeated themes is also about the “insecurity” Syrians create and robbery and crime Syrians are said to have committed, as the quotations below illustrate:
“We should not classify them all in one group, but when they first came, they created an insecure environment”.
“I cannot get on my own property without fear because of you. I cannot visit my own field after six o’clock”.
“They increased the crime rate”.
“For example, there is a park near our house, a very big area and a very nice cafe place in the middle. We were going for summer, for example, we could sit comfortably there. That park, all the grass in summer and winter are full of them, they are occupied, when we are dressed up and passing through them, they look at us, disturbing. So you do not understand their language, they are talking and they may be insulting us, maybe they are swearing”.
Lower moral standards
Syrians are accused of having lower moral standards and being unclean. One of the participants explained why he would not eat in a Syrian restaurant:
Let me put it in that way. Filtered us left pebbles. Our point of view is “Syrians are filthy”. I saw the crap, and I saw the cleaning. I also saw them perform an ablution.
It’s not about making good food. Do you know what my client? I eat this food, but is it clean?
According to participants, the presence of the Syrians in Adana also affects the locals’ behavior and moral corruption becomes contagious: “It was also my own idea, but of course people living here, as beggars, doing bad things, they start to pretend that they are Syrian, behave like Arabs and start to work more comfortably”. The blaming of Syrians for perceived hypocrisy and betrayal is part of ordinary language and the subject of many stories:
You go shopping at the market, most places are so it’s officially Syrian here, everyone is Syrian. You are going to market, I get my essential needs. Both from the kitchen and cleaning . . . You are looking at them, they are buying chocolate, ice cream, I do not know what. I said, “Life is nice to you”, beside the cashier. “Life is good for them”, I said, “Look at it. We are now forcing ourselves to live, look at this shopping, how many bags are filled. Always cookies, always fun”.
One participant explained that “I won’t let them work, whatever happens. Even if I accept, society would not. Who would come to my shop. Syrians are dirty”, while another saw Syrians displacement as proof of their unreliability: “I would not do anything with a man who left his country. He would also leave my job too”.
Adana’s residents regard these perceived differences as threats to “their” culture:
Money does not change your internal order. They are from another culture. They want to keep it alive. They don’t adapt. They don’t say it is the way we should accept. If they accept, believe me, money, then, is nothing.
This sense of threat is also linked to the persistence of the presence of Syrians in Adana’s communities:
Definitely they are to blame. If water is four lira here, in Syria two lira, one lira. They were very comfortable. These sectarian divides, Shi’ite or Sunni, if they escaped from that, I can understand. However, if you left that water . . . they have a free life. They left that life and they came. They have four or five kids. It has been four years war began. They have many kids. How comfortable they are.
These negative perceptions and scapegoating narratives led us to investigate different dimensions of perceptions about Syrians in Adana using quantitative data. When we developed our questions, we drew both on the research literature and on the findings that had emerged from our qualitative research. We were seeking to note negative and positive perceptions, but also to understand whether people thought the rights given and/or rights to be given to Syrians should be approved or denied.
In our analyses, we will focus on different dimensions of the attitudes participants expressed towards Syrian immigrants, namely negative perceptions which we conceptualize as a threat, positive perceptions, and the denial or approval of rights. Our independent variables include some demographic factors such as age, gender, level of education, household income, socio-economic status, work status, immigrant status, and Kurdish origin due to internal migration from the southeast of Turkey. In addition to these variables, we will also employ contact as a mediating variable.
Negative perceptions – threats
As the perception of threat from immigrants by individuals increases, they tend to have more hostile attitudes towards immigrants. The idea of an “economic threat” is largely based on a competitive model of society. According to this approach, increasing international labour flows result in increased feelings of insecurity. Citizens of the host country have to compete with newcomers, and this competition is vital especially for low-skilled labour. Since the traditional flow of migration brings unskilled, low-wage human capital (“Polish plumbers”) to developed countries; it is not surprising that unskilled segments of society have to compete with these immigrants. Consequently, anti-immigrant attitudes are widely observed among these citizens. However, recent studies have shown that economic threat involves something more than citizen’s simple rational calculations about the odds of being replaced by immigrants. Economic threat is based on perceptions, rather than just crude statistical facts. Blalock distinguished actual competition (macro- or meso-level socioeconomic conditions, such as the availability of scarce resources) from perceived competition, which is founded on subjectively perceived socio-economic threats. Actual competition is largely affected by economic factors such as unemployment, inflation, economic growth, the structure of the labour market and one’s relative position in it, and it is easily operationalized. Meanwhile, perceived competition includes some intangible factors and is more difficult to measure and operationalize.
First of all, threat perceptions should not be understood in a limited way in terms of threats towards the individual. Identities are constructed within society, and social identity, in terms of social category memberships, is self-defined. Individuals try to achieve a positive social identity and they tend to perceive their own groups as superior to outgroups. Consequently, any threat to their in-group is perceived as an individual threat, especially when this threat comes from an easily identifiable outgroup of the kind represented by immigrants. This social identification process also plays an important role in the emergence of “symbolic” threats.
Secondly, economic threat is not limited to competition in the market. If we use Blalock’s definition of threat, which understands it as “competition for scarce resources”, it also means competition for public goods, and especially social services provided by the welfare state. US-based research shows that citizens of wealthier societies worry about the additional demands immigrants create on social service budgets. Increased numbers of immigrants mean more pressure on welfare expenditure already limited by neoliberal economic policies.
In our questionnaire, we listed a group of threats and asked participants to what degree they agreed with them. Our list was developed using standard batteries of questions supplemented with additional arguments that had arisen during group discussions. The table below presents the percentages of those who agreed with these arguments.
According to these figures, more than 80 per cent of participants believe that Syrians are a burden to the health and education systems in Turkey. The percentage of those who perceived a threat in terms of job competition is about 83 per cent, while 79 per cent of participants agreed with the argument that Syrians are threatening Turkey’s moral values. More than three quarters of participants believed that Syrians had made crime rates rise in Turkey. Using the answers given to these questions, we constructed a threat index by using a factor analysis (Cronbach α=0.88).
Figure 1. Negative Perceptions – Threat (% of “Agree” and “Strongly Agree” Answers)
Although in the focus groups, positive perception about Syrians were very limited, we employed positive perceptions about Syrian immigrants as a variable in our analysis.
As part of our questionnaire, we asked participants three questions about their positive perceptions of Syrians. Only 10 per cent of participants agreed with the argument that Syrians recompense the economy for the health and education services provided to them by the Turkish government. A similar percentage agreed with the argument that Syrians contribute to a tolerant atmosphere in Turkey. Finally, the percentage of those who believe that Turkey needs the Syrian population to work in different economic sectors in Turkey is 7.3 per cent.
Alongside a threat index, we constructed an index of positive perception, again using the factor analysis method. The final index explains 66 per cent of the total variance (Cronbach α=0.74).
Figure 2. Positive Perceptions (% of “Agree” and “Strongly Agree” Answers)
Denial/approval of rights
While positive and negative perceptions will clearly affect the possibility of future co-existence, it is also important to take into account views about the denial and/or approval of rights already given or which might be given to Syrians in Turkey. The dataset we produced included three different questions that investigated respondents’ attitudes to immigrants’ rights. The first question offered two options. First, it asked whether equal rights should be given to legal immigrants living in the country in a move that would indicate acceptance of immigrants as part of the society and situate them as equals, or whether they should instead be given rights but not citizenship. The second question, which is similar to the first one, focused on the specific domain of education and used Marshall’s assertion that education is a socioeconomic right. Respondents were asked to state whether immigrants have a right to education or not. The third question offered a statement on citizenship and asked whether or not it might be considered as a right by immigrants.
Figure 3. Rights given to Syrians (% of “Partially Against” and “Totally Against” answers)
We asked these three questions in order to understand current trends that might indicate social approval for giving further rights to Syrians. The answers presented in Figure Four demonstrate that 92 per cent of respondents are against giving citizenship to Syrians and they also suggest that support for social rights for Syrians is relatively low: 76 per cent of respondents are against giving free access to education and 74 per cent of respondents do not support the idea of giving Syrians free health services. These figures show that people living in Adana are generally against providing citizenship status to Syrians in Turkey. The index of rights was constructed by using a factor analysis method and a single factor explaining 74 per cent of the total variation (Cronbach α=0.82) was obtained.
Figure 4. Contact (% of Yes Answers)
Figure Five shows the degree of contact that residents in Adana had with Syrians. According to this figure, 65 per cent of respondents live with Syrians in their neighbourhood, and, considering the relatively high number of refugees, this situation is not surprising. On the other hand, only eight per cent of respondents have worked together with a Syrian, and the percentage of those who have Syrian friends is 11 per cent. By using the answers given to these questions, we constructed an index of contact, from 0 to three which indicates respondents who gave positive answers to this question (Cronbach α=0.64).
The independent variables employed in the empirical part of this study are listed below.
Gender: According to our data, 51 per cent of participants were male and 49 per cent of them were female. We included this variable in our models in order to understand whether a gender gap existed or not.
Age: Age is another independent variable which was used in our models. Twenty per cent of participants were aged between 18 and 21, 33 per cent were between the ages of 22 and 31, and 22.4 per cent were between the ages of 32 and 41. Meanwhile, 24 per cent were over the age of 42.
Education: We included education level as an independent variable in our model. We found that 34 per cent of participants had not had a secondary education, 50 per cent had had a secondary education, and the percentage of those who had some higher education was 16 per cent.
Household income: According to our survey, the percentage of those who had an average family income less than or equal to 1500 TL (350 USD, official minimum wage) is 49 per cent, while 38 per cent of participants had an average income between 1501 and 3000 TL (350 USD-750 USD), and only 13 per cent of participants had an average monthly income greater than 3000 TL (750 USD).
Being immigrant or not: We also included a variable indicating the respondent’s immigration status. It is known that Adana is attractive for immigrants and our data shows that 15 per cent of respondents had immigrated to Adana.
Having Kurdish origin: In order to measure the ethnic status of respondents, we asked a question about which languages respondents were able to speak. We assigned the 15 per cent of participants who were able to speak Kurdish as having Kurdish origin.
Work status: Previous studies have shown that those who are most vulnerable to competition in the job market feel themselves most threatened. We asked a question about participants’ employment status. According to our findings, 14 per cent of participants were unemployed, while 16 per cent of participants were students. The percentage of housewives was 25 per cent; white-collar workers made up 12 per cent of the sample, and skilled workers formed 16 per cent. The unskilled workers’ share was six per cent, and 10 per cent of participants were working as craftsmen or were in other similar jobs.
Socioeconomic status: We used respondents’ ownership of household items to calculate an index of socioeconomic status, from 0 to 11, with a mean of six and standard deviation of two.
Table 1. Bivariate Analyses
|Household Income||< 1500 TL||0.05**||-0.01||0.00||0.89|
|1501 – 3000 TL||-0.11**||0.03||-0.04||0.80|
Table One presents the results of bivariate analyses or relationships between our independent and dependent variables. According to this table, the most important difference is observed between males and females. Our bivariate analyses show that men are more supportive of giving rights to Syrians and feel themselves less threatened. The same analyses show that the average level of social contact between Turkish and Syrian men is relatively and statistically high. Similarly, our bivariate analyses do not provide any evidence for differences in our dependent variables according to age groups or the levels of education of respondents. Observed differences are not statistically significant. Meanwhile, it seems that those who have an average household income of between 1501 TL and 3000 TL are relatively less supportive of giving more rights to the Syrians. Our post hoc analyses show that averages of lower and higher income categories are statistically higher (Scheffe p.<0.1 for both categories).
An important finding is that those who have an immigrant background have different attitudes towards the Syrians compared with other individuals. They feel themselves less threatened, they have higher positive perceptions about Syrians, and their contact score is significantly higher. The same table also shows that individuals of Kurdish origin have relatively higher contact scores. This situation is not surprising considering the fact that both Syrians and Kurdish people are living in the same neighbourhoods; nevertheless, we did not observe the same statistically significant differences in other variables.
Our bivariate analyses also show that the work status of respondents also matters in terms of threat perceptions and level of contact with the Syrians. According to our table of unqualified workers, unemployed people and housewives have relatively higher scores for threat perception; however these differences are not statistically significant. On the other hand, according to our analyses, skilled employees have significantly higher levels of contact compared to students (0.84 vs. 0.69, Scheffe p. <0.05).
In order to present a more accurate picture of the situation, we went on to conduct multivariate analyses to try to explain the differences in negative perception/threat, positive perception of refugees, contact with them, and views on their rights. Since we think that the approval/denial of rights is crucial to social integration, we started by using a Rights Index as a dependent variable and the variables discussed above as independent variables.
Table 2. Determinants of the Acceptance of the Rights (OLS Regression Results)
|Household Income||< 1500 TL||0.000||(.)||0.000||(.)|
|1501 – 3000 TL||-0.077*||(0.03)||-0.049+||(0.10)|
p-values in parentheses
Standardized beta coefficients; p-values in parentheses
+ p < 0.10 * p < 0.05, ** p < 0.01, *** p < 0.001
Table Two presents the results of a multivariate analysis where the acceptance of rights is used as the dependent variable. The first column shows the model with basic demographic and economic variables and a small number of variables have statistically significant effects after controlling for other variables. First, age matters. Respondents older than 42 are relatively less supportive of giving rights to Syrians in Adana (β=-0.14). Moreover, bivariate analyses showed that those who have a household income of between 1501 and 3000 TL are also against the granting of rights to Syrians (β=-0.08). Pre-existing populations have similarly negative scores on this issue (β=-0.06). None of the other variables have a significant effect on the independent variable. This picture may lead us to think that attitudes towards Syrians’ rights are almost independent of demographic and socioeconomic variables, including gender, the level of education, or having Kurdish origins. Even work status does not matter here, a finding which runs counter to the expectations that would arise from a theoretical approach focused on threat.
When we included three other variables – contact, negative perceptions/threat, and positive perceptions – the picture changed. First of all, these three variables have significant effects on the independent variable, and including these variables suppressed the effects of other variables; in other words, they mediated their effects. As we noted earlier, contact has a positive effect on the acceptance of rights given to Syrians: as a person’s contact with Syrians increases, his support for giving rights to Syrians also increases (β=0.12). The index of positive perceptions also has a positive coefficient (β=0.28), meaning that having positive perceptions about Syrians’ contributions to the labour market increases support for the rights given to them. Finally, as foreseen by the threat approach, as the perception of threat from Syrians increases, support for the rights given to them decreases (β=0.28), and this effect is the strongest in the model. It means that, despite the positive effects of contact and positive perceptions, perceived threat matters.
Table 3. Determinants of Perceptions of Threat and Positive Perceptions
|Threat||Threat||Positive P.||Positive P.|
|Household Income||< 1500 TL||0.000||(.)||0.000||(.)||0.000||(.)||0.000||(.)|
|1501 – 3000 TL||0.015||(0.69)||-0.000||(1.00)||-0.014||(0.71)||-0.011||(0.77)|
p-values in parentheses
Standardized beta coefficients; p-values in parentheses
+ p < 0.10 * p < 0.05, ** p < 0.01, *** p < 0.001
Table Three presents the findings of our multivariate analyses where threat and positive perceptions are used as dependent variables. The first model is composed of demographic and socio-economic variables and shows that being female or an immigrant matters. According to the table, female participants’ threat perceptions are relatively higher than those of their male counterparts, after controlling for other variables (β=0.09). Another significant difference is observed between immigrants and locals. A person who did not immigrate to Adana has a higher threat perception compared to immigrants (β=0.12). This finding may lead us to conclude that being immigrant contributes to individuals having empathy towards newcomers, thus they feel themselves less threatened. The above table shows that neither education, income, being Kurdish, socio-economic status, or the work status of respondents matters; they feel themselves equally threatened. It shows that the realistic threat approach is not valid in Adana’s context, because according to this approach, vulnerable persons – people with little education, the poor, unemployed, or unskilled – should have higher threat perceptions, and they are not evident in our case.
Our second model shows the results of our analysis after including contact as another independent variable. Including this variable suppressed the previously significant effects of gender; however, being immigrant still matters (β=0.09). Meanwhile, the contact index has a very strong significant effect (β=-0.23), meaning that those who have higher levels of contact with Syrians feel themselves less threatened.
From the perspective of positive perception, the picture is different. Among the demographic and socio-economic variables we included, only gender and education have a significant effect. Female participants have more positive perceptions about the contributions of Syrians to the job market and economy (β=0.09), whereas those who have a higher education level have more negative perceptions (β=-0.08). Locals also have more negative perceptions (β=-0.07) and these effects do not disappear after including the contact variable in the model. In this case, the contact variable does not have any statistically significant effect, meaning that having contact with Syrians does not contribute to positive perceptions.
Table 4. Determinants of Contact with Syrians
|Household Income||< 1500 TL||0.000||(.)|
|1501 – 3000 TL||-0.062+||(0.08)|
p-values in parentheses
Standardized beta coefficients; p-values in parentheses
+ p < 0.10 * p < 0.05, ** p < 0.01, *** p < 0.001
Our previous models show that the contact variable has an important role as a determinant of support for rights and threat perceptions. Table Four presents the determinants of the contact variable. First, gender still matters; female respondents have significantly less contact with Syrians (β=-0.07), compared to their male counterparts. Secondly, those who are located in the middle income categories also have limited contact with the Syrians (β=-0.06). Meanwhile, the model shows that locals’ contact level with Syrians is also limited (β=-0.11), and those who are not of Kurdish origin have relatively lower levels of contact (β=-0.07). The same model shows that students do not have contact with the Syrians (β=-0.09).
Figure 5. Interaction between Contact, Threat, Positive Perceptions and Denial/Approval for Rights (SEM Output, Beta Coefficients)
Figure five presents the interactions between our independent variables by using the Structural Equation Model (SEM). All of our three variables have significant effects on support for Syrians’ rights, and the most powerful one is the negative perceptions/threat variable, which has a negative coefficient. Coefficients related to contact and positive perceptions are positive. Contact’s effect on the negative perceptions/threat variable is negative, and this coefficient is positive for positive perceptions, which means that having contact with Syrians decreases the negative perceptions and increases positive perceptions. Through its effect on negative and positive perceptions, contact also has indirect effects on approval for rights. The total effect of contact is 0.26 and almost half of it comes from indirect effects (0.12).
As most countries in the world are currently both sending and receiving migrants, it is important to study how foreigners are perceived within different countries to understand commonalities and differences. In order to live in harmony with newcomers to a city in a given country, civil society, government institutions, community leaders, and the business community must address locals’ attitudes and perceptions and understand which factors play a role in negative and positive attitudes and perceptions. Within the limits of this article we have focused on the case of Adana to record scapegoating narratives there, and then we have examined those narratives through the lens of literature on anti-immigrant attitudes.
Although our bivariate analyses show that gender and immigrant background affect perceptions, we conducted three multivariate analyses and used SEM models to provide a more robust picture of perceptions towards Syrians in Adana. We began by using approval of rights as the dependent variable, and we found that age matters. Respondents older than 42 are relatively less supportive of giving rights to Syrians in Adana. This picture might lead us to conclude that attitudes to the rights given to Syrians are almost independent of demographic and socio-economic variables. Even work status does not matter, a finding which undermines expectations that perceptions about economic threat should dominate. However, when we include three other variables – contact, negative perceptions/threat, and positive perceptions – the picture changes. First of all, all three variables have significant effects on the independent variable and including these variables suppresses the effects of other variables; in other words, they mediate their effects. Contact has a positive effect on the acceptance of Syrians’ rights. Having positive perceptions about Syrians’ contributions to the labour market also increases support for the rights given to them. As expected, when the perception of threats from Syrians increases, support for their right to rights decreases, and this is the strongest effect in the model. It means that despite the positive effects of contact and positive perceptions, perceived threat matters.
In line with the literature, we found that perceived threat leads to varying levels of anti-immigrant sentiment. The findings of both the qualitative and quantitative study both cast doubt on the idea that the struggle for scarce resources or an increased crime rate emerge out of the realities of the field or personal experience. Instead, Syrians are blamed for economic and social problems which were there long before the arrival of the Syrians.
When we conducted another multivariate analysis which used threat and positive perceptions as dependent variables, we found that being female or immigrant matters. Female participants’ threat perceptions are higher compared to their male counterparts, after controlling for other variables. Another significant difference is observed between immigrants and locals. Immigrants have lower threat perceptions than people who did not immigrate to Adana. When we include contact as another independent variable, the previously significant effect of gender decreases; however being immigrant still matters. Controlled for other demographic factors, being an internal migrant also plays an important role that needs to be further studied as this experience leads to fewer negative perceptions despite the internal migrants being more vulnerable and insecure than many others in the population.
The contact index clearly has a very strong, highly significant effect, meaning that those who have higher levels of contact with Syrians feel themselves less threatened. According to the SEM findings, contact with Syrians decreases negative perceptions and increases positive perceptions. Contact also has indirect effects on approval for rights. Our findings show that any type of contact matters: whether contact is based on true or casual acquaintance, contact reduces people’s sense of threat and this finding needs to be further analyzed as it challenges the assumptions made in earlier literature.
When we look to the positive perception case, among the demographic and socio-economic variables we included, only gender and education have a significant effect. Female participants have more positive perceptions about Syrians’ contributions to the job market and economy, whereas those who have experienced higher education have more negative perceptions. These findings need to be studied further as they question the argument that the higher the level of a person’s education, the lower their level of intolerance is likely to be. Locals also have more negative perceptions and these effects do not disappear after including contact in the model. In this case, contact does not have any statistically significant effect, which raises the question as to why this is the case. Whether strength of belonging to a locality is a powerful determinant of anti-immigrant feeling or not must be scrutinized in future studies.
Table 5. Descriptives of Independent Variables
|Household Income||< 1500 TL||49.20%|
|1501 – 3000 TL||37.90%|
|Positive Perceptions||Average=0, sd.=1|
Arendt, H., The Origins of Totalitarianism, New York, Harcourt, Brace & Co., 1951.
Blalock, Hubert M., Toward a Theory of Minority-Group Relations, New York: John Wiley & Sons, 1967.
Blumer, Herbert, “Race Prejudice as a Sense of Group Position”, Pacific Sociological Review, Vol. 1, No 1, 1958, p.3-7.
Bogardus, Emory S., “Measuring Social Distance”, Journal of Applied Sociology, Vol. 9, No 2, 1925, p.299-308.
Buzan, Barry et al., Security: A New Framework for Analysis, Lynne Rienner, London, 1998.
Castles, Stephen and Miller, Mark J., The Age of Migration, Palgrave Macmillan, New York, 2009.
Gherghina, Sergiu and Chiru, Mihail, Physical Insecurity and Anti-immigration Views, COMPAS Working Paper, No 98, University of Oxford, 2012.
Çetin, İhsan, “Labor Force Participation of Syrian Refugees and Integration: Case of Adana and Mersin Cities”, Gaziantep University Journal of Social Sciences, Vol. 15, No 4, 2016, p.1001-1016.
Coenders, Marcel et al., Majority Populations’ Attitudes Toward Migrants and Minorities, Report for the European Monitoring Centre on Racism and Xenophobia, Netherlands, April 2003.
Curşeu, Petru Lucian, et al., “Prejudice Toward Immigrant Workers among Dutch Employees: Integrated Threat Theory Revisited”, European Journal of Social Psychology, Vol. 37, 2007, p.125-140.
Cukurova Development Agency, A Research on Potential Investment Areas for Adana, Turkish Development Bank, 2016.
Çukurova Young Businessman Association, “Global Trends in the Manufacturing Sector and Adana: Analyses and Proposal”, http://www.pglobal.com.tr/upload/%C3%87ukurova%20G
%C4%B0AD%20K%C3%BCresel%20ve%20B%C3%B6lgesel%20%C3%96l%C3%A7ekli%20Imalat%20Sanayi%20Analizi%2010062016.pdf (Accessed 2 January 2018).
Dedeoğlu, Saniye, “Tarımsal Üretimde Göçmen İşçiler: Yoksulluk Nöbetinden Yoksulların Rekabetine”, Çalışma ve Toplum, No. 1, 2018, p.37-67.
Esmer, Yılmaz, “Islam, Gender, Democracy and Values: The Case of Turkey (1990-2001)”, Pettersson, T., and Esmer, Y. (eds.), Changing Values, Persisting Cultures: Case Studies in Value Change, Brill, Leiden, 2008, p.275-304.
Fetzer, Joel S., Public Attitudes Toward Immigration in the US, France, and Germany, Cambridge University Press, Cambridge, 2000.
Freeman, Gary P., et al. (eds.), Immigration and Public Opinion in Liberal Democracies, Routledge, Abingdon, 2013.
Fasel, Nicole et al., “Facing Cultural Diversity: Anti-Immigrant Attitudes in Europe”, European Psychologist, Vol. 18, No 4, 2013, p.253-262.
Gorodzeisky, Anastasia, and Semyonov, Moshe, “Terms of Exclusion: Public Views toward Admission and Allocation of Rights to Immigrants in European Countries”, Ethnic and Racial Studies, Vol. 32, No 3, 2009, p.401-423.
Gümüş, Adnan. “Yasadan, kültüre işçi sömürüsü: Üçte bir ücretle Suriyeli işçi”, Evrensel, https://www.evrensel.net/yazi/80383/yasadan-kulture-isci-somurusu-ucte-bir-ucretle-suriyeli-isci (Accessed 2 January 2018).
Hainmueller, Jens, and Hiscox, Michael J., “Educated Preferences: Explaining Attitudes toward Immigration in Europe”, International Organization, Vol. 61, No 2, 2007, p.399-442.
Karyotis, Georgios, and Patrikios, Stratos, “Religion Securitization and Anti-Immigration Attitudes: the Case of Greece”, Journal of Peace Research, Vol. 47, No 1, 2010, p.43-57.
Kleemans, Marieke, and Klugman, Jeni, Understanding Attitudes Toward Migrants A Broader Perspective, UNDP, Human Development Reports Research Paper, October 2009.
Marshall,T. H. Citizenship and Social Class, Oxford, Oxford University Press, 1950.
McLaren, Lauren M., “Anti-Immigrant Prejudice in Europe: Contact, Threat Perception, and Preferences for the Exclusion of Migrants”, Social Forces, Vol. 81, No 3, 2003, p.909-936.
Milliyet, “Suriyeli Siğinmacilarin Yüzde 57’si Yoksulluk, Yüzde 31’i Açlik Sinirinin Altinda”, 4 September 2015, http://www.milliyet.com.tr/suriyeli-siginmacilarin-yuzde-57-si-adana-yerel
haber-956828 (Accessed 2 January 2018).
Ministry of Development of Turkey, Socio-Economic Development Ranking Survey of Provinces and Regions-2011, 2013. Ankara, Turkey
Ministry of Interior Directorate General of Migration Management, “Migration Statistics”, http://www.goc.gov.tr/icerik6/temporary-protection_915_1024_4748_icerik (Accessed 2 January 2018).
Rokeach, Milton, The Nature of Human Values, The Free Press, New York, 1973.
Semyonov, Moshe et al., “The Rise of Anti-Foreigner Sentiment in Europe Societies 1988-2000”, American Sociological Review, Vol. 71, No 3, 2006, p.426-449.
Scheve, Kenneth F. and Slaughter, Matthew J., “Labor Market Competition and Individual Preferences over Immigration Policy”, Review of Economics and Statistics, Vol. 83, No 1, 2001, p.133-145.
Schneider, Silke L., “Anti-Immigrant Attitudes in Europe: Outgroup Size and Perceived Ethnic Threat”, European Sociological Review, Vol. 24, No 1, 2008, p.53-67.
Social Security Institution, “Statistical Yearbook”, http://www.sgk.gov.tr/wps/portal/
sgk/tr/kurumsal/istatistik/sgk_istatistik_yilliklari (Accessed 2 January 2018).
Stephen, W. G. et. al., “Prejudice Toward Immigrants: An Integrated Threat Theory”, Journal of Applied Social Psychology, Vol. 29, 1999, p. 2221-2237.
Şahankaya Adar, Aslı, “Türkiye’de Yeni Prekarya Suriyeli İşgücü mü?”, Çalışma ve Toplum 1, 2018, p.13-35.
TUİK. Hanehalkı İşgücü Anketi 2016, TURKSTAT, Household Labor Survey, 2016.
Uyan Semerci, Pınar, and Erdoğan, Emre, Ben Kendim Büyüdüm Demiyorum, Research Report, 2016, http://www.bilgi.edu.tr/site_media/uploads/files/2017/03/30/goc-merkezi-ben-kendim-b-y-d-m-demiyorum-rapor.pdf (Accessed 2 January 2018).
Uyan Semerci, Pınar, and Erdoğan, Emre, “Guests to Neighbours: the Difficulty of Naming Syrians in Turkey”, Refugee Watch, Vol. 48, 2016, p. 20-34.
Weber Hannes, “National and Regional Proportion of Immigrants and Perceived Threat of Immigration: A Three-Level Analysis in Western Europe”, International Journal of Comparative Sociology, Vol. 56, No 2, 2015, p.116-140.
 Freeman et al., “Immigration and Public Opinion in Liberal Democracies”.
 Castles and Miller, The Age of Migration.
 Kleemans and Klugman, “Understanding Attitudes toward Migrants”.
 Coenders et al., “Majority Populations’ Attitudes Toward Migrants and Minorities”; Kleemans and Klugman, “Understanding Attitudes toward Migrants”.
 Stephan et al., “Prejudice toward Immigrants: An Integrated Threat Theory”’; Curşeu et al., “Prejudice Toward Immigrant Workers among Dutch Employees”.
 Buzan et al., Security: A New Framework for Analysis; Karyotis and Patrikios “Religion Securitization and Anti-Immigration Attitudes”; Chiru. and Gherghina, Physical Insecurity and Anti-immigration Views.
 Gorodzeisky and Semyonov, “Terms of Exclusion”; Semyonov et al., “The Rise of Anti-Foreigner Sentiment”; Scheve and Slaughter, “Labor Market Competition”.
 Blumer, “Race Prejudice as a Sense of Group Position”.
 McLaren, “Anti-Immigrant Prejudice in Europe”; Fetzer, Public Attitudes toward Immigration; Weber, “National and Regional Proportion of Immigrants”.
 Schneider, “Anti-Immigrant Attitudes in Europe”; Green, “Facing Cultural Diversity”; Weber, “National and Regional Proportion of Immigrants”.
 Syrians in Turkey are not officially accepted as “refugees”; instead they have “temporary protection status”. See Uyan Semerci and Erdoğan, “Guest to Neighbors”.
 TURKSTAT, Household Labor Survey.
 Ministry of Development, “Socio-Economic Development Ranking Survey”.
 Çukurova Young Businessman Association, “Global Trends in the Manufacturing Sector and Adana”.
 Social Security Institution, “Statistical Yearbook”.
 Çukurova Young Businessman Association, “Global Trends in the Manufacturing Sector and Adana”.
 Here, we are using the official definition of unemployment, which excludes individuals who are not looking for jobs. Labour force participation rates are significantly different between men and women (45/19 per cent between ages 15-19; 76/39 per cent between ages 20-24; and 89/42 per cent between ages 25-34).
 Cukurova Development Agency, “A Research on Potential Investment Areas for Adana”, 52.
 Ibid., 27.
 Uyan Semerci and Erdoğan, Ben Kendim Büyüdüm Demiyorum!
 Ministery of Interior, Directorate of Migration “Migration Statistics”/
 Milliyet, “Suriyeli Siğinmacilarin Yüzde 57’si Yoksulluk, Yüzde 31’i Açlik Sinirinin Altinda”.
 Çetin, “Labor Force Participation of Syrian Refugees and Integration”.
 Adar, “Türkiye’de Yeni Prekarya Suriyeli İşgücü mü?”.
 Dedeoğlu, “Tarımsal Üretimde Göçmen İşçiler”.
 Freeman, Gary P., et. al., Immigration and Public Opinion in Liberal Democracies.
 Blalock, Hubert M., Toward a Theory of Minority-Group Relations.
 Hainmueller, Jens et al., “Educated Preferences”.
 Hannah Arendt, The Origins of Totalitarianism.
 T. H. Marshall, Social Class and Citizenship.