While status has typically been approached in social sciences through the notion of objective status (e.g., the legal status of a group defined by country of birth, or objective indicators of racism and discrimination), for social psychologically oriented research, the notion of subjective status is especially relevant. Subjective status has been defined as the perceived position of oneself and one’s ingroup in the prevailing social hierarchy (e.g., Jackman & Jackman, 1973; Demakakos, Nazroo, Breeze, & Marmot, 2008). In previous social and acculturation psychological research on immigrant integration, subjectively perceived group status has been largely studied along two separate lines of research that have focused either on material indicators of disadvantage such as economic and social well-being and socio-economic status (SES; e.g., Michalikova & Yang, 2016) or its social indicators such as social distance or ethnic discrimination (e.g., Jasinskaja-Lahti, Liebkind, & Solheim, 2009).
Importantly, not everyone experiences their ingroup’s status in the same way (for subjective deprivation, see Smith, Pettigrew, Pippin, & Bialosiewicz, 2012), and the effects of low status can be group and context specific. For example, different immigrant groups may react to stressors like perceived discrimination in a different degree (Jasinskaja-Lahti, Liebkind, & Perhoniemi, 2006). Thus, belonging to a disadvantaged immigrant group is not inevitably reflected in poor integration outcomes. Moreover, previous studies also acknowledge that SES and ethnic discrimination are closely intertwined, with members of many minority groups experiencing both lower SES and higher levels of discrimination than members of majority groups (Chen, Martin, & Matthews, 2006). Indeed, as we will discuss next, especially the cumulation of perceived low status is a risk for social integration of immigrants and other minority group members.
In this study, we focus on the national identification of Russian-speaking immigrants in Finland and Estonia, and aim to complement previous research by examining the interaction between two indicators of subjectively perceived group status: perceived socio-economic status in relation to that of the majority group, and perceived ethnic discrimination targeted towards the ethnic ingroup. More specifically, we are interested in how the cumulation of perceived low status potentially hinders immigrant integration in the form of national disidentification.
Identity Reactions to Perceived Cumulative Group Disadvantage
In intergroup context, it is not individual but rather group status that is at stake. Perceived group status often corresponds to the group’s position in the societal ethnic hierarchy (Hagendoorn, 1995). Furthermore, even if an individual does not perceive being socio-economically disadvantaged, perceived group discrimination may still have psychological consequences due to the internalization of ingroup members’ experiences (Swim & Stangor, 1998). Low status also has societal consequences. If immigrants and other minority members perceive that their ingroup is treated with less respect than it deserves, they will be less motivated to belong to the social system that they perceive to be unfair (cf., group engagement model, Tyler & Blader, 2003). In this case, they are likely to react to this rejection by distancing themselves psychologically from the national majority group and society which it represents (for studies on national disidentification, see Jasinskaja-Lahti et al., 2009, and Martinovic & Verkuyten, 2012; for a distinction between dis- and nonidentification, see Becker & Tausch, 2014).
Previous research has shown that while an advantageous position within society leads to even greater advantages (e.g., DiPrete & Eirich, 2006), disadvantage also tends to accumulate and hinder immigrant integration (e.g., Brekke & Mastekaasa, 2008). For example, labor market discrimination based on ethnic group membership makes it difficult for immigrants to achieve a better economic position in society (e.g., Larja, Warius, Sundbäck, Liebkind, Kandolin, & Jasinskaja-Lahti, 2012, p. 12). Cumulative disadvantage can be understood both in terms of growing inequality over time (i.e., when current levels of disadvantage directly predict future levels) or in terms of disadvantage in different spheres of life (e.g., economically, educationally and socially) (DiPrete & Eirich, 2006; see also Riosmena, Everett, Rogers, & Dennis, 2015). According to Mannila and Reuter (2009), for social exclusion to be a risk factor, one must experience disadvantage in at least two spheres of life (e.g., unemployment, subjective poverty or perceived poor health). In their study conducted among Russian and Estonian immigrants in Finland, they found evidence of accumulating risks, especially among Russian speakers who had been living in Finland either less than five years or longer than ten years. This results shows how integration is not necessarily linear. More research is needed to examine the reasons for and ramifications of cumulative disadvantage among newcomers, as well as among groups who have lived in the country of migration for a longer time.
Following Mannila and Reuter (2009) and Riosmena and colleagues (2015), in this study, we define cumulative disadvantage as the experience of both high ethnic discrimination and low socio-economic status of the ingroup. To our knowledge, there are no previous studies on how these two types of subjectively perceived group status jointly affect feelings of national belonging in different societal contexts. As a recent example of research relevant for the present study, Hadjar and Backes (2013) compared the subjective well-being of immigrants and native populations in Europe, utilizing the European Social Survey. In their multi-level study, they included subjectively perceived economic deprivation (i.e., feeling about household’s income) and a macro-level indicator of xenophobia (attitudinal climate within society), but immigrants’ personal experiences of discrimination against the ingroup were not evaluated. To respond to this gap in existing research, we study Russian speakers in a relatively under-studied Baltic context. We focus on two countries, Finland and Estonia, where Russian speakers represent the biggest ethnic minority group with an immigrant background. In both contexts, Russian speakers are quite highly educated and can be considered as quite similar culturally to the national majority groups. However, for historical and political reasons, Russian speakers are in both countries socially disadvantaged, as will be further discussed in the next section.
Contexts of the Study
As an intergroup context, Finland is characterized by a relatively small but increasing immigrant population (ca. 7%, Statistics Finland, 2017). Historically, the Russian-speaking population in Finland has been quite small, but it started to grow considerably in the 1990s after the collapse of the Soviet Union. Today, around 21% of those speaking a foreign language as their mother tongue are Russian speakers (Statistics Finland, 2017). Largely due to historical conflicts between Finland and the former Soviet Union, intergroup relations between the Finnish majority and Russian-speaking immigrants have been characterized by considerable discrimination towards this minority (Larja et al., 2012). Despite being highly educated, the employment rate of immigrants from Russia and the former Soviet Union is considerably lower (61%) than that of, for example, Estonians (76%) (Nieminen, Sutela & Hannula, 2015). Longitudinal studies conducted among different groups of Russian-speaking immigrants have shown the detrimental effects of perceived ethnic discrimination on identification with Finnish society (Jasinskaja-Lahti et al., 2009; Jasinskaja-Lahti, Mähönen, & Ketokivi, 2012).
The Estonian Republic regained independence in 1991 after being part of the Soviet Union for half a century. Russian speakers currently form the largest minority group in Estonia, constituting around 25% of the total population according to the 2011 census. Although a small number of Russian speakers have lived in Estonia for centuries, the majority arrived during the Soviet period. Because of a number of structural factors and policy decisions in 1990s, the status of the Russian speaking population in Estonia has declined considerably during the period of independence. One reason is citizenship; most Estonian Russians found themselves to be non-citizens after the adoption of the Citizenship Act in 1993 (Barrington, 1995). This forced most of them to go through the process of naturalization to receive Estonian citizenship. Currently, there are still about 80,000 stateless individuals in Estonia, and only about a half of Estonian Russians hold Estonian citizenship. This has a negative impact on their status in the labour market and, more generally, society (Vetik & Helemäe, 2011). As a result, the Russian-speaking minority has reacted by forming strong defensive identities (Nimmerfeldt, 2009; Vetik & Helemäe, 2011).
Aim and Hypothesis
Based on the research literature discussed above, we expect perceived cumulative disadvantage to hinder immigrant integration. Thus, we hypothesize national identification to be the lowest when perceived high ethnic discrimination coincides with perceived low SES of the ingroup. Conversely, we expect national identification to be the highest when there is cumulation of relatively high perceived ingroup status (i.e., when ingroup’s SES is perceived as equal to that of the majority group, and ethnic discrimination is perceived as low). We test this hypothesis in two contexts, Finland and Estonia. Given the lack of previous research on the interactive effects of different types of subjective status on national identification in different societal contexts, we explore the possible contextual differences in the associations studied.
The analyses are conducted while controlling for the effects of ethnic identification. The reason for this is that ethnic and national identities of ethnic minority group members are often found to be interrelated (e.g., Benet-Martinez & Haritatos, 2005), and disadvantage may be observed to a higher degree among highly identified ingroup members (e.g., Major, Quinton, & Schmader, 2003). Additionally, demographic background variables age and gender are controlled for. To ensure the robustness of the findings, the analysis is also conducted without these control variables.
Participants and Procedure
This study is a part of the international project Mutual Intercultural Relations in Plural Societies (MIRIPS).1 The Finnish data was collected between June and November 2012 with a postal survey. The sampling was conducted by the Finnish National Population Register Centre, and the sampling criteria were that participants spoke Russian as their mother tongue, were 18 to 65 years old, lived in mainland Finland, were born in the former Soviet Union or the Russian Federation, and had resided in Finland for at least five years. The response rate was 39%, which can be considered less than optimal but typical for postal surveys conducted in Finland. The final sample consists of 316 individuals (77% female; mean age 45, ranging from 18 to 65 years of age).
The Estonian data was collected in May 2015 among Russian-speaking adult population. Proportional random sampling was based on the 2011 census data in Estonia. Eighty-one percent of Russians in the sample were born in Estonia and represented the second or third generation of Russians in the country. More than 90% said that they have lived in Estonia more than 25 years (i.e., the whole post-Soviet period). The survey questionnaires were completed during face-to-face interviews, and the response rate was 65%. The final sample consists of 501 participants (54% female; mean age 48, ranging from 18 to 74 years). Participation in both national surveys was voluntary and confidential.
Regarding the demographic characteristics of the two samples, participants from Finland were slightly younger than those from Estonia, t (777) = –3.38, p = 0.001. Also, the proportion of men to women was statistically and significantly smaller among Russian-speaking immigrants in Finland than among Russian-speaking immigrants in Estonia (χ2(1) = 44.61, p < 0.001). As for socio-economic characteristics, the level of education was measured differently in the two samples. In Finland, educational background was measured in terms of years of education; while in Estonia, participants were asked to report their highest level of education. Russian speakers in Finland had spent on average 15.37 (SD = 3.19) years in education, ranging from 7 to 26 years in education. Among Russian speakers in Estonia, 7% had tertiary education, and 9% had elementary education (including vocational) and 84% had secondary education (including specialized education). Thus, both samples can be regarded as relatively well educated. Russian speakers in Finland had experienced longer phases of unemployment (M = 50 months, SD = 43 months) than Russian speakers in Estonia (M = 18 months, SD = 22 months), t (332) = 6.765, p < 0.001. There was no statistically significant difference between the subjective evaluation of one’s economic situation (i.e., if participants felt that they had enough money for living), t (798) = –0.396, p = 0.692, nor in the correspondence of current work and educational background between participants in Finland and Estonia, t (524) = 0.022, p = 0.982.
The measures used in the analyses were identical in both contexts. The response scales for all items ranged from one (completely disagree) to five (completely agree). The questionnaire was back-translated from English into Russian by native speakers of the Russian language.
Perceived ethnic discrimination targeted against one’s ingroup by the national majority was measured with two items adapted from Berry, Phinney, Sam and Vedder (2006) “In my opinion, <OUTGROUP> has treated <INGROUP> unfairly or otherwise negatively,” “I think that <OUTGROUP> does not accept <INGROUP>.” Higher scores reflect higher levels of perceived discrimination (Spearman-Brown reliability statistic for two-item scales: Estonia: ρ = 0.90; Finland ρ = 0.74).
Perceived socio-economic status of the ingroup was measured with one item adapted from Louis, Duck, Terry, Schuller and Lalonde (2007), which taps participants’ perception of their ingroup’s socio-economic status as compared to that of the national majority group. The item was: “Compared to <OUTGROUP>, the social and economic standing of <INGROUP> is much worse.” To facilitate the interpretation, the scores were reverse-coded so that the higher the score, the higher the perceived socio-economic status of the ingroup in comparison to the majority group.
National identification was measured with four items adapted from Mlicki and Ellemers (1996) and Phinney and Devich-Navarro (1997). The items covered the affective aspect of participants’ identification with the mainstream society of their country (“I feel that I am a member of Estonian/Finnish society,” “I am glad to be a member of Estonian/Finnish society,” and “I am proud to be a member of Estonian/Finnish society”), as well as commitment to this society (“It is important to me to be a member of Estonian/Finnish society”). Higher scores reflect higher national identification of participants (Estonia: α = 0.92; Finland: α = 0.90).
Corresponding to the measure of national identification, ethnic identification (control variable) was measured with four items adapted from Mlicki and Ellemers (1996) and Phinney and Devich-Navarro (1997). The items covered the affective aspect of ethnic identification (“I feel that I am Russian,” “I am glad to be Russian,” and “I am proud to be Russian”) and commitment to one’s ethnic group (“It is important to me that I am Russian”). Higher scores denote higher ethnic identification of participants (Estonia: α = 0.87; Finland: α = 0.90).
Missing data (4% of values) were imputed with with the maximum likelihood (ML) estimation method. The hypotheses were tested with conditional process analysis (Hayes, 2013), using the PROCESS tool for SPSS. The strength and significance of the conditional effect were assessed with a non-parametric bootstrapping method using 5,000 resamples. All regression coefficients in the study are reported in an unstandardized form (B).
We conducted confirmatory factor analyses (CFA) to test the three-factor structure of ethnic identification, national identification and group discrimination items, as well as to test factorial equivalence of these measures across the two groups. These analyses were run in Mplus 5 (Muthén & Muthén, 1998–2010) with the MLM estimator. Goodness of fit indices generally indicated that the three-factor model had a good fit with the data for Russians in Estonia: χ2(32) = 46.291, p = 0.049; comparative fit index (CFI) = 0.995; Tucker-Lewis index (TLI) = 0.993; root mean square error of approximation (RMSEA) = 0.030; standardized root mean square residual (SRMR) = .017. This was true also for the Finnish data: χ2(32) = 57.830, p = 0.003; CFI = 0.983; TLI = 0.976; RMSEA = 0.051; SRMR = 0.036. Each item also had a moderate or high loading on its respective factor in the two groups (factor loadings varied from 0.52 to 0.93 for Russians in Estonia and from 0.59 to 0.96 for Russians in Finland).
The test of the factorial equivalence of the three measures across the two groups consisted of several steps (see Kline, 2016, p. 394–402). We found support for partial metric invariance (equality of factor structure and factor loadings) across Russians in Estonia and in Finland. That is, after releasing an equality constraint for one factor loading (an item “I feel that I am Russian” of the ethnic identification measure), the Satorra-Bentler scaled chi-square difference of ΔMLM χ2(6) = 8.150 between the partial metric invariance and the configural invariance models became statistically non-significant (p > 0.05), indicating no need for further releasing any other factor loading equality constraints in the model. As the detected non-invariance was caused only by the control variable and not by any of the predictor and outcome variables, we decided to proceed with the main analysis.
Correlations, means and standard deviations of the variables used in the analyses are shown in Table 1. Russian-speaking immigrants in Finland identified less with Russians, t (427) = –9.09, p < 0.001, and more with Finnish society, t (815) = 4.45, p < 0.001, than Russian-speaking immigrants in Estonia identified with their ethnic ingroup and Estonian society.2 As regards perceived socio-economic status and perceived ethnic discrimination, their mean levels were quite close to the midpoints of the scales in both countries. Russian-speaking immigrants in Finland perceived, however, their ingroup’s socio-economic status to be higher in relation to the majority than their counterparts in Estonia, t (577) = 12.87, p < 0.001. As regards the level of ethnic discrimination, Russian-speaking immigrants in Finland perceived it to be lower than Russian-speaking immigrants in Estonia, t (815) = –4.10, p < 0.001.
|3. Ethnic identification||3.83||1.06||1||–0.07||0.02||–0.16||**|
|4. Socio-economic status||2.92||1.14||1||–0.30||***||0.06|
|5. Ethnic discrimination||2.73||1.01||1||–0.26||***|
|6. National identification||3.76||0.91||1|
|3. Ethnic identification||4.39||0.55||1||–0.04||0.07||–0.06|
|4. Socio-economic status||1.93||0.94||1||–0.33||***||0.28||***|
|5. Ethnic discrimination||3.02||0.95||1||–0.30||***|
|6. National identification||3.47||0.87||1|
To examine cumulative (dis)advantage, the focus of our study, we compared the proportion of participants experiencing cumulative (dis)advantage by splitting the scales measuring perceived discrimination and ingroup’s socio-economic status from their midpoints. We noticed a statistically significant difference between the contexts studied, χ2(1) = 39.06, p < 0.001. In Estonia, there were relatively more cumulatively disadvantaged Russian-speaking minority group members (50% disadvantaged and 50% advantaged) than in Finland (19% disadvantaged and 81% advantaged). We argue that this speaks for a higher level of inequality experienced by our participants living in Estonia.
We began the testing of the hypothesis of this study by exploring possible between-country differences, and found a statistically significant three-way interaction between perceived socio-economic status, perceived ethnic discrimination, and country (B = –0.11, p = 0.048, 95% CI [–0.22, –0.00]). The results are presented in Table 2 and Figure 1.
|Gender (0 = male)||0.15||*||00.06||2.43|
|Socio-economic status (SES)||–0.19||00.11||–1.67|
|Ethnic discrimination (ED)||–0.42||***||00.12||–3.45|
|Country (C; 0 = FIN; 1 = EST)||–1.44||**||00.48||–3.01|
|SES × ED||0.06||00.04||1.52|
|SES × C||0.50||**||00.16||3.07|
|ED × C||0.32||*||00.15||2.10|
|SES × ED × C||–0.11||*||00.06||–1.98|
|F change for R2||14.20***|
Among Russian speakers in Estonia, higher perceived socio-economic status was related to higher national identification. This association was slightly stronger among those who perceived low levels of ethnic discrimination (1 SD below the mean; B = 0.22, p < 0.001, 95% CI [0.11, 0.33]) than among those who perceived high levels of ethnic discrimination (1 SD above the mean; B = 0.12, p = 0.035, 95% CI [0.01, 0.24]), thus confirming our hypothesis about cumulative (dis)advantage. In Finland, we found no evidence of interaction: perceived socio-economic status was unrelated to national identification both among those who perceived low levels of ethnic discrimination (1 SD below the mean; B = –0.08, p = 0.152, 95% CI [–0.18, 0.03]) and among those who perceived high levels of ethnic discrimination (1 SD above the mean; B = 0.04, p = 0.505, 95% CI [–0.08, 0.17]). In Finland, we only found a direct negative effect of perceived ethnic discrimination. A higher level of ethnic discrimination perceived by Russian-speaking immigrants in Finland was associated with lower identification with Finnish society (B = –0.42, p < 0.001, 95% CI [–0.66, –0.18]). Thus, we found support for our cumulative (dis)advantage hypothesis only in Estonia.
The aim of this study was to show how feelings of national belonging among immigrants and ethnic minority members depend on their perceptions of cumulative disadvantage (i.e., their ingroup’s societal standing reflected in ethnic discrimination and socio-economic situation). The main contributions of the study are related to the simultaneous examination of these two different indicators of subjective group status, and to the examination of the same ethnic minority group in two relatively under-studied intergroup contexts, Finland and Estonia.
Several valuable studies on different indicators of perceived group disadvantage have showed us the detrimental effect of low social status and perceived discrimination (e.g., Jasinskaja-Lahti et al., 2009; Kulich, Lorenzi-Cioldi, & Iacoviello, 2015; Smith et al., 2012) on the development of intergroup relations. Complementing this research literature, the present results speak for the importance of studying the role of cumulative disadvantage in immigrant integration, and suggest that reactions to group-based disadvantage could also be context- and/or group-specific. According to our descriptive results, the proportion of participants experiencing cumulative disadvantage (vs. advantage) was higher in Estonia than in Finland. Moreover, in Estonia, there was a clear sign of cumulative disadvantage predicting national identification of Russian speakers. Namely, their national identification was lower when they perceived both disadvantages (i.e., high level of ethnic discrimination and low socio-economic standing of their ingroup). However, in Finland, only perceived ethnic discrimination (but not socio-economic status) was associated with lower national identification. As a post-hoc explanation to this between-country difference, we consider two inter-related factors to be relevant: 1) socio-historical differences between the intergroup contexts studied and the consequent differences in the experiences of Russian speakers in Finland and Estonia; and 2) differences between long-established and more recent immigrant groups (cf. Mummendey & Wenzel, 1999, on the role of intergroup context in intergroup evaluations).
First, despite the many similarities in their backgrounds, different socio-historical circumstances probably contribute to a different perception of group status among Russian speakers in Finland and Estonia. Specifically, while Russian speakers in Finland are voluntary immigrants from Russia, Russians in Estonia are internal migrants from the times when the country was a part of the Soviet Union. As a result, Russians in Estonia have lived there for generations and do not regard themselves as immigrants at all, as they did not cross state borders, but the borders crossed them (see Brubaker, 1996). Moreover, as discussed above, only about half of Estonian Russians are citizens of Estonia, while the rest are either stateless persons or citizens of Russia. Those who are citizens of Estonia have mostly acquired citizenship through the naturalization process. However, the demand to go through such a process for those who did not migrate, and even for those who have been born in Estonia, is widely regarded as discriminatory and unjust (Vetik, 2012). Thus, the intergroup context of Estonia can be regarded as more conflictual than that of Finland. As we will next discuss, we argue that an established minority group facing pervasive disadvantage might be especially prone to react to it with disidentification from the national majority.
Namely, and as our second point, it is possible that perceived low status of the ingroup is particularly difficult to accept for bigger and older minority groups such as Russian speakers in Estonia. They have strived for generations to achieve a better status, but still experience negative treatment and perceive that their socio-economic status does not match that of the national majority. As personal- and group-level investments are found to be related to feelings of ownership and commitment (for organizational contexts, see Brown, Lawrence, & Robinson, 2005; for immigrant integration, see Verkuyten, Sierksma, & Thijs, 2015), we claim that minority group members’ problems in identifying with broader society reflect a tension between investments to society and negative outcomes. In contrast, the more recent immigrant group of Russian speakers in Finland may perceive their relative socio-economic status to be quite high, still regard their situation optimistically and perceive possibilities for social mobility. Recently, Owuamalam, Rubin and Issmer (2016) showed that strong ingroup identifiers tend to justify system inequalities if they hope that their ingroup status could improve in the future (see also Hadjar & Backes, 2013, about expected improvement in socio-economic status). This might explain why Russian-speaking immigrants in Finland perceive only ethnic discrimination as a threat, and react to it with lower national identification.
As regards to directions for future research, it would be interesting to include a measure of a more politicized form of collective identification than the general measure of civic national identification used in this study. In previous research, national identification is seen as a prerequisite for the integration and political activation of members of disadvantaged immigrant groups (e.g., Simon & Grabow, 2010). However, given the difficulties of Russians in Estonia in claiming a full political citizenship in the country (Vetik, 2012), it might be that the general measure of civic national identification used in this study did not enable us to fully grasp identity reactions to perceived unjust treatment from the state. As recently pointed out in a study by Verkuyten (2016), the meaning of national identity depends on how it can be used in each societal and political context. The fact that we have used a general measure of civic identification with society might explain why the level of national identification was relatively high among both Estonian and Finnish Russians, despite the highly politicized intergroup context in Estonia and the lack of equal opportunities for immigrants in Finland. In future studies, it could be interesting to differentiate between a ‘societal identity’ and a ‘state identity.’ We would expect identity reactions to perceived low ingroup status and cumulative disadvantage be stronger when assessed through political state identity (i.e., identification with the state that is regarded as unjust), compared to assessment of societal identity (i.e., identification with civic society and the people living in it) (cf., Becker, Tausch, Spears, & Christ, 2011).
Further, the hypothesis of this study should be tested in other intergroup contexts where the content of national identity is not as ethnocentrically construed as in Finland (e.g., Mähönen et al., 2015) and Estonia (Nimmerfeldt, 2009; Vetik, 2012), and among immigrant groups that are culturally not as close to the national majority group as Russian speakers in these two countries. As pointed out in studies on ingroup projection among both majority and minority group members, claiming the attributes of the ingroup to be norms of a more inclusive category has negative ramifications on intergroup relations and social cohesion (e.g., Mummendey & Wenzel, 1999; Verkuyten & Martinovic, 2016). In future studies with more contexts for comparison, it would be interesting to see how subjective status perceptions are possibly affected by perceived ingroup prototypicality and indispensability in different immigration contexts. It would also be interesting to include a measure of social dominance orientation when studying reactions to perceived low status and cumulative disadvantage in different intergroup contexts. For example, Levin (2004) showed that even though preference for social inequality is generally quite robust from situation to situation, it may still vary according to how the intergroup situation is framed and which outgroup the ingroup is juxtaposed with (for further discussion, see Sidanius, Pratto, Van Laar, & Levin, 2004).
Regarding the limitations of this study, the most significant relates to the cross-sectional design. As a strong sense of attachment to host society has been found to undermine concern for one’s disadvantaged ingroup (Kulich et al., 2015; Mähönen & Jasinskaja-Lahti, 2015), it is possible that immigrants with high national identification evaluate the position and treatment of their ethnic ingroup more positively than immigrants with weaker ties to the majority group. Future experimental and longitudinal research should keep both causal directions in mind. Also related to the interpretation of the findings, it should be noted that the magnitude of the effects (evaluated in unstandardized betas) was not great, and the regression model with the two countries explained only 15% of the total variance. However, the associations found are still substantial and theoretically meaningful. Given the theoretical broadness of the concept of national identification—the dependent variable of the present study—it is understandable that many other factors outside of those studied here account for its variation.
It is also important to point out that our primary research question was based on the similarity of the two samples, and the data was not optimal for making post-hoc comparisons between them—especially due to the time lag between the data collection in Finland and Estonia. However, we are not aware of any big changes in legislation that would have changed the formal situation of the minority groups studied between 2012 and 2015. Instead, the integration and status of Russian speakers has been a recurring topic in societal discussion for a long time in both countries.
As a further limitation, perceived ethnic discrimination was assessed only with a two-item scale, and perceived socio-economic status of one’s ingroup was assessed with a single-item measure. Even though these measures were adapted from previous research, more extensive measurement of status is recommended in future studies. Finally, we were not able to control for the educational background and length of stay or the immigrant generation of our participants. In future research, the samples should optimally be matched to be as identical as possible.
To sum up, based on the research information gained so far, it is clear that perceived status inequalities provoke a backlash in the form of immigrants’ psychological distancing from the national ingroup. However, the results obtained suggest that the magnitude and mechanisms of this distancing differ between more established and more recent immigrant groups, or between intergroup contexts with different socio-historical backgrounds. While perceived discrimination remains to be one of the most severe forms of social exclusion, there is also cumulation of disadvantage which undermines ethnic minority members’ possibilities to feel that they are full members of the national community. This is something to keep in mind also when developing index measures of citizenship and integration policy (see Goodman, 2015). Despite the importance of objective macro-level indicators, we should not forget to include also subjective perceptions of social and economic disadvantage in our models on the precursors of immigrant integration. Thus, we call for future interdisciplinary research to examine the cumulative effects of subjectively perceived ethnic discrimination and socio-economic status with a more extensive comparative design. To promote equal opportunities, and to avoid a culmination of conflicts between immigrant groups and host nationals, it would be critical to acknowledge minorities’ claims for recognition and implement anti-discriminatory policies more effectively at the grass-root level of everyday life. Lessons learned from the integration problems of long-established minorities such as Russian speakers in Estonia should be remembered also when targeting policies at more recent immigrant groups such as Russian speakers in Finland. This is especially important for countries inhabited by diasporic communities; messages of unequal treatment are transmitted beyond national borders (e.g., Weidmann, 2015). To avoid spreading of ethnic tensions, it is important to prevent the formation of monolithic, transnational ethnic hierarchies.