Conditions and constraints of political participation among Turkish students in Germany

Abstract What motivates conventional and unconventional participation in the political realm? Specifically, what are the facilitators and constraints of political participation in the views of Turkish migrants? The present study explored these conditions using semi-structured interviews with a small number of university students of Turkish descent in Germany, as these may be more likely to become spokespersons of the Turkish people in Germany than other members of their ethnic minority in-group. The interviewees draw attention to characteristics of different kinds of political activities, linking them to a lack of trustworthiness, frustration, disappointment, and reflection. Most of these factors can be related to rational actor approaches of political behavior; hence, the findings are discussed with respect to rational choice theory and with regard to the importance of sociopolitical and emotional integration of Turkish migrants. The role of political trust is discussed as that of a potential moderating force in the facilitation of political participation.


PUBLIC INTEREST STATEMENT
The discrimination of immigrants and social inequities can pose serious threats to social peace and the health of a nation, and are therefore pressing challenges for many Western societies. The Turkish community forms the largest migrant group in Germany, whose social and political discrimination has been of ongoing public interest. Such discrimination can be challenged through the co-shaping of public policies. This requires the members of the Turkish community to get politically active, because others may be unaware of or uninterested in overcoming existing inequalities. However, educational and language barriers can prevent minorities from co-shaping the public. Highly educated members of the Turkish minority may therefore be most influential in promoting the interests of the Turkish community. This research interviewed university students of Turkish descent to identify obstacles to their active participation in the co-shaping of public policies, which can help us to address existing inequities and to encourage them to help co-shaping our shared communities.

Introduction
For modern democracies, the participation of citizens in the political realm is an important characteristic of legitimate political decision-making. As many modern democracies face the challenge of the social and political integration of immigrants, the participation of immigrants and their descendants is of particular significance. This requires us to know how we can effectively stimulate political participation and, as a consequence, the political integration of immigrants. This paper employs a qualitative approach to explore the motives and facilitators of, and obstacles to, unconventional and conventional political activities among Turkish migrants 1 in Germany. The focus is on Turkish migrants as these represent the largest group of immigrants in Germany (17.4% of all immigrants; Statistisches Bundesamt (Destatis), 2015, p. 7). Hence, their integration into political life is a significant challenge, in particular as Germany for a long time did not define itself as a country of immigrants (Bommes, 2010;Joppke, 1999;Santel, 2006). Although the present study does not give a representative account of that population, this study aims to provide insights into conditions and intervening variables that may be of relevance in the facilitation of political action in addition to those conditions and constraints that were examined in a standardized survey, which focused on cognitive political mobilization as well as collective identity.

Political participation
"Political participation" consists of every voluntary activity a person performs in their role as a citizen to influence authoritative regulations and decisions of the political system or political processes (van Deth, 2014, p. 351f.). "Political action" and "political behavior" are often used as synonyms, and intentions to participate in politics are frequently understood as political participation, too. Based on existing literature (Barnes et al., 1979;Steinbrecher, 2009), we may distinguish four kinds of political participation: Electoral participation-voting-does not require intense effort, nor is it bound by a strong commitment. Voting is only constrained by formal regulations (e.g. citizenship). Conventional political activities are traditional, party-related forms of participation, and are sometimes called "party politics." These activities are often institutionalized and require a certain degree of commitment as well as a higher investment of time by the activists (e.g. supporting an election campaign). Unconventional activities refer to a broad range of less time-intensive or committed political activities outside the realm of political parties. In fact, these activities have a long tradition in many Western countries and are nowadays often referred to as "protest activities" (e.g. signing a petition, distributing leaflets). Finally, non-normative, illegal political activities are located outside the legal framework (e.g. attending a violent demonstration).

Key concepts of the present study
The present study is a follow-up to a panel survey which focused on the distinctive role of political knowledge, interest in politics, and political self-efficacy as precursors of different kinds of political participation (Reichert, 2013). These core concepts were defined with reference to common conceptualizations in the field. That is, political knowledge was defined "as the range of factual information about politics that is stored in long-term memory" (Delli Carpini & Keeter, 1996, p. 10). Political interest was understood as the "degree to which politics arouses a citizen's curiosity" (van Deth, 1990, p. 278). This includes citizens' awareness of and attention to political issues (Zaller, 1992). Political self-efficacy was defined as the feeling of being politically powerful on one's own. That is, political self-efficacy was conceptualized as an individual's perception of being capable to understand political facts and processes, and to take political influence (Almond & Verba, 1963;Balch, 1974;Campbell, Gurin, & Miller, 1954). This conceptualization aligns with psychological concepts of internal efficacy (Bandura, 1977) and the concept of perceived behavioral control in the Theory of Planned Behavior (Ajzen, 2012), and it is a key variable that has been used in empirical tests of rational choice models (e.g. Lüdemann, 2001) (cf. 2.3 Political Participation: A Rational Decision?). Finally, political trust turned out as an important concept. It was loosely defined as the belief that politicians and political institutions would not treat people unjustly (Newton, 2007).
Rational choice approaches often utilize some of the variables identified and introduced abovevalues, norms, attitudes, motivations-to explain political behavior (Steinbrecher, 2009, p. 67). Rational choice models generally assume that "individuals engage in political activity to pursue particular goals, and they decide to participate when the benefits of such activity outweigh the costs" (Leighley, 1995, p. 192). Essentially, this thesis comprises three basic assumptions (Opp, 2009, p. 2ff.): (1) Individuals engage in political activity to pursue particular goals and to satisfy their own needs (preference hypothesis).
(2) The probability that their goals are achieved and/or that their needs are satisfied depends not only on themselves, but it is constrained by contextual factors, hence the likelihood that an individual will engage in a certain political activity also depends on those external factors (constraint hypothesis).
(3) Rational actors will not act if the costs of action outweigh the returns. More specifically, rational actors try to satisfy their needs to the highest possible extent and at the same time try to keep their own costs of acting at a minimum (maximization hypothesis).
However, decisions that result from participation in the political arena are almost always the consequence of the political acts of many individuals. That is, making politics is a form of collective action to define common rules (e.g. making laws) and to achieve the common good, and thus political participation is often irrational if it is seen from such an economic rational actor perspective.
Indeed, unless the number of individuals in a group is quite small, or unless there is coercion or some other special device to make individuals act in their common interest, rational, selfinterested individuals will not act to achieve their common or group interests. In other words, even if all of the individuals in a large group are rational and self-interested, and would gain if, as a group, they acted to achieve their common interest or objective, they will still not voluntarily act to achieve that common or group interest. (Olson, 1965, p. 2) In quite many situations, one individual cannot make a huge difference and the effect of an individual's act remains extremely small (e.g. one vote out of several million votes). In addition, if all group members benefit from a specific political decision, it is more rational to let others engage in the decision-making process: in fact, free riders act more rationally than those who actively contribute to the making of politics. Consequently, nobody would engage in political action if everyone acted entirely rational; yet individuals do participate in politics. This "paradox of participation" has struggled researchers for decades, but most research in the rational actor framework has examined voter turnout, which (at least in consolidated democracies) is a very low-cost activity and not wellsuited for examination in this framework (Leighley, 1995).
"Incentives" have been proposed as the solution of the paradox of participation, such as psychological selective incentives, policy dissatisfaction, feelings of personal influence and group success (i.e. internal, external and group efficacy), and a sense of duty to participate (Leighley, 1995). Arguably, diffuse trust in political institutions can be one of these incentives and its effect may depend on the political goals to be achieved. A recent study showed that a lack of such trust is positively associated with non-institutionalized political participation, whereas diffuse trust in political institutions is not only positively associated with electoral and institutionalized participation, but the latter effects cumulate with higher levels of political awareness (Hooghe & Marien, 2013). Leighley (1995) also emphasized the role of group membership and that participation might be expressive rather than instrumental (e.g. to express dissatisfaction about policies), and that expressing political opinions could be a motivating benefit in itself.
One rational approach that combines some of those suggested variables to explain participation is the Theory of Planned Behavior (Ajzen, 2012;Eckstein, Jugert, Noack, Born, & Şener, 2015;Jugert, Eckstein, Noack, Kuhn, & Benbow, 2013). It assumes that human action is guided by three kinds of considerations: beliefs about the likely outcomes of the behavior and the evaluations of these outcomes (behavioral beliefs), beliefs about the normative expectations and actions of important referents and motivation to comply with these referents (normative beliefs), and beliefs about the presence of factors that may facilitate or impede performance of the behavior and the perceived power of these factors (control beliefs). (Ajzen, 2012, p. 18) Recent studies have expanded that framework in the political realm to account for political knowledge as an additional condition, or precursor of individuals' beliefs of internal control (i.e. internal political self-efficacy, see above) (Reichert, 2016). Yet despite supporting evidence for these approaches, debate remains about the applicability of rational actor models. One question is about the behaviors that are supposed to be rational, as partially planned or unplanned behavior might be automatically activated (Bargh, 1997;Fazio, 1990;Kahneman, 2003;Strack & Deutsch, 2004). Also, as was mentioned earlier, it is challenging to solve the paradox of political action using such a framework. In addition, the measures used in (quantitative) studies on rational actor theory have been criticized (Leighley, 1995), because the incentives, benefits, and costs that are associated with political participation may vary largely between individuals.
The reflective-impulsive model (RIM), which was proposed by Strack and Deutsch (2004), might provide a solution to this paradox: These researchers have developed a dual process model, according to which two pathways lead to behavior. "The reflective system generates behavioral decisions that are based on knowledge about facts and values, whereas the impulsive system elicits behavior through associative links and motivational orientations." (Strack & Deutsch, 2004, p. 220) Moreover, the RIM assumes that both systems operate simultaneously: On the one hand, affective processes are always involved in the initiation of behavior and are dominant in spontaneous activities. On the other hand, reflective processes are only activated in addition to impulsive processes when reasoned action is required. Thus, the reflective system has precursors that are considered in rational actor models, but it is only active if the "stakes are high," i.e. when the consequences are not immediately clear or when the costs of an activity are quite high. However, this system is also more prone to disruption than the impulsive system, because the former depends on the available capacities of our working memory. On the other hand, the impulsive path depends on particular habits and the satisfaction of basic needs as well as motivational orientations which ease information processing and behavior.
The RIM has not been applied to political theory, and it is questionable whether political participation could be entirely impulsive. Still, it would be helpful to learn whether differential pathways that might represent highly reflective vs. less reflective processes can be explored for political participation. More specifically, questions that need to be answered are: What are the goals and the expected benefits of those who decide to participate in politics? Do the expected costs and benefits of, as well as the reasons for participation in party-political activities differ from those that are linked to participation in protest activities? Can we identify, in fact, two different pathways to political action, of which one is more rational than the other in that costs and benefits are weighed against each other more explicitly (completely unreasoned political action might be less common though)? Finally, can we link these pathways to differences in the types of political activity as well as to differences in political interest and political self-efficacy?

The present study
The present study was part of a larger project which examined the role of collective identities among highly educated Turkish migrants (Simon, Reichert, & Grabow, 2013;Simon, Reichert, Schaefer, Bachmann, & Renger, 2015). For migrants, the costs of participation are usually higher than for nonmigrants, which makes Turkish migrants a particularly interesting population. Given lower levels of participation among Turkish migrants, we had chosen university students as our population, due to the well-known positive correlation between education and political participation, and because those migrants were more likely to become spokespersons of their own minority group and, thus, to initiate change. Results from the standardized survey suggest that political interest directly predicts unconventional political participation, but it needs to translate into the explicit intention to participate with respect to electoral and conventional political participation. Political interest appeared as the more proximal predictor of political action compared with political self-efficacy. However, the latter was almost as important as political interest in the prediction of conventional political participation (Reichert, 2013). Moreover, identification with both their minority in-group and the German majority appeared to promote normative political participation among Turkish migrants, a group that reported a significant level of collective grievances (i.e. they perceived being treated unfairly because of their Turkish origin, or being discriminated against as members of the Turkish community) (Simon et al., 2015).
It seems plausible to assume that other conditions and additional intervening variables may be relevant in the prediction of political behavior, however. The present study presents results from a small number of qualitative interviews with participants of that panel survey. These interviews were analyzed with respect to additional conditions that may or may not relate to individuals' interest in politics, political self-efficacy, and their knowledge and skills in the political realm. The aim of this analysis was to gain insights that are not constrained by a standardized questionnaire, and to gather qualitative, open responses on study participants' decisions to participate, or not to participate, in politics, and to link these responses to their political interest, self-efficacy, and costs and benefits of participation. Therefore, this research focuses on individual (and not collective) participation, and it tries to explore further rational and non-rational "conditions" of political participation (for more details, see Section 4). An important feature is its focus on Turkish migrants, who represent the largest group of immigrants in Germany (Statistisches Bundesamt (Destatis), 2015, p. 7), as they may experience higher hurdles to participation in politics than native Germans.

The situation of Turkish migrants in Germany
With almost three million Turkish migrants, this group represents the majority among migrants in Germany (Statistisches Bundesamt (Destatis), 2015). The first Turks arrived in Germany in the 1950s, and due to political regulations, the immigration of Turks increased in the 1960s, and Germany became the home for many Turkish families in the 1970s (Böcker & Groenendijk, 2006;Joppke, 1999;Meier-Braun, 2007;Şen & Faruk, 2002;Thränhardt, 1999Thränhardt, , 2004. Although Germany abandoned the ius sanguinis in 2000 in favor of the ius solis, so that newborns can acquire the citizenship of their country of their parents' citizenship, Germany still does not fully accept dual citizenship. Hence, the offspring of Turkish immigrants have to choose between the German and the Turkish citizenship when they are adults. Those who choose to keep their Turkish citizenship will be excluded from electoral participation. Many Turkish migrants have been living in Germany for more than two decades (Babka von Gostomski, 2008). Although most of them are Muslims (Haug, Müssig, & Stichs, 2009), Turkish immigrants are by no means a homogenous group-and this not merely because Muslims are a diverse group or because some Turks may be Muslims in name only (Halm, 2013). Yet despite the vibrant social life and a vast number of club activities in the Turkish communities, they are actually disadvantaged and perceive being deprived: compared to children without migration background, Turkish migrants perform much worse in skills assessments even after controlling for other influences (Diehl & Fick, 2016). No other migrant group counts a relatively higher number of members without a secondary school qualification and with fewer members who have received a university entrance certificate (see also Avcı, 2006;Woellert, Kröhnert, Sippel, & Klingholz, 2009). Only among Turkish migrants is it possible to explain their lower position in the German labor market not almost exclusively by differences in education (Granato & Kalter, 2001). Youths from (Turkish) immigrant families are significantly disadvantaged with respect to vocational training and job search compared to their German peers-even if they have equivalent school leaving degrees (Goldberg, Mourinho, & Kulke, 1995;Granato, 2003;Hunkler, 2016;Kaas & Manger, 2010). Yet Turkish youth are more likely than Germans to enter tertiary education, which could be due to their unfamiliarity with the German system of vocational training (Kristen, Reimer, & Kogan, 2008).
Turkish migrants also experience rejection in everyday life, and politics and fellow citizens refuse to treat them equally (Böltken, 2000;Sachverständigenrat Deutscher Stiftungen für Integration und Migration, 2014;Şener, 2007;Wasmer & Koch, 2000), in particular if they are Muslims (Brettfeld & Wetzels, 2008). As Germans seem to have a more critical perspective on Turkish immigration compared to immigration of other populations (Yavçan, 2013), 2 Turkish migrants feel more disadvantaged and pessimistic compared to other migrant groups in Germany (Fassmann & İçduygu, 2013;Sachverständigenrat Deutscher Stiftungen für Integration und Migration, 2014). Therefore, the basic premise of active participation in politics clearly exists; which is: to complain about social deficiencies and to demand political change, and therefore to get involved in the political process. Social psychological theory and research on intergroup behavior indeed suggests that the perception of illegitimate injustice of one's own in-groups, such as the ethnic collective, compared to other groups within the same society can trigger grievances that are shared among in-group members (e.g. Simon & Klandermans, 2001). If in-group members attribute the causes for their perceived injustice to other groups within the wider society, and if these out-groups do not accept responsibility for that injustice, then in-group members may become politically active in order to obtain a more just situation (Simon & Klandermans, 2001). Clearly, these conditions are present in the case of Turkish migrants in Germany.

Political participation of migrants in Germany
The few existing studies on migrant political participation in Germany indicate that legally existing opportunities for political participation correspond with an ineffective advocacy of migrants (Diehl, 2002;Ergi, 2000;Wiedemann, 2006). Accordingly, migrants usually lobby for political, cultural, and social rights (Rucht & Heitmeyer, 2008), and Turkish migrant organizations address both the situation of Turks in Germany and political developments in Turkey (Sezgin, 2010). Although there exist many opportunities to individual political participation even without the possession of German citizenship (Diehl, 2002), citizenship is important for the alignment of political participation of migrants in Germany (de Wit & Koopmans, 2005;Koopmans & Statham, 1999).
Related to a lack of legal requirements is a significantly lower level of political interest among migrants, which is higher among Turkish migrants though, compared with other migrant groups (Diehl & Blohm, 2001;Diehl & Urbahn, 1998;Wächter, 2005;Wüst, 2003). However, the patterns of political participation among young migrants differ only little from native Germans if it is accounted for the legal framework-participation rates are merely at a lower level, and these differences disappear almost completely when controlling for education (Gaiser & de Rijke, 2006;Gille, Krüger, & de Rijke, 2000;Heß-Meining, 2000;Wiedemann, 2006). Yet people of Turkish origin are less connected with German society, and previous research found that they report the lowest willingness to participate in politics as well as the lowest propensity to participate in politics (Glatzer, 2004;Wiedemann, 2006). This also applies to the electoral participation of naturalized Turks who have the lowest turnout rates compared with almost all other naturalized citizens (Wüst, 2007). Probably due to their lower levels of education, Turkish migrants also prefer to participate in Turkish associations rather than German associations. However, more recent works found indication that young immigrants may be more active than non-immigrant youth (Eckstein et al., 2015), and particularly Turkish migrants might be more active than Russian migrants, and potentially also compared to non-migrant Germans (Jugert et al., 2013;Simon et al., 2015).

Methodology
The present, qualitative study was conducted as a follow-up to a longitudinal survey study which focused on the cognitive political mobilization of highly educated Turkish migrants (Reichert, 2013). Given lower levels of participation among Turkish migrants, we had chosen university students as our population, due to the well-known positive correlation between education and political participation, and because those people were more likely to become spokespersons of their own minority group and, thus, to initiate change. This was an important choice, because well-educated migrants are more likely to have the skills to influence and convince members of their own minority group. In addition, less well-educated migrants would be less likely to engage with the German society and, hence, to address the perceived injustice of their in-group. At the same time, there might be a higher risk for highly educated migrants to feel aggrieved if they sense discrimination despite their high level of education, which might turn their grievances into unproductive forms of participation. 3 On the other hand, this selection also means that the findings may disregard to some extent the role of Turkish associations, which are commonly used by Turkish migrants as a means of addressing their political claims. The present explorative study aims to give our respondents a voice and to enrich the findings from the quantitative study with qualitative insights from the respondents. The primary goal of the qualitative interview study was to gain deeper insights on the factors that motivate or hinder political participation among our participants, for the purpose of a more thorough interpretation of our quantitative findings. Thereby, this analysis intends to yield additional insights into constraints and further variables that may or may not be relevant to facilitate different kinds of political participation.
Semi-structured interviews were conducted in order to broaden our survey results by inviting some of the survey respondents to speak on their own, without the standardized format of a survey questionnaire. For this purpose, individuals who had previously participated in the online panel survey were selected for interview participation using a criteria-based case selection based on our quantitative data (Reichert, 2013). This specific sampling plan leads to a heterogeneous sample, which has the power to identify typical patterns (Merkens, 2005;Schreier, 2010). More specifically, the larger project on identity, within which the current study was conducted, aimed to conduct interviews with Table 1 Characteristics of the interviewees a Not born in Germany (immigrated to Germany in childhood or early childhood). b "Dual" means the possession of both the German and the Turkish citizenship. c Range is from 0 = little interest/efficacy to 4 = high interest/efficacy. d Self-reported political participation since the first survey. V = Voting (0 = no; 1 = yes), C = Conventional participation (0 = no activity; 1 = at least one of the following: contacted a politician, actively supported a political party campaign, membership in a political party, or participation in a political committee or working group), U = Unconventional participation (sum index of the following activities: signed a petition, distributed political leaflets, consumer boycott, participated in a legal demonstration, and participated in a citizens' initiative), N = non-normative political action (0 = no activity; 1 = at least one of the following: wrote a political slogan on a public wall, participated in an illegal demonstration, blocked a road for political reasons, occupied houses or offices, participated in a violent demonstration, damaged other people's property). e According to self-report, there was no opportunity/legal right to participate in elections between the first and the last survey. an equal number of (1) respondents who self-identified as both Turkish and German, and those who did not identify as both;

ID Gender
(2) respondents who thought there was vs. was not an incompatibility between being both; and (3) male and female respondents. In a second step, for this present study and after the interviews had been conducted, interviewees were selected according to their participation in conventional and unconventional political activities, as they had reported it themselves in the online surveys before the conduct of the qualitative interviews. These patterns were then linked to the key variables of the quantitative analysis, that is, respondents' self-reported interest in politics and their political self-efficacy. This procedure ensured that these interviewees covered the entire range from low to high political interest and self-efficacy, and it enables the comparison of the quantitative and qualitative data. Table 1 shows the composition of the interviewees.
Interviewees were assured anonymity and written consent was obtained from all interviewees at the end of the interview, allowing us to use the data obtained during the interview for our research purposes. No incentive for participation was offered to motivate respondents to participate in the interview study. However, those respondents who participated in the interview received either a book voucher or cash (each worth ten Euro) after the interview. The present study analyzed 10 interviews that were collected by either a female or a male interviewer in German language. These interviews were 92 min long on average (ranging from 74 to 128 min), and all interviews were transcribed and analyzed after data collection.
The interviews always started with a standardized introduction before the interviewees were asked to talk about several issues. Here interviewees were informed that there were no "correct" or "incorrect" answers to our questions, and that they could take all the time they needed to talk about everything they would associate with the questions, or consider relevant for answering our questions. The interview guideline itself was structured around three key questions: (1) "What does it mean for you that you have Turkish roots on the one hand, and live in Germany on the other?" (2) "How do you integrate the different aspects of your 'self'?" (3) "What do you think of the fact that people who live in Germany and have Turkish roots are politically engaged in Germany?" The guideline also contained additional questions in case that narrative flow did not come about. Those most relevant for the present analysis were listed under the third key question and covered the topics of competence, efficacy, engagement, motivation and obstacles, duty, goals, and significance. Examples are: "Some people think it is much too complicated to understand what is going on in politics. What do you think?"; "What motivates you to engage in the public?"; or "What are the goals of your public engagement?" Finally, continuative keywords were only used when no flow of conversation came about or when it had stalled. Some examples of these keywords are: "importance of politics for you;" "effects of participation;" "influence;" "risks;" etc.
The following analysis focuses on subjective complexity, the interviewees' understanding of and interest in politics, and statements on political behavior. These narrations were meant to enable us to draw conclusions about the interviewees' political activities, and the specific nature of these political activities, as well as particular motives or action conditions underlying their political participation that our standardized survey may have missed. That is, the goal of this analysis was to examine the three "types" of participants (no participation-only unconventional participation-conventional and unconventional participation) with respect to their qualitative understandings of political participation, their motivations to or not to participate, and how the qualitative reports "match" the quantitative self-reports on interviewees' interest in politics and political self-efficacy. The expectation was that the resulting narratives might provide further insights into the specific kinds of and motivations behind actual political participation, or subjective conditions of participation. It seemed reasonable that this would help to expand our understanding of the conditions and constraints of political participation.
The steps that guided the analysis of the qualitative data were driven by content analysis as suggested by Mayring (2003), and case-based data exploration similar to qualitative evaluation as described in Kuckartz (2007). Each case was explored individually in the first place, before a more systematic analysis of the interviews was conducted based on five categories: motives of political participation; conventional political participation; unconventional political participation; political interest; and political efficacy. All texts were analyzed with the aim to draw generalizations based on "typical" statements of the interviewees, to explore correlations between the mentioned categories, and to identify statements that are potentially inconsistent within the interviewees or in relation to the quantitative data. The following section summarizes the qualitative interview findings.

Politically inactive interviewee
The guided conversation with the only person who had participated in neither conventional nor unconventional political activities (ID F) was not very informative, and the interviewee reported a somewhat contradictory attitude. The interviewee expresses "unfortunately" to have a basic interest in politics, which essentially is an interest in everyday politics ("politically charged topics", "panel discussions" on television). However, the respondent also claims to be in a position to critically question current political events: she deliberately does not call herself "politically informed," but rather means to have a "political awareness." She clarifies this by stating that she no longer can inform herself about politics (as this is too time-consuming, such as reading a daily newspaper), but is only aware of the political events without having a deeper knowledge of the details. At the same time, this person shows little appreciation of conventional, party-political activities: "I get too irritated. […] Particularly about these party-political conflicts; and this party-political demeanor is beastly unnerving." This interviewee explains her political inactivity by the absence of political goals. In her opinion, the prerequisite for political action is to have a specific objective, as well as the individual need to contribute to social change and "to be a help" to other people or society. However, none of these two prerequisites were met in her case, according to herself; instead, the interviewee claimed to lack of interests and objectives, "because I myself do not know what is stopping me now. Yes, maybe because I cannot accumulate for myself any interests that are so important to me that I would now bother to be politically active." She can still imagine being politically active, though, if a particular political topic would become "so politically charged that it is impossible for me not to do it" or if she was not required to develop her own agenda, but if she was told specific political objectives-such as in the form of a party programme, for example. Also, this interviewee does not believe in political violence: "I would not have the balls for that." In summary, this person watches political events suspiciously though she endorses political activity in general. However, although she reports an interest in politics, this does not lead to political activity-in her opinion, her interest does not yield a sufficient level of cognitive involvement in the sense of a productive reflection on how the alleged weaknesses of the political process can be eliminated by her own contribution. Her interest rather appears to be a manifestation of her distrust in the polity. This person simply seems to be uninterested in making an active contribution to political life herself. Instead she is satisfied with the current division of responsibilities and her own role as a "critic of political events," for she leaves political work to other people.

Interviewees who participated only in unconventional political activities
These interviews (IDs A, C, E, H, and I) suggest that the respondents participate in unconventional ways because they do not trust conventional structures which they think are nontransparent and inefficient. "Actually, I was always very interested and then I realized, 'Oh, they get the money and do anything to boost the media and the like.' I just realized that this is not my world. I do not know any true, reasonable politician who really wants the best for their country." (H) Thus, politics is considered a "dirty deal" (H). One of the interviewees explains his distrust in political parties, "because I have such an attitude that I do not trust politics so much and so on, and I would rather prefer to have direct access and not to rely on someone else." (A) Moreover, these interviewees do not feel sufficiently politically competent and efficacious themselves. Hence, these students might either indeed be not as competent as necessary for participation in conventional politics, or they do not put enough effort in making politics transparent to themselves. One of the interviewees justifies this by her "sciolism," and she further notes: "I think I have other interests [than politics and political participation]. I do not know if it has ever interested me or I do not know. I guess I have just left it to others.
[…] I mean if you want to be politically active, you should have a firm political stance." (C) Conditions for political participation that are mentioned by this interviewee are time, political interest, and perseverance, owing to the perception that individuals and small groups have no influence: politics "is decided elsewhere. As a small group? I do not know." (C) Most of the respondents who participated in unconventional political behavior, but not in conventional activities, really want to reap the fruits of their labor directly. Thus, they prefer possibilities of direct democratic participation-activities that "can really have a direct impact on society" (A) and activities in non-governmental organizations (A, I). These people also talk about the realization of "projects" instead of persistent, "ideologically" coined party politics (A), and they want to recognize the results of their political activities immediately. They aim to "support the community" (E), "give something back to society" (E), "show solidarity with other people" (I) and promote a just society. Hence, it comes without surprise that these interviewees also suggest that social awareness and the individual's need to bring about social change are necessities. Political action is also considered to be a "hobby" (E) and should be anchored in the living environment, everyday life and day-to-day needs. For interviewee I, for example, political action "is a very personal matter," because she feels excluded and "does not feel to be free" in her own daily life. It is for this reason that she prioritizes the fight for freedom and justice in her own engagement. Though at the same time, she reports that she deliberately decided to participate in an international nongovernmental organization, because people in Germany would have a good life. Yet this emphasizes her concern about social justice, and her interest in freedom and justice further explains why her activities are at the international level.
Those interviewees' unconventional political participation whose political interest is relatively low are likely to be encouraged by fellow activists. "I guess I have just left it to others. […] If someone would come and say, 'Well we do this and that and do you want to work with us, help us?' I think that would be the way where I would say yes, gladly, so what can I do? Then I would be helpful of course." (C) If an opportunity to express one's own political concerns opens up and one is not alone, you just participate, in particular spontaneously and on short notice, by going on a political demonstration, for example. Personal interests are supposed to reflect the activities of those interviewees, and short-term, well-defined commitments are often preferred. These, in turn, are characteristic of unconventional political activities. The concise statement of interviewee H puts it succinctly: "If you are interested, then go ahead." This suggests a direct link between political interest and issue-related political action (see also earlier, interviewee C).
Particularly interesting and somewhat different is interviewee E, who clearly prefers conventional ways of participation, though not yet being actively engaged in these. In his opinion, political decisions can only be influenced by elections and direct contact: "You talk until you are blue in the face, so to speak. I always think that people are demonstrating against a wall, because politics can only be influenced by elections." His level of political self-efficacy is relatively above his interest in politics, and he does not think that politics is complicated: "One does not have to have studied politics to become a politician." Moreover, he has already plans to further political participation using conventional means in the future.
Overall, it is amazing that those migrants who had participated in many unconventional activities simultaneously tend to report lower levels of political self-efficacy, and sometimes also lower levels of political interest, compared with those interviewees who had performed only one of the unconventional activities that we tapped in our survey. Whether this is because the interviewees understand "political interest" as an interest in traditional or conventional politics could not be clarified.

Interviewees who participated in conventional and unconventional political activities
The interviews with those who additionally participated in conventional political activities (IDs B, D, G, and J) suggest that conventional political action requires carefully planned decision-making and a consideration of costs and benefits. All four interviewees point out that political action requires organization, planning, and a long-term commitment-things that other interviewees had also mentioned. Interviewee B: "An individual cannot achieve that much. If you manage to organize yourself and collect a few votes, be it a signatory campaign or a demonstration, you can achieve a lot more." Conventional, particularly party-political activities are described as tedious; our interviewees emphasize that persistence is actually a personal requirement. They suggest that one needs to refrain from the desire to recognize immediate success of political actions, since doing successful politics is quite time-consuming and proceeds in small steps. Interviewees say that one has to be "deathly cold" (J) and should have a "thick skin" (G). Acting in a conventional way is considered as a kind of "self-flagellation" (J). J further clarifies that she has "given up" and is now "depoliticized," because people would always talk but not act, and even in those groups that she worked in, hierarchies persisted with no societal change in sight. According to the interviews, one clearly needs to be able to emotional distancing, and has to accept social exclusion. For example, interviewee G reports a clear example when she describes that her circle of acquaintances shows a lack of understanding and appreciation of her participation in the German Social Democratic Party. This is because the party is seen as the wrong place to advocate for Turkish people, as it dithered and eventually did not exclude Thilo Sarrazin, who wrote the book "Germany is abolishing itself." In his book, he argued that Muslim immigrants-and particularly those from Turkey-are hard to integrate into society, and that this is due to those immigrants' attitudes. This interviewee characterizes herself as a "warrior by nature," because she has not given up her conventional participation despite her enormous frustration about the "cliquing" in her own party (as well as "lobbying," as B puts it) and about the fact that politics "always operates on a very short-term basis and always just survive the next election." Another thing that individuals may need to engage in conventional action, according to these interviews, is to overestimate their own contribution. This might result in frustration though. Like previous interviewees, social change is an important objective, but most of the interviewees who report conventional political participation confirm that it is a conscious decision to participate that accounts for the costs of conventional participation. "I cannot save the whole world, but at least I have this ideal to change something and improve something." (G) Finally, the interviewees also talk about political competence in a broad sense, and objectively incorrect knowledge appears to be an obstacle to conventional political action for one of the interviewees: "I would also like to get involved in politics, but I lack the German nationality.
[…] According to German law, I believe that one must be a German citizen to get active in a party and unfortunately, God forbids me to move up to the Federal Chancellor. I might have had that ambition." (B) In fact, non-German citizens can legally participate in a political party. However, this interviewee is not just unaware of this fact, but he possesses incorrect knowledge, which seems to prevent him from more conventional participation. Unsurprisingly, a lack of participation is attributed to a lack of political knowledge and political skills, and this does not only concern the knowledge of political rights and opportunities to participate, but linguistic skills such as the knowledge of technical terms, completing forms or writing political proposals are discussed as well. The command of the German language is also meant to be important, in particular when it comes to conventional political participation. An instance reported by interviewee G makes this very clear: During a meeting of district delegates, her friend-who had no university degree, as emphasized by G-asked her about the meaning of the word "pragmatic," which she could not make sense of. All these aspects show the significance of being conscious of your activity, your aims and the time frame until you will succeed, which skills you need and what you might have to spare meanwhile. This actually corresponds to the high levels of political self-efficacy among these interviewees.
Unconventional political activities were carried out by all of these interviewees, too, which also matches their high political interest. As noted earlier, the aims of participation, and particularly of unconventional participation, are similar to those mentioned by the interviewees who only participated in unconventional activities, are related to social change and justice. "When I see something unjust, I try to get it right as quickly as possible." (D) This interviewee strives for a community that is "worth living in:" "So if you can change something in their lives, this is politics for me.
[…] For me, politics is to change the active life here in Germany-with the rights one has got." Similar is interviewee G: "I cannot save the whole world, but at least I have this ideal to change something and improve something." These interviewees believe that change can be achieved by multiple means-conventional and unconventional.
One exception is interviewee B though, who has the lowest levels of political interest and efficacy among the 10 interviewees, which might correlate with his distrust in German politics. It should also be noted that immigrants and their descendants are often excluded from the political process so that they do not know their options to participate, and interviewee B is apparently unaware of his right to participate in political parties (see above). Thus political knowledge is important.

Discussion
The explorative interviews analyzed here followed a survey study which examined political interest and political self-efficacy as precursors of different kinds of political action. This qualitative study focused on a well-educated subpopulation of Turkish migrants in Germany, and the results may not be generalizable across the entire population of Turkish migrants. The presented interviews draw attention to the characteristics of the different kinds of political action and they shed light on further processes and conditions of political participation beyond interest and efficacy. In particular, among the interviewees political inactivity was associated with distrust in (conventional) politics. Instead of active participation, primarily criticism was practiced and a rather individualistic strategy was pursued. In this study, political distrust and dissatisfaction with politics also did not necessarily pose an incentive to get politically active, although dissatisfaction was consistently associated with the wish for a better society.
Resignation and a lack of credibility of political parties and politicians were linked to the lack of conventional political activity, but as indicated, unconventional participation was frequently reported by the interviewees who indicated that they were unsatisfied with conventional politics. A lack of confidence in their political self-efficacy and skills, which are often considered to be prerequisites for reasoned, rational, reflective behavior, also prevented some interviewees from engaging in conventional political participation. On the other hand, a basic level of distrust and possibly a moderate interest in political issues were associated with increased unconventional activity among the Turkish migrant interviewees. Other research has shown that despite the often positive ramifications of political trust, more trust does not necessarily implicate more participation in any political activity (Torney-Purta, Barber, & Richardson, 2004;. Recent research has provided a more nuanced answer to the role of political trust and suggested that more institutional trust might be positively correlated with more electoral and conventional political participation, whereas less political trust might trigger unconventional participation (Hooghe & Marien, 2013). That study also found a cumulative effect of political trust and political self-efficacy with respect to electoral and conventional participation. The findings of the present study align well with that (see below). In terms of rational choice theory, electoral and conventional political activities are probably instrumental forms of participation that serve a specific goal. On the other hand, unconventional or "protest" participation might be regarded as more expressive in nature, so that dissatisfaction and distrust in political institutions might primarily serve the purpose of expressing one's dissatisfaction, which then would provide instant gratification. Major societal changes were indeed rarely the primary purpose of respondents' unconventional acts, rather, they wanted to see immediate results and help their local and immediate environment(s).
Also, we could speculate that political trust might influence outcome beliefs and the perceived power of impeding factors to the successful execution of a particular behavior; hence, participation within the traditional system may be believed to be costlier, and combined with limited political selfefficacy, this type of political participation might be less common. If trust took the role of a switching point that facilitated conventional participation, whereas distrust triggered unconventional participation, a pronounced withdrawal from conventional politics could be the consequence: Turkish migrants are disadvantaged in German society, which arguably could promote distrust in the established institutions. In return, this could consolidate their disengagement with conventional politics, a means of participation that is more likely to produce authoritative and binding decisions and regulations. Consequently, that development might have the potential to promote a democratic deficit, if our study findings would be generalizable across the whole population of Turkish migrants.
An important task for future research will be the examination of the relationships between political trust, political self-efficacy, and political participation among different groups in society, of which some might be more disadvantaged than others, to learn whether the mentioned mechanism is conditional of social disadvantage of a (minority) in-group. For example, as Muslims-who are "dominated" by the Turkish population in Germany (Halm, 2013;Haug et al., 2009)-are more politically trusting than non-Muslims (Doerschler & Irving Jackson, 2011), it is worth examining how this trust may moderate the intensity of political participation between Turkish migrants and Germans without any immigration background, as well as between different migrant groups. An additional question that needs to be answered is about the role of religiousness in general, and whether the strength of religious belief is a moderator that works above and beyond the pure membership in the Muslim community.
In addition, in a time when "fake news" has become a serious threat to society, it is also essential to connect trust, efficacy, and actual knowledge about politics, and to study whether correct and incorrect knowledge of politics (and the degree of certainty about the correctness of the knowledge one beliefs to have) make a difference in how trust and efficacy relate to different types of political participation. Although prior research has suggested that political knowledge has indirect effects on political participation (Reichert, 2016), it would be useful to study the interactions between the three mentioned variables. Arguably, respondents' confidence in the accuracy of their political knowledge might affect the level of their trust or distrust in politics, which then should affect the strength of the relationship between political trust and the different types of participation.
Furthermore, the interviewees found it important consciously and deliberately to opt for participation in conventional political activities: Those who were, are, or plan to get active in conventional ways reported about well-planned decisions regarding their political participation, and that their choices were made despite the acknowledgement of significant "costs" (emotional distancing, social exclusion). These respondents weighed the costs of participation, although some also mentioned that the cost is quite high and that their participation does not provide immediate benefits. Individuals may (initially) need to overestimate their own contribution to politics in order to get involved in conventional activities (Opp, 2009), but this might also increase the risk of political resignation. In any instance, participation in conventional activities always was a highly reflective decision, not primarily an affective "impulse." Although unconventional activities were not always as strongly thought through and might be related primarily to specific political issues, interests and current topics, these more expressive activities also required reflection.
Another inevitable aspect that is linked to reflective decision-making is political competence, which comprises the skills that are required for successful participations (e.g. knowledge, behavioral skills, language skills): In particular linguistic skills to articulate political concerns, the knowledge of certain political concepts and practices within legal structures were understood as necessary conditions of political activity; the lack of communication competence was even considered to be a cause of political violence. Communication scholars also argue that communication competence, understood as the capacity to interpersonal communication and the adequate use and interpretation of news media, is an essential requirement for political activism (Shah, McLeod, & Lee, 2009). Developing such competences may be more demanding and difficult for (Turkish) migrants, however, as they are disadvantaged with respect to their educational achievement. Moreover, their families and peers might prefer to speak Turkish at home or to discuss Turkish politics.
Members of the Turkish community also do not feel accepted in German society (Sachverständigenrat Deutscher Stiftungen für Integration und Migration, 2014). Hence, non-participating peers or acquaintances may find it hard to understand why, and to accept that, a member of their ethnic ingroup would engage in conventional politics and, thus, become part of the political establishment. This would raise the psychological costs of participation. In fact, research suggests that norms of significant others such as family and peers can play a major role in motivating behavior (Ajzen, 2012), and it seems that for young Turkish migrants, significant others' views are indeed relevant for offline participation (Jugert et al., 2013). Being surrounded by people who do not understand or accept such an involvement in the established political processes, resignation or frustration could be likely consequences. Such undesirable attitudes might be even more likely to develop the less the expected outcomes are achieved.
In particular, if participation is not backed by the support (in form of acceptance) by those significant others and at the same time the expected outcomes are not achieved within a given timeframe, it is reasonable to expect that this will affect political self-efficacy or the perceived influence as a collective of Turkish migrants, which rational actor theory has identified as important predictors of intended and actual behavior (Ajzen, 2012; but see Eckstein et al., 2015; who found significant influences of political self-efficacy only among non-immigrant youth-yet their measure of participation comprised no conventional activities; Jugert et al., 2013;Opp, 2009). That is, the incentives that rational actor theory has identified (Leighley, 1995) would not be present in order to balance the costs that are associated with an engagement in political action. Consequently, behavioral beliefs, normative beliefs, and control beliefs would be negative and no participation would be the likely consequence (Ajzen, 2012). Our study provides some evidence that political self-efficacy is in fact a requirement for political participation, in particular regarding conventional activities (Reichert, 2013). The interviewees of the present study furthermore pointed out the significance of other factors that theory has found to be important precursors of (reasoned) action.
Yet it is an undeniable truth that the marginal position of Turkish migrants in Germany requires them to tackle higher hurdles to engage in a conventional manner than, for instance, native Germans without a migration history. Therefore, weighing costs and benefits could be more significant compared to other populations such as native Germans and other migrant groups. Recall, for instance, that some of the obstacles that the interviewees mentioned included, for instance, experiences of exclusion (also from their own social group), but also missing opportunities to participate in general, or an absence of a lobby within their own minority in-group in Germany.
In addition, experiences of exclusion and the absence of a lobby within their own minority ingroup indicate little approval for such participation, that is, a lack of norms of participation among the interviewees' significant others. It is worth mentioning that interviewee B also suggests a need to identify yourself with your representatives in order to develop an interest in politics. Such identification can be regarded as an important aspect of normative beliefs, which provide directions for action (Ajzen, 2012). The absence of the mentioned factors, however, could reduce an individuals' collective efficacy and, eventually, their political self-efficacy. That is, if political participation is not valued among members of the minority in-group, and if there is no lobby that advocates for the goals and needs of that group, then the expectation of group success may be limited-even if all group members share a sense of injustice against their group and even if they attribute the responsibility to another group (which are conditions in the model proposed by Simon & Klandermans, 2001). However, the interrelationships between collective and self-efficacy need more attention in future research: If the group cannot achieve the goals as a group, how would an individual feel (externally and internally) efficacious to do so, in particular if there are no role models that would show how to perform political action successfully? Apparently, significant contextual conditions for participation, especially participation in conventional politics, would not be met, and only little incentives to engage in politics would prevail.
In such a context, respondents would probably have two options to justify their political participation: either they raise the value of the expected outcome of participation, or they value less the opinions of their significant others (cf. Ajzen, 2012). This could cause issues of incompatibility between being a member of the disadvantaged minority in-group and participation within the majority society. Recent research has shown that such an incompatibility can become a liability in that it might increase the readiness for radical political action (Simon et al., 2013). Therefore, future research needs to identify the specific conditions and how these would enable migrants to channel their feelings of disadvantage, unacceptance of participation by other minority in-group members, and the aim to improve society toward the use of legitimate means, rather than to give up or turn to radical forms of participation.
In particular, interviewees mentioned needs such as bringing about social change, an anchoring of political participation in the living environment, and an immediate recognition of action results. These needs are not easily satisfied through conventional means of participation. Those interviewees who utilized unconventional means of participation reported that in their opinion an engagement in these activities demands social awareness, some distrust in "traditional" politics, and a need for immediate change (or "improvement") of social or political conditions. "If you are interested, then go ahead," was the slogan of these interviewees, without bothering about institutionalized means of participation and all the hassle that seems to be associated with them (see also the above discussion on expressive behaviors). We may note here that scholars have argued that younger cohorts may be driving a value change and generally prefer unconventional political activities instead of traditional ways of participation. 4 It would be worthy to examine whether this development is more pronounced among migrant youth. 5

Conclusion
The present study used a selective sample of well-educated university students with a Turkish migrant background. Although it does not allow us to draw general conclusions, it explored and confirmed further conditions and variables that may be required in addition to political interest and self-efficacy to promote political participation of Turkish migrants. This paper did not test any theory, but building on a survey study, we were more prone to identify variables that may be considered in rational actor theories. The results could be easily linked to rational choice models and the Theory of Planned Behavior (Ajzen, 2012), as we found that normative beliefs were important conditions of conventional participation among our study participants-or at least were these constraints taken into consideration by the interviewees. Control beliefs, defined by political self-efficacy and actual political skills (e.g. communication competence) were positively associated with political participation and, again, primarily with conventional activities among the respondents. Motivations such as a just society or positive contributions to society were easily identified by all interviewees. Apparently, striving for a just society, potentially an expressive goal, was the major incentive for political participation, although it was mingled with (a lack of) trust in politics, political self-efficacy and skills.
In conclusion, this study has confirmed the role of several incentives and conditions of participation, as these were identified in rational actor and action theoretic models, among a small and unrepresentative group of highly educated Turkish migrants. However, future research is needed to examine the specific relationships that we tentatively identified.
(1) Future research should study the relationship(s) between political trust, political self-efficacy, (objective and subjective) political knowledge and participation among different groups with different levels of (perceived and actual) disadvantage.
(2) The relationships between collective and self-efficacy deserve more attention in future research and should be examined with samples from different ethnic groups. (3) Finally, future research needs to examine how feelings of disadvantage, the rejection of participation and participants by other minority in-group members, and the wish to improve society work together, and under which conditions members of the minority group will opt for legitimate forms of participation, radical means of participation, or no participation.
As for the Turkish migrants in our study, we may conclude that due consideration of their legitimate interests as a collective is important for their social integration and their readiness to participate in political life. This seems crucial to make the political game also their game, and not just the game of the majority. Otherwise, and if the findings of this study apply to the majority of Turkish migrants in Germany, the liberal democratic society would remain a hollow promise. As European societies are faced with large numbers of immigrants, we have to keep that promise in order not to cultivate disenchanted, distrustful, and frustrated immigrants. Otherwise, democracy may as well abolish itself. If our societies fail to integrate young migrants socially and politically, these young people could endorse authoritarian or totalitarian forms of government, and only then would they become actual threats to democracy.