What Is Youth Political Participation? Literature Review on Youth Political Participation and Political Attitudes

Looking at political participation behavior of young adults in contemporary Europe, this paper provides the reader with a map of different terminologies and logics that are used to discuss youth political participation. The existing literature is examined through the lens of five guiding questions: what defines youth political participation? How does youth political participation differ from adult political participation? How do young adults develop political attitudes? How does youth political participation differ across Europe? What methods are being used to analyze youth political participation? For those researching youth political participation for the first time, this paper offers a useful overview of the topic. At the same time, it gives researchers who are already well-informed the opportunity to reflect on the current state of research in this field. Finally, this paper indicates where future research is needed.


INTRODUCTION
Looking at the political participation behavior of young adults in contemporary Europe, one is faced with a contradiction. Representatives of the disengagement paradigm within the literature underpin their argument with empirical findings, such as young adults being the least likely to vote in national elections, the drop of youth membership in political parties, and generally low levels of political interest. On the other hand, the literature on an engagement paradigm of youth participation represents a more optimistic view as it is based on findings in the context of new forms of political participation, which are more appealing to and are used more frequently by young adults.
Both perspectives raise questions about the role of young adults in European democracies. The two mentioned positions represent the respective end points of a much more nuanced line of research on this topic. Research in this area can appear confusing, but overall it is clear that a comprehensive picture of both the degree and the modes of youth political participation is lacking. This paper tries to take a first step in the direction of addressing this problem. The goal is to provide the reader with a map of the different terminologies and logics that are used to discuss youth political participation. To attain this goal, this paper presents insights from the existing literature on the following guiding questions: y What defines political participation? y How does youth political participation differ from adult political participation? y How do young adults develop political attitudes? y How does youth political participation differ across Europe? y What methods are being used to analyze youth political participation?
The first step is to provide a structured inventory. On the one hand, this paper will be helpful for those encountering this research area for the first time as it provides an overview of the previous research in the field of youth political participation in Europe. On the other, it offers well-informed researchers the opportunity to reflect on the current state of research in this field. In addition, this paper clearly points toward where further research is needed.
With this in mind, I develop three main arguments within this paper. First, although existing definitions of political participation are adequate to capture youth participation, the current literature is inconsistent in the inclusion of new modes of participation that are increasingly common among young adults. Second, there are both methodological and substantive problems within the existing literature, which emerge from young adults' different conceptions of politics as well as from their differing awareness to adults of what constitutes political participation. Third, and resulting from this, the current state of research in this area lacks larger cross-national studies that take into account an adequate conception of how the youth define political participation and that conduct comparative research on youth political participation behavior, which is necessary if we agree that young people hold the key for the future functioning of our political systems (Hooghe et al., 2004).
The rest of this paper is structured as follows. The next section gives an overview of the development of political participation research. In doing so, it includes a review of the definitions of political participation then and now and clarifies why it is important to be familiar with those definitions when looking at the political participation behavior of young adults. The following section deals with the (potential) differences between the political participation behavior of young adults and adults. Besides an overview of the ongoing debate on whether and, if so, what kind of differences there are, this chapter clearly indicates which role the question of measurement plays in this. The fourth section focuses on the political attitudes of young adults after reviewing political socialization research, for political socialization plays an important role in the formation of the political attitudes of young adults. After this, section 5 gives an overview of youth political participation across Europe. Section 6 then presents methods previously used in the context of youth political participation. As usual, the final section summarizes the previous sections and highlights which questions remain unanswered. It thereby tries to provide an answer to the question of youth political participation as it actually is and indicates where future research is needed.

THE DEVELOPMENT OF POLITICAL PARTICIPATION REPERTOIRES AND RESEARCH
Political Participation research has undergone significant developments over the course of the last few decades. Multiple disciplines have contributed to broadening our understanding of the field, but because of this multidisciplinary input it has become less clear what the underlying core assumptions and definitions are that make up the term Political Participation (PP). This section therefore sketches the development of the term and answers the core question of what defines political participation.
Signing a petition, joining a party, or casting a vote are the most commonly accepted actions deemed as PP. But that's about as far as agreements go. To answer the question of how PP can be defined, one has to go back a few decades. In 1973, Robert Dahl offered a first glimpse of what it might mean. In "Poliarchy: Participation and Opposition" he declares PP an essential part of modern democracies as it enables citizens to hold their governments accountable (Dahl, 1973). However, Dahl didn't explicitly define his concept of participation. His definition only implicitly covered actions within the given institutional framework of a nation, meaning that actions such as consumerism (Stolle et al., 2005) or just hitting a "like" button wouldn't be categorized as participation, even though they could be seen as holding governments accountable. His works nevertheless contain some fundamental elements of our modern conception of PP-namely accountability as well as the dichotomy of private citizens and professional politicians, which can also be found in the well acknowledged works of Verba and Nie (1972).
To these researchers, political participation is "those activities by private citizens that are more or less directly aimed at influencing the selection of governmental personnel and/or the actions they take" (Verba and Nie, 1972, p. 2). According to Verba and Nie, private citizens have the ability to participate in politics not just by casting votes or joining parties but through numerous other activities. Their suggested typology consists of voting, campaign activity, contacting public officials, and cooperative or communal activities.
This definition has paved the way for the analysis of actions such as protests, strikes, or petitions as activities that participate in politics through other means than elections (Verba and Nie, 1972, p. 47). Similar concepts have been presented by Parry et al. (1992) or Pattie et al. (2004), who, in contrast to Verba and Nie (1972), stress that political participation does not necessarily have to address governments but could also target other institutions or even organizations. PP can therefore affect the policymaking process as well as services provided by governments, such as education or health care (Pattie et al., 2004;Fox, 2014). Brady (1998) adds that, in order to qualify as PP, actions taken by private citizens must be observable, manifest, and voluntary, but he also focuses on interactions between citizens and political elites.
Parallel to developments in participation-research, authors such as Flanagan (2013), Norris (2002), Putnam (2001), Zukin et al. (2006), or Daskalopoulou (2018 have been working on the concept of civic engagement, which has several intersections with PP research. The concept of civic engagement has been used to analyze all kinds of citizen behavior, including activities and actions, which can but don't necessarily have to be political. Putnam's "bowling alone, " e.g., also includes going to a bowling alley as a vital indicator of engagement. The ever-growing repertoire of indicators has therefore led to accusations of conceptual stretching (Berger, 2009), meaning that the conception is too broad and therefore not suitable for researchers. Most authors' conceptions of PP from the first period of research have three aspects in common: Actions have to be taken by private citizens, not politicians; these actions have to be voluntary, meaning structural forces that require citizens to take certain actions wouldn't count as participation; and their actions need to target governments, institutions, organizations or NGOs. These three aspects are at the "hard core" (Lakatos and Musgrave, 1970) of almost every single contemporary definition of PP.
van Deth (2001) nicely summarizes the evolution of political participation repertoires between the 1940s and the 1990s by tracing the publication of landmark studies. In the 1940s and 1950s, PP was mainly restricted to voting and campaign activities. In the early 1960s, though, appeared the later so-called "conventional" modes of PP. At this time, "political participation was broadly understood as activities concerned with traditional conceptualizations of politics as campaigning by politicians and parties, and with well-accepted contacts between citizens and public officials" (van Deth, 2001, p. 5). During the 1970s, these conventional forms were expanded and "unconventional" forms, which were not in line with the societal norms of the 1970s, appeared. These unconventional forms included, among others, protest and rejection as well as new social movements, such as women's or pacifist movements (van Deth, 2001). Later, in the 1990s, the borderline between the political and non-political spheres of modern society disappeared as the political participation repertoire came to include "civil" activities such as volunteering and social engagement (van Deth, 2001). Nowadays, further forms of PP have emerged and challenge PP research. The new forms use non-political behavior to express political opinions, and what was once defined as unconventional or elite-challenging is now commonplace. Therefore, these forms can no longer be captured by a distinction between conventional and unconventional PP (Teorell et al., 2007). Furthermore, García-Albacete (2014) has found that citizens' political involvement has changed recently and argues that these changes characterize today's PP repertoire and have led to the distinction between institutionalized and non-institutionalized PP. First, "the agencies or structures through which citizens are mobilized and participate have (...) been transformed, with the spread of new social movements and advocacy networks" (García-Albacete, 2014, p. 15). Second, individualized patterns of participation are growing as ties to political and civic organizations become weaker (García-Albacete, 2014). The now widely used distinction between institutionalized and noninstitutionalized PP capture forms of PP which happen within the institutional framework (e.g., voting or party membership) and those which happen outside of the institutional framework (e.g., protest or boycotting). This distinction between institutionalized and non-institutionalized PP is particularly important for any kind of research on youth participation, given the fact that young adults are disproportionately more likely to participate through non-institutionalized means.
Adapting to or being challenged by new forms of participation is a continuous process. One of the more recent developments in this regard is online participation. The debate about how and if online participation fits into existing concepts is ongoing and vibrant (Gibson and Cantijoch, 2013;Dayican, 2014;Halupka, 2014;Kristofferson et al., 2014). Authors such as Morozov (2009) declare it as an illusion of participation, whereas Rojas and Puig-i-Abril (2009) see it as "expressive participation" which constitutes a "subdimension" (Rojas and Puig-i-Abril, 2009, p. 907) of political participation. Because of this heated debate and the numerous ways of integrating online participation into existing forms of participation, Theocharis (2015) warns that the entire concept of PP could face a risk of overstretching.
In order to avoid this fate, van Deth (2014) has offered a distinct concept of PP, which should enable researchers to "recognize a mode of participation if [they] see one" (van Deth, 2014, p. 5). In order to "see one, " researchers should look for these characteristics of participation: it is an activity; it is voluntary and not ordered by a ruling class or required by law; it refers to people in their role as non-professionals or amateurs; and it concerns government, politics, or the state (van Deth, 2014). This description represents the minimum definition to which further variants are added, namely "two additional variants based on the target (politics/government/state or problems/community), and two based on circumstantial evidence (contextual and motivational)" (Theocharis and van Deth, 2018a, p. 81). This conceptual map results in five analytically unambiguous modes of political participation (Theocharis and van Deth, 2018a). Thus, the first form (minimal definition) focuses on the arena of participation rather than its outcomes, while the second and third forms deal with the targets of the activities rather than relying on the goals or intentions of the people. In the fourth form, the political nature of the activities is based on contextual evidence, and only at the very last stage (form five) are the intentions/aims of the participants considered in order to identify a form of political participation. The authors therefore illustrate that "the advantage of following these decision rules is not only that we can distinguish between political acts that fit into definitions with stricter or loser requirements, but also that we can systematically exclude those who do not meet the definitional requirements" (Theocharis and van Deth, 2018b). Based on this concept, online PP could be recognized as a form of PP. However, this example also reveals that PP cannot be defined in a simple way, which is also reflected in the existing literature. Instead, it raises the question of whether a definition such as the one by van Deth does permit the development of means for unifying the existing discussion. At the same time, such a broad and yet clearly defined definition offers the possibility of being able to classify forms newly emerging in the literature. In the course of ever-changing social situations and behaviors, this seems to be a key aspect of developing a definition of PP, which can be used over the long term.

DIFFERENCES BETWEEN YOUTH AND ADULT POLITICAL PARTICIPATION
The previous section has already shown that there is no need for a separate definition of youth political participation; instead, the various forms of political participation used by young adults is of central importance. Following this, the question arises as to what extent the PP behavior of young adults differs from that of other groups. A look at the previous research shows a perceived gap between young adult and adults. Many studies show the lowest scores in almost all areas of political participation for young adults and thus the image emerges that young people are not sufficiently engaged in politics. This perception is based upon trends such as voting in elections, where young adults have the lowest rates, and these rates continue to decrease just as the level of youth membership in political parties is decreasing (Kimberlee, 2002;Hooghe et al., 2004;Fieldhouse et al., 2007;Cross and Young, 2008). To capture young people's disengagement in politics, it can be said that "young people are less concerned with politics, less politically knowledgeable, do not participate in social or political activities, are more apathetic, and have low levels of political interest" (Quintelier, 2007, p. 165). Even if this representation seems clear, the disengagement of young adults in politics remains a contested issue in the literature. At this point, three central questions need to be clarified. First, is there a real difference between the political participation behavior of young adults and adults? Second, which factors lead to a different behavior between young and old? Third, does this really mean that young adults are politically disengaged?
The first, and to some authors most important, reason for differences between youth and adult political participation is lifecycle. Here, one can find a curvilinear effect of age, which means that participation rises from youth until middle age, then decreases with old age. Scholars have been researching this trend for decades (Jennings, 1979), and it must be clearly differentiated between lifecycle effects and generational effects. In the context of lifecycle effects, what matters is the increase or decrease of political participation resulting from different stages of life (Nie et al., 1974).
In this sense, political participation is nobody's priority as it competes against more pressing personal concerns, especially for young people (Highton and Wolfinger, 2001). As people have a finite amount of time, engagement with politics is more costly for those who have not yet sorted out their lives (ibid.). In addition to the fact that young adults gain more experience with the electoral and political process as they age, some specific steps of adulthood have proven to have an effect on political participation behavior. These include, among others, settling down, marriage (Stoker and Jennings, 1995), graduating and getting a job. Some authors state that these results relate purely to the influence on voting turnout and that the influence of lifecycle effects on other forms of political participation can only explain minor differences (Quintelier, 2007). However, not all researchers share this view. Research in the area of non-institutionalized participation shows that lifecycle effects are also relevant here. Specifically, they concern personal availability and refer to "the absence of personal constraints that may increase the costs and risks of movement participation, such as full-time employment, marriage, and family responsibilities" (McAdam, 1986, p. 70). The social movement research shows relevant influences on young adults, in the sense that the absence of these kinds of constraints facilitates their participation (Beyerlein and Hipp, 2006;Saunders et al., 2012;Earl et al., 2017).
The theory of generation effects is based on the assumption that pre-adult socialization exerts enduring effects on political socialization. In this sense, the adolescence of each individual is the period relevant for the development of political thinking.
Building on this argument, some authors assume that as today's young adults are less active, they will never reach the level of political participation of the current elderly (Martikainen et al., 2005). One explanation for this is that young adults today are having more difficulty in reaching the milestones of adulthood (Arnett, 2014;Tagliabue et al., 2014) and that this results in an irreversible delay in political participation. Studies show that young adults retain these characteristics that distinguish them from previous generations and that this will lead to a replacement of the current electorate by a more passive generation of political participants (Quintelier, 2007).
Another reason for the perceived different behavior between the age groups derives from the varying definitions of the political or of political behavior. Every researcher needs to base his or her research on a clear definition. At the same time, this definition of the political or what is defined as political participation must also be used and accepted by the survey population. This is exactly where differences between young adults and adults emerge. Do young adults and adults view the same activities as political? Generally, studies show a difference between the definitions of researchers and survey participants. For example, Parry et al. (1992) found that only 18% of their survey participants interpreted a list of activities as political, which the researchers also defined as political. For this reason, some researchers call for a broader definition, which would lead to higher noted levels of political engagement (e.g., Roker et al., 1999). The definition question is also relevant when thinking of non-institutionalized forms of political participation. Young adults might not define their actions as political, even though they are actually political. Therefore, it is both about the individual's conception of politics/the political as well as their awareness of doing something political. Only a few studies focus on young adults' definition of the political, but they show that young adults use a narrower definition than both researchers and adults (e.g., Bynner and Ashford, 1994;Andolina et al., 2002). This results in young adults being less interested than adults in politics, because they do not view politics, in their narrow definition, as relevant to their lives (Andolina et al., 2002). In this sense, "the low political participation rate among youth is a by-product of their narrow conception of politics and their impression that politicians do not truly care about their needs" (Quintelier, 2007, p. 169). Hence, youth disengagement is a result of the organization of politics rather than of the youth's own lack of interest. For this reason, research is growing on how young adults define political participation and what they perceive as political participation (Henn et al., 2002(Henn et al., , 2005O'Toole, 2003;O'Toole et al., 2003a). They show that previous studies used a concept of participation that is too narrow and that, e.g., the topic of non-participation as an act of political action has so far not been sufficiently addressed (O'Toole, 2003).
Finally, a study by Quintelier (2007), which specifically examined the differences between the age groups, revealed that young adults and adults seem to be similar in their political attitudes, with the exception that young people have fewer opportunities to participate politically. Furthermore, they state that there are differences with regard to the engagement in specific forms of political participation as young adults tend to participate more in non-institutionalized forms. This leads to the conclusion that "it seems as if the problem of youth political participation is less a matter of whether they participate, and more a matter of where they participate" (Rainsford, 2017, p. 2).

POLITICAL ATTITUDES OF THE YOUTH
Just like the research on political participation, contributions to the field of political attitudes have also broadened our understanding of how political attitudes develop and how the political attitudes of young adults differ from those of adults. This section takes the different approaches to socialization and the debated inputs from other fields and focuses on development, maturation, and the stability of attitudes in order to answer how young adults develop political attitudes. Hyman (1959, p. 25) thought of political socialization as an individual's "(...) learning of social patterns corresponding to his societal positions as mediated through various agencies of society." Considering that it is one of the most commonly used definitions of political socialization, it is surprising that researchers had mostly analyzed family influence first and foremost and neglected various agencies of society. Furthermore, Sapiro (2004) points out that, in its early days, dedicated research on political socialization (Easton et al., 1969;Searing et al., 1973;Jennings and Niemi, 1974) mostly focused on shared party affiliations, participation in voluntary organizations, or the genuine political interest of children and their parents' possible influence on it. However, scholars have repeatedly faced the same methodological challenge, since young children do not possess many issue beliefs at all (Searing et al., 1973). This makes it hard to identify inferences valuable to political science. Hess and Torney-Purta (1967), on the other hand, claim that children are able to express political opinions and partisanship. This uncertainty caused researchers (Hanks, 1981;Percheron and Jennings, 1981;Nieuwbeerta and Wittebrood, 1995) to shift their focal point toward adolescents and young adults instead of children, because their issue beliefs could be accessed more easily due to the wider scope of methods available for gathering data. The driving force behind socialization research until now has been a biologically sound core assumption: The neurological structure, senso-motoric skills, as well as temperament, reactivity, semantic networks and behavior of infants and adolescents differs from adults (Kagan, 2003, p. 6-8). Dollard and Miller (1950) argue that this difference slowly deteriorates through learning because "human behavior is learned" (Dollard and Miller, 1950, p. 25).
In the "heyday" (Niemi and Hepburn, 2010, p. 10;van Deth et al., 2011, p. 48) of political socialization research the Columbia school (Berelson et al., 1954;Butler and Stokes, 1974) and Michigan School (Campbell et al., 1960;Easton et al., 1969) dominated the discourse. Both schools found that political affiliation and attitudes toward institutions and the authorities strongly correlate with whatever interests one's parents had and that these interests didn't change much over the span of a lifetime. Socialization research was equal to research on preference or opinion inheritance; almost all research focused exclusively on the United States and also suffered from selection biases as they mostly included white middle-class Americans. Niemi and Sobieszek (1977) note that this bias was compensated for mainly by Abramson (1977) and García (1973), who tried to answer why people of color feel less politically efficacious throughout multiple generations. Researchers posed interesting questions but couldn't identify any causal mechanisms. Clarke (1978) and Percheron and Jennings (1981) dissected differences between American and French families, concluding that "(...) the object of partisan socialization within the family is country specific" (Percheron and Jennings, 1981, p. 434), which remains true today.
However, political socialization is not only country specific; it also depends on the respective political context. This raises the question of generational dependency, i.e., if it makes a difference whether young adults themselves or their parents have been socialized in a specific political context. The various studies on this question reveal that political socialization is influenced by the broader context both during one's own political socialization (Grasso et al., 2019) and during the transmission from parents to children. In this way, researchers have shown that "if parents are politically engaged and frequently discuss politics with the child, transmission rates rise substantially, particularly on topics of general political significance and salience" (Jennings et al., 2009). Here, regular political events, as well as more episodic events, offer socialization opportunities for parents (Valentino and Sears, 1998). Furthermore, Jennings (1984) demonstrated that socialization can also be observed through social class and not just the direct transmission from parents to their children. This made a multitude of arguments part of the socialization process. According to Niemi and Hepburn (1995), up until the 1990s research had been suffering from two flawed implicit assumptions: Political attitudes, opinions and assumptions of today remain mostly the same tomorrow, and early learning is more important than learning in later life. Instead, they argue that adolescents' attitudes do change, often substantially, and do not necessarily settle just because they turned 18 and/or moved out. Only emerging longitudinal studies (Hanks, 1981;Alwin and Krosnick, 1991;Smith, 1999) made such findings possible. Niemi and Hepburn (1995) therefore demanded a revitalization of political socialization theory and research that would abandon these flawed assumptions. As if they had heard the call, Nieuwbeerta and Wittebrood (1995) challenged one of the cornerstones of socialization research: the idea of fathers being dominant in the transmission of party preferences. They found that there was a gender specific difference in the transmissions of these preferences, challenging decades of previous research. Their Dutch case showed that daughters were more likely to share their mother's party preference and sons were more likely to share their father's. With the focus on women, this influence of a mother on her daughter was also confirmed 15 years later in a Canadian context (Gidengil et al., 2010). Family settings change over time and different kinds of settings-such as stay-at-home parents, patchwork or single parenting-grow in numbers, which could lead to individualization and growing issue heterogeneity (Du Bois-Reymond et al., 2001;Inglehart and Welzel, 2005). The realization that "(...) socialization nowadays clearly occurs under different circumstances" (van Deth et al., 2011, p. 148) has cast doubt on most previous findings regarding the influence of parents. Many of core assumptions of socialization theories could not be reproduced with more sophisticated methods (Sears, 1990;Jennings, 2007), and research therefore still suffers from significant blind spots. "The questions, methods, and assumptions have been changed by 40 years of scholarship, political experience including regime change into and out of democracy, and altered political sensibilities" (Sapiro, 2004, p. 19). Political socialization theory struggles to deliver on its promises: "correlations between parents and their (mostly) biological children, with no way of separating the effects of the environment the parents provide from the effects of the genes they provide, and no way of separating the effects of the home environment from the effects of the environment outside the home. The evidence, in other words, is ambiguous." (Harris, 2000, p. 626).
Thus, it cannot only be parents who exert influence. While most researchers still assume that family has some influence, they still do not know how much of an influence that is. Other places of socialization that receive a lot of attention are the school, peers, and the media (Blais and Carty, 1990). Research on the influence of school has existed for a long time, and from the beginning its results have been in the area of conflict between those who see an influence (Himmelweit and Swift, 1969;Palonsky, 1987) and those who do not (Hyman, 1959;Easton et al., 1969). A central problem here is the difficulty of isolating the school effect from other effects (Banks and Roker, 1994). For this reason, researchers especially in more recent studies, try to keep the framework conditions constant, e.g., by looking at samples that vary only in one characteristic, such as the type of school. This should facilitate the isolation of the influence of the various factors from each other. Examples for Finland (Koskimaa and Rapeli, 2015) and Belgium (Quintelier, 2015) show that school has an influence, without being the most central one. Instead, in addition to family influence, the influence of peers is in the foreground. It has even been shown that "peers, through discussion and diversity, are even more influential and successful in creating greater political participation" (Quintelier, 2015, p. 65) than the family. Nevertheless, the media are also assigned a relevant role here. While earlier studies dealt with the influence of different forms of media, such as television news or newspapers (Atkin and Gantz, 1978;Garramore and Atkin, 1986), researchers have only recently begun to assess the influence of social media on the process of political socialization. The argument in relation to social media would be that they are characterized by less distinct boundaries between non-political and political activities, thereby lowering the thresholds of political engagement (Ekström and Shehata, 2018). However, previous studies can only partially confirm this (ibid.). This research strand therefore requires supplementary studies.
In addition to studies on parental and other social as well as structural transmission, a branch of interdisciplinary research is steadily growing which focuses on the genetic inheritance of attitudes and norms (Martin et al., 1986;Bouchard et al., 1990;Bouchard and McGue, 2003;Alford et al., 2005;Bell et al., 2009;Hatemi et al., 2011;Kudrnac and Lyons, 2017). These studies look at the development of attitudes, norms and values amongst twins or parents and their offspring either in a setting of continuous exposure to the parent/sibling or in a setting with deliberate discontinuities in their biographies. Moreover, the results appear promising: "If father and mother both hold a highly intensive justworld belief, the probability that their child will also hold a strong belief in a just world is very high" (Schönpflug and Bilz, 2009, p. 229). It is important to stress that researchers also warn that most genetic association studies greatly overinterpret their findings (Benjamin et al., 2012).
Beside these studies, and in the context of a more interdisciplinary view of the topic, psychologists like to refer to attitudes as "a person's general evaluation of an object (where 'object' is understood in a broad sense encompassing persons, events, products, policies, institutions and so on)" (O'Keefe, 2015, p. 13). However, social scientists struggle with this definition as it is far too broad to operate with. Batista Foguet and Saris (1997) would argue that the outcome of the aforementioned evaluation would have to be stable over time and that it would have to be consistent with previous evaluation in order to constitute an attitude. Researchers seem to agree on the fact that the backbone of an attitude is stability (Alwin and Krosnick, 1991;Wilson and Hodges, 1992;Zaller et al., 1992), even though critics argue that stability is not necessary for attitudes (Kahneman et al., 1999). But what happens to evaluation during maturation? Hooghe and Wilkenfeld (2008) argue that attitude development during maturation is not the same thing as changing an attitude: "Attitude development requires change in the quality of thinking, rather than merely change in thinking" (Hooghe and Wilkenfeld, 2008, p. 156). Previous authors had denied the existence of attitudes in young adults (Marsh, 1971;Searing et al., 1973).
Because of this maturation process, the attitudes of young adults differ from those of adults in many policy fields. But what attitudes are we talking about here? "Civic culture" by Almond and Verba (1963) is often (Galston, 2001;Sapiro, 2004;Dalton, 2008;Kam and Palmer, 2008;Quintelier and Hooghe, 2011;Hoskins et al., 2015) cited as encompassing a vital set of attitudes, including political interest and political trust, and, depending on the research design, researchers often measure civic culture by surveying interest and/or trust in politics. The stereotypical picture would be that young adults are less interested, more negative and that they don't trust political elites as such (Quintelier, 2007). As Rekker et al. (2015) has shown, multiple longitudinal studies reproduce the same result: Younger cohorts are less conservative on cultural issues but not on economic issues. Two specific fields of this are ethnocentrism and egalitarianism. Furlong and Cartmel (2012) confirmed these findings as well. Young adults also appear to be less materialistic (Rudig and Bennie, 1993). Alwin and Krosnick (1991) argue that the maturation process interfered with the core characteristic of attitude, namely stability. In their setting, the youngest group, whose members were aged between 18 and 25, was the least stable as far as their attitudes were concerned. Quintelier and Hooghe (2011), on the other hand, argue that attitudes among adolescents develop early and are likely to remain stable until adulthood. Eckstein et al. (2012) found common ground between both realms and argue that most young adults agree on aspects of good citizenship such as voting, helping others or taking part in organizations. But Henn et al. (2005) point toward a difference between attitude and action in the UK as young people are less likely to vote and less likely to even register for it in the first place. Eckstein et al. (2012) also mention a key issue of the entire field: "(...) there is still a lack of studies explicitly investigating young people's orientations toward political behaviors over a longer period of time in order to depict development. Furthermore, longitudinal studies that did account for changes revealed no coherent pattern of results" (Eckstein et al., 2012, p. 491).
The scarce research shows that young people's attitudes appear to be somewhat different from those of adults. In particular, the relationship between development, maturation and the stability of one's attitudes seems to be one of the most researched topics, yet it offers only a few insights. Eckstein et al. (2012) can only be supported in their demand for more longitudinal studies in this field.

DIFFERENCES IN YOUTH POLITICAL PARTICIPATION ACROSS EUROPE
Having discussed the possible differences that exist between young adults and adults and the role that political socialization plays, the next step is to look at how young adults and their participation differ across Europe. This section aims to illustrate the diversity of participation of young adults, which has already been covered by existing research. Of course, this cannot be an exhaustive view of all existing studies. Instead, it offers a nuanced view into different regions of Europe and, together with the following section on methods, provides the basis for identifying the research gaps in this area. Generally, each of the EU-member states' polities offers distinct institutionalized ways of participating. In 26 of the 27 member states, citizens need to be at least 18 years old in order to be eligible to vote; Austria, with its active voting-age of 16, is the exception. Keeping in mind these structural differences, this overview will nevertheless look at both the institutionalized and non-institutionalized participation of young adults in different regions of Europe.
Research on Northern European countries has had a great impact on questions of association membership and its effects on political participation. Torpe (2003) indicates that, among Danish youth, membership in associations is becoming looser and that this membership don't necessarily influence the likelihood of political participation. Coe et al. (2016) took a different approach and directly surveyed 10 political activists aged between 17 and 19 in Northern Sweden. On the basis of this study, Coe et al. conceived the concept of "Youth Politics as Multiple Processes" (Coe et al., 2016, p. 6), which indicates that youth political participation is characterized by very distinct restrictions such as age limits, adults' disinterest in youthdemands, and state-centered definitions of politics. Nygard et al. (2016) focused on variables deriving from "resource models" to explain different forms of political participation amongst Finnish 9th graders and found higher rates for alternative forms of political participation among this age group, given the right socio-economic resources. Wass (2007) emphasizes this by pointing out that the concepts of family socialization alone lack explanatory power, a point which was already discussed in section 4.
Youth political Participation in Eastern European countries has so far mostly been analyzed comparatively and with a focus on the anticipated effects of previous communist regimes in those countries. Slomczynski and Shabad (1998) argued for the polish case that democratic principles can be successfully taught in school in order to avoid extreme left or right tendencies which could result from a lack of democratic experience. Roberts (2003) partly contradicts these findings, arguing that, amongst other actions, political participation amongst young adults is deeply connected to the social environment as well as structural effects. Research on 10 eastern European countries conducted by Letki (2004) has shown that, in many cases, political participation in eastern European, post-communist countries is very similar to established western democracies. Association membership and established institutions also increase the chances of political participation in post-communist countries. In addition to this, Ådnanes (2004) found that young Bulgarians with a high degree of formal education consider migrating partly because they perceive their ways of participation as restricted and are unsatisfied with their political system, thereby confirming the importance of an established institutional framework. Burean and Badescu (2014) show that similar triggers of participation can be seen at the core of the protest movements against the Romanian government in 2012, where thousands of students took to the streets to protest against their government.
Apart from these countries, some EU-Member states, namely Greece, Spain, and Portugal, have been severely hit by the financial crisis and have also been suffering from a high degree of youth unemployment (Tosun et al., 2019), which appears to go hand in hand with decreasing institutionalized and increasing non-institutionalized forms of political participation among young adults. As a result of this crisis and its severe effects on young adults, the research on southern European countries has, e.g., and beside other forms of political participation (Sloam, 2014), provided valuable insights into youth political participation online. Online participation is genuinely perceived as less costly and therefore more easily accessible even during times of crisis, which is when Pacheco and Plutzer (2008) expect decreasing levels of participation. Espinar-Ruiz and Gonzalez-Rio (2015) as well as Calenda and Meijer (2009) have shown through large-N surveys that there is a significant relation between multiple forms of political participation and time spent on the internet. Theocharis (2011) research on Greece partly contradicts these findings. He argues that while the online realm is more likely to cultivate a post materialist mindset, it is also the case that this mindset seems to go hand in hand with a genuine disinterest in political participation. In addition to research on online participation, the financial crisis has granted remarkable insights into the relation of neoliberal policies and informal youth political participation (Sotiris, 2010;Sakellaropoulos, 2012;Zamponi and Gonzalez, 2017). This also applies to extremist positions (Koronaiou et al., 2015), showing that neoliberal policies often serve as the initial spark of protest or extremism, even though they do not represent the actual underlying cause.
Much like the research on northern European cases, research in central and Western Europe has significantly contributed to our knowledge of similarities between European nations in the forms of participation and political attitudes. These comparative studies (Timmerman, 2009;Cammaerts et al., 2014) show similarities in the participation of Europe's adolescents. Cammaerts et al. (2014) found that insufficient participation in the United Kingdom, France, Spain, Austria, Finland, and Hungary is due to the existing structural nature of the political systems and its discourse as adolescents mainly feel excluded from it. Within her research on municipalities in the UK and the Netherlands, Timmerman (2009) found that neither country offers enough entry points for young adults to contribute or participate in debates or the democratic process in general. Hooghe and Stolle (2003) found that adolescents in Germany, France and the UK are less likely to vote or participate through institutionalized means than adults, though their willingness to participate through non-institutionalized means is proportionally higher. Hooghe et al. (2004) and Quintelier and Hooghe (2011) also find this to be true for the Belgian case.
Previous research thus reveals a wealth of different forms of participation among young adults in Europe. The studies focused on very different areas, from membership in associations or voting behavior to political activism, e.g., in the form of protest. Here, young adults are exposed to different contexts, as, e.g., the case of Eastern Europe with many post-communist countries shows. The results of the studies also show which new spaces young adults use for participation and that participating in "older" spaces or institutionalized forms of participation can be problematic for them.

PREVIOUSLY USED METHODS TO STUDY YOUTH POLITICAL PARTICIPATION
In this final step, the focus is on how and with which methods youth political participation has been investigated so far. In the past, some authors addressed one of the central questions-namely how young adults perceive and define politics and political participation-and developed tools for assessing youth definitions of politics. These consisted, e.g., of a threeyear qualitative longitudinal study (Lister et al., 2003) and a quantitative survey (Vromen, 2003) of young adults and their perceptions of citizenship or of qualitative focus group studies that examined young adults' understanding of political engagement (Pontes et al., 2018). Researchers also tried to get closer to the "vocabulary" of young adults with regard to political participation (O'Toole, 2003;O'Toole et al., 2003a,b).
Furthermore, a number of recent research projects have analyzed youth political participation. They mostly used a cross-national comparative design combined with a mixed methods approach to emphasize different focal points. Within the YOUNEX (Youth unemployment and exclusion in Europe, Lorenzini and Giugni, 2012) project, e.g., researchers shed light on the consequences of long-term unemployment for youth political participation by both conducting in-depth interviews and original survey data. The EURYKA (Reinventing democracy in Europe: Youth doing politics in times of increasing inequalities, Kousis and Giugni, 2019) project, meanwhile, conducted both panel survey analysis and biographical analysis to investigate how inequalities are experienced by young adults and how these conditions can stimulate youth political participation. A third project worth mentioning here is EUYOUPART (Political Participation of Young People in Europe, Spannring et al., 2008), which was specifically concerned with the development of comparatively usable indicators that would facilitate the study of youth political participation. Here, three key points were identified that may limit the comparative usability of indicators. These limitations can stem from "failed or inaccurate translations of central terms used in a question, different opportunity structures in the countries that facilitate or hamper a form of activity or different political cultures that embed an activity in a different institutional context" (Ogris and Westphal, 2005). The importance of such an approach was also shown by later investigations using existing survey datasets. García-Albacete (2014), e.g., used data from the European Social Survey to show that indicators need to be tested for their usability both across countries and age groups.
When looking at the development of research on the political participation of young adults, the first thing that emerges is a clearly positive trend. Older studies mostly focused on establishing how adolescents are different from their adult counterparts in a descriptive manner. These studies therefore described youth participation behavior ex negativo in almost all designs. This begs the question of whether there is more to adolescents than just being non-adult. More recent studies have shown this to be the case and now hardly use this exclusive approach of comparison between young and old. Nevertheless, three points arise from this and the previous section that have so far received insufficient attention. First, recent studies do not always take into account our existing knowledge on the 'vocabulary' of young adults. At this point, it would also be worth discussing whether the existing knowledge is even sufficient or whether newer and updated studies are needed, since the possible fields of participation are constantly developing. Secondly, there is a lack of large cross-national studies that take several different contexts into account and thus explore how young adults resemble each other in their participation behavior or do not. Third, with few exceptions, the use of existing survey data sets not designed for young adults has so far paid insufficient attention to the suitability of the items used/developed for the study of young adult participation behavior.

CONCLUSION-WHAT IS YOUTH POLITICAL PARTICIPATION?
This review article pursued several goals, among which were to give an overview of the landscape of definitions of the term political participation and to work out the specific features of youth political participation. Furthermore, it aimed to shed light on the state of youth political participation in the European context and the methods previously used to investigate this, in order to be able to identify gaps in the literature and to suggest avenues for further research.
In the first step, it became clear that the decades-long debate on the definition of political participation has produced many small-scale definitions. The (few) broader definitions seem to be more helpful, even when considering that there is no independent definition of youth political participation. Although these recent definitions of political participation are adequate for capturing youth political participation, the current literature is inconsistent in the inclusion of new modes of participation that are increasingly common among younger generations. Resulting from this one major shortcoming is the fact that non-participation has not yet been problematized adequately. Although this issue was addressed a long time ago (O'Toole, 2003), it is still the case that research so far has paid little careful attention to this (Theocharis and van Deth, 2018b). This results in the danger of more frequent support of the disengagement thesis, which does not necessarily correspond to the actual participation situation of young adults.
In the second step, this paper sought to answer the question of how youth political participation differs from adult political participation. In general, it was shown that existing differences are interpreted differently by researchers (engagement vs. disengagement thesis). In addition, it also became apparent that the classic research design of comparing young and old, which was mainly used in older studies, is used less frequently in more recent studies. This is due to the existence of differences between what young adults define as "political" and what researchers define and interrogate as such. These definitions can differ not only between young adults and researchers, but also between young adults and adults. Inconsequently, problems can emerge from young adults' varying conceptions of politics and the "political" as well as from their differing awareness to adults of what constitutes a political act. Although some researchers tried to solve this problem by conducting research to get closer to the "vocabulary" used by young adults (O'Toole, 2003;O'Toole et al., 2003a,b), youth-specific explanations of what being politically engaged really means remain insufficient (Pontes et al., 2018). This results in a clear call for future research: It is necessary to develop further youth-specific explanations and definitions of what political participation means, which new studies should then employ accordingly. This aspect of youthadequate definitions and measurements must also be considered when using existing datasets.
Another, third major shortcoming is the lack of larger cross-national studies that take into account a youth-adequate definition of political participation and conduct research on the political participation behavior of youths. This certainly results from the absence of a unified theoretical foundation for studying "European" youth political participation. This is unfortunate considering the enormous amount of data available, especially from the EU. In addition, implications for European policy research can only be made on the basis of cross-country consistent studies.
In conclusion, it can be said that the definition of youth political participation is currently nothing more than general political participation. However, the question remains regarding the use of forms of political participation by young adults. Hopefully, this article will trigger other researchers to spend more time on this topic and both to resolve the mismatch between the definition of political participation and the perception of young adults regarding what is "political" and to review existing and upcoming datasets so that they can scrutinize this concept.

AUTHOR CONTRIBUTIONS
JW conceived and designed the article, wrote the manuscript, revised the manuscript, reread it, and finally approved the submitted version.

ACKNOWLEDGMENTS
The author would like to thank the reviewers for their helpful suggestions and comments. The author gratefully acknowledges the comments from Jale Tosun and all other members of the project team in Heidelberg. Marcel Katzlinger deserves credit for his research assistance and comments on previous versions of the paper. Finally, the author thanks Laurence Crumbie for language editing.