In recent years, there has been seen an increase in political participation among young people. The U.S. 2008 presidential election registered a record-high level of youth voters (Kirby & Kawashima-Ginsberg, 2009) . This continued support of young people was crucial in determining Obama’s re-election four years later (Pew Research Center, 2012) . Youth voters became decisive in the 2012 U.S. president election and their electoral participation was likely to be encouraged by their close social networks or campaign and civil organizations (Circle, 2012). Also, the 2011 General Election (GE 2011) in Singapore has been regarded as a “distinct shift” in the political landscape, stemming from a young generation of voters with higher political engagement than before (Lin & Hong, 2015) .
In light of this heightened political engagement among youth, there has been growing research on various channels by which youth acquire political norms and attitudes. Such studies on political socialization have centered on the integrality of communication at home, among friends, and through media in helping youth develop their abilities and motivations to engage in political activities (Lee, Shah, & McLeod, 2013). Due to fast technological advancements, there have been sharp changes to the socio-political milieu that young people now grow up in, with evolving individual values and family roles (van Deth, Abendschön, & Vollmar, 2011) . Young people’s political engagement was usually influenced by parents’ political engagement (e.g., Hyman, 1959 ), school political activities (e.g., Hess & Torney, 1968 ), and online media use (e.g. Warren & Wicks, 2011 ). For instance, according to Cross & Young (2008b) , young people join a political party because their parents are also a member of the party. In addition, Lopes, Benton, & Cleaver (2009) pointed out that school activities are able to affect students’ political efficacy. Besides, according to Dalton (1996) , online media use can enhance young voters’ political discussion.
This study will thus focus on the different ways that Singaporean youth are socialized politically, also taking into consideration of changes in the technology savvy, tightly regulated, Asian political context of Singapore.
Prior studies found that election campaigns were usually for increased consolidation of partisan attitudes among adolescents (Sears & Valentino, 1997). During the campaigning period, news of politicians and parties can be more easily accessed by young voters on multiple platforms. Young Singaporeans also actively sought GE 2011 election and party information from traditional media and the Internet. They tend to perceive online media’s political and election content as more credible sources than their older electorates (Lin & Hong, 2015) . Young people’s heavy dependence on online media means that they are likely to engage with politics online than participating in traditional political gatherings (Zhang, Johnson, Seltzer, & Bichard, 2010) . This study distinguishes political participation into offline participation and online participation in order to examine how they are affected by the political socialization process. Offline political participation refers to the respondents’ non-Internet involvement in some political activities, such as attending a political rally or being volunteers to help a political party. Online participation refers to the respondents’ political activities on the Internet, such as writing about the election or matters related to the election on their Facebook pages or taking part in online forums to discuss political issues related to the election.
Singapore is a unique context for studying the role of youth’s political socialization. In this country, as political discussions often take place more in close social networks than in public, young voters’ political socialization influences from families and peers are likely to be more significant than other contexts, which is worth investigating. A holistic understanding of social influences from family and peer in shaping youth political attitudes is pertinent, especially when youths play an increasingly active role in the local political sphere. In addition, compared with the tight censorship of traditional media, the flow of online information is relatively free under lighter content regulation. Political websites and blogs have been found to be influential sources to Singaporeans as these online media provide alternative perspectives rarely shown in traditional media (Hong, Lin, & Ang, 2015) . During GE 2011, young Singaporeans consumed more online media content than older voters and considered them to be more important (Lin & Hong, 2015) . Thus, it is also crucial to investigate how old and new media might impact Singaporean young voters’ political participation. The findings will contribute to the understanding of how political socialization plays a role in shaping youth’s political participation in a context where drastic differences exist in traditional and new media.
2. Literature Review
2.1. Singapore Youth in the Local Political System
In past years, youth political apathy in Singapore has been the subject of many parliamentary discussions and national speeches. With the domination of the People’s Action Party (PAP) since the 1950s, Singapore’s status as a virtual single party state has led to a worrying lack of interest in the local political scene among young people, as they have been made to feel powerless in influencing government policy decisions (Noh & Tumin, 2008; Tay, 2011a) .
In 2000, the Speakers’ Corner at Hong Lim Park was opened to promote free speech and act as a venue for the free airing of views (Lee, 2005) ; in 2004, the Prime Minister Lee Hsien Loong declared that Singapore would work towards an open and inclusive government in future politics, and to further “promote civic involvement” (p. 134). Regardless of the criticisms such gestural policies have faced, other incidents over the years seem to support the notion of changing levels of participation in societal issues. Skoric, Ying, & Ng (2009) cite events such as public protests and the breaching of the Film Act by political filmmaker Martyn See in 2005 as evidence of an evolving political landscape. More recently in 2008, 60 students from Singapore’s Nanyang Technological University (NTU) gathered in the Speakers’ Corner to protest the censorship of an article related to a major opposition political party in the university news bulletin (Reuters, 2008).
GE 2011 had consequently represented a major turning point in Singapore politics. In the lead up to GE 2011, all Singapore political parties reported an increase in youth involvement (Lim, 2010) . The actual elections brought forth an overwhelming number of young voters aged between 21 and 34, making up about one-quarter of the total 2.21 million voters (Singapore Elections Department, 2011) . The unexpected results of GE 2011, which saw the highest proportion of contested seats since independence and a 6-percentage point swing against the PAP, cemented the increasing importance of young people in adding alternative voices to the local political scene.
Following from the above, this study intends to explore the political participation of young people in Singapore today, specifically, whether their political participation had been affected by the political socialization processes of parent and peer discussions and media use during GE 2011.
2.2. Youth and Political Participation in Singapore
Political participation can be defined as “legal activities by individual citizens that are more or less directly aimed at influencing the selection of governmental personnel” (Verba, Nie, & Kim, 1971: p. 9) . It can be in “any form or activities that impact on decisions” (Saad & Salman, 2013: p. 10) , including voting or supporting or withdrawing support in the political arena. In addition, according to delli Carpini (2004: p. 418) , the essence of political participation is citizens acting “through voting and other forms of electoral involvement, contacting public officials, membership in civic organizations, volunteering in their community, or even protesting and demonstrating.”
Traditional political participation might include electoral related activities such as voting and joining campaign activities, as well as non-electoral activities such as citizen-initiated contacts with governmental officials and cooperatively joining organizations (Verba et al., 1971) .
Away from traditional political participation, there is now a new way for people to engage in political activities, i.e. via online platforms such as the Internet (Gainous, Marlowe, & Wagner, 2013) . The affinity with new media displayed by young people has further brought hopes of reconnecting them with civic and political life (McMillan & Morrison, 2006) . The impact of new media on youth political participation could be seen clearly in the 2008 U.S. presidential election, where young people mostly relied on new media as their primary means of sharing election information (Wu, 2009) . The Internet allows people to share and exchange opinions on political issues through forums, blogs, discussions boards, and other interactive platforms (Althaus & Tewksbury, 2000) . Political participation can thus happen in another form, in which users become influencers through writing articles and comments (Kim & Rhee, 2006) .
In Singapore, a similar phenomenon of increased youth political participation via new online platforms could also be observed. According to Tan, Chung, & Zhang (2011) , young Singaporeans seem to be more politically active than their older counterparts, and to consume more political content through online channels. Despite much evidence pointing towards the increased participation of youth in politics, the unique local political system has made the study of this concept a complex issue. Singapore is one of 30 democracies worldwide where voting is compulsory for citizens over 21 years of age (Hill & Louth, 2004) . A survey by The New Paper found that this law was the only reason why 40% of Singapore youth actually voted (Tay, 2011b) .
To eliminate this potential bias, our investigation of offline political participation among the young will include attendance of political gatherings and volunteering as well as voting, whereas investigation of online political participation will include involvement in forums and commenting on Facebook. One of the key issues for this study is how the effect of political socialization relates to these two forms of political participation.
Liu’s study (2009) noted a low sense of political efficacy in Singapore university students. Seemingly content with their affluent lives and satisfied with the government’s performance as long as their material needs are met, most Singapore undergraduate students tend to be uninterested in the state’s political development and display a tolerance towards the many essentially social control measures exercised by the government under the guise of “parental guidance.” In other words, they are unconcerned with personal liberty and rights as long as the government can provide for and ensure their material comfort. As a result, there is a general attitude of indifference to governance and civic affairs.
Improvement in material security may partly explain Singapore youth’s political apathy, but are there other causes to the lack of interest to learn about and participate in politics? We believe that, with young people, political socialization is a more significant factor in affecting political attitudes and behavior. In the sections below, we will discuss how this learning process works via communications from parents and peers.
2.3. Political Socialization
Political socialization is defined as the learning process in which people accept various political attitudes, values, and actions from their surroundings. According to Shah, McLeod, and Yoon (2001), it is important to consider how communication patterns shape individuals’ perceptions about the government and their political orientations. For the young, such perceptions are usually derived from communications from their parents, peers and school ( Jennings, Stoker, & Bowers, 2009; Schwarzer, 2011 ; Lee, Shah & McLeod, 2012). The above channels, or agents and sources of political socialization, exert a major influence in shaping youth political attitudes (Easton & Hess, 1962) .
For instance, there is a positive correlation between political discussions with parents and adolescent political participation (Hwang & Kiousis, 2010) . An individual’s school experience also has significant impact on his or her future political participation (Humphries, Muller, & Schiller, 2013) . Studies have pointed out that political behavior is acquired through learning from as early as childhood (Sigel, 1989: vii) . Through the political learning process of discussion with family members, individuals may find out and then adopt the party affiliation or other political behavior of their elders, while a clearer understanding of why certain political issues are supported or opposed can be obtained in the more equal discussion with peers of a similar age. To establish social identity, young people tend to “follow the crowd” and make themselves belong to particular groups (Ashforth & Mael, 1989) , therefore, they are likely to choose to support a stance or candidate that has the majority’s backing. At the same time, when some members of a group show an intention to participate in political activities during discussions, others may be influenced to join in and become similarly involved.
Based on the above, this study will examine the political socializing agency of family and peers and the relationship with young people’s political participation.
2.4. Political Socialization within the Singapore Family
The family is usually accorded the importance of being the primary agent or the “foremost among agencies” in the political socialization process (Hyman, 1959) . Jennings & Niemi (1968) were the first to use social learning theory as foundation for research on political socialization, finding a high correlation between the political attitudes and behavior of parents and those of their children. Other studies (Davies, 1970; Renshon, 1973; Calavita, 2003) further affirmed the strength of family influence in children’s early formative years, terming this the cornerstone of one’s political attitudes. The home context is an irreplaceable socializing agent on the political beliefs of young people, to the extent that other positive influencing factors (e.g. the local political context) cannot work in the absence of this social construct (Pacheco, 2008) .
Political interest in teens is especially affected by the political socializing agency of parents (Cross & Young, 2008a) , and the effect is even stronger when both parents share similar political ideologies (Jennings & Niemi, 1974) . There is a positive correlation between the political participation of parents and that of their teenage children (Warren & Wicks, 2011) , and a young person usually joins a political party because at least one of his or her parents is also a member (Cross & Young, 2008b) . There is no doubt that the family is a powerful agent in affecting youth political attitudes and behavior.
In the Singapore context, the influence of the family unit is also indubitable. Minister Mentor Lee Kuan Yew termed the family “the building brick of society” when referring to the family-oriented culture of Asian societies like Singapore (Kim, 1994: p. 190) . In the close-knit Asian family, information exchanges might be even more frequent than in Western families, and the positive effect of political discussions with parents on adolescent political participation (Hwang & Kiousis, 2010) ought to be also more evident. Thus, we propose the following hypotheses with respect to the impact of political socialization by family on the political participation of Singapore youth:
H1: Political socialization by family is positively related to offline political participation. H2: Political socialization by family is positively related to online political participation.
However significant the influence of family, we cannot ignore the impact of Western culture on Singapore youth either, as manifestation of major differences between the younger and older generations can already be seen. For instance, in a 2011 study by Tan, Chung, and Zhang, the younger generation’s comparative leaning towards a less authoritative political stance suggests that their political attitudes can no longer be entirely accounted for by the influence of their older counterparts.
In further support of this alternative view, Calavita (2003) found that some youth reported different political values and beliefs from their parents. Such research points to the fact that, while the important role parents play in shaping their children’s political beliefs cannot be disputed, other social factors besides the family also contribute to the political socialization process. In the following section, the influences of peers will be discussed.
2.5. Political Socialization in Singapore Schools & the Workplaces
Past research has shown that education plays a critical role in the political socialization of young people, allowing for social mobility and affecting one’s perception of political values and norms (Abramson, 1967) . According to Lopes, Benton & Cleaver (2009) , at the school level, student demographics, school programs, and activities are found to affect students’ political knowledge and efficacy. Citizenship education is also able to promote civic and political engagement. The school environment has further been widely identified as a major source of peer influence on political attitudes (Hess & Torney, 1968; Nie, 1996) . Chandler (1974) suggested that although the family’s influence is hardly deniable, peer’s opinions also hold a strong force on young people’s cognition of things around. He explained that while in the household the parents usually reserve the strongest opinions; in the group of peers, expressing thoughts and disagreements is a back-and-forth interaction where everyone could hold a standpoint. The possibility of a more democratic scenario could explain why the peers may affect youth’s political preference.
In Singapore however, efforts such as the 1997 introduction of National Education into the school curriculum, designed to prepare students for important political roles in the future, have been criticized, partly due to the PAP’s involvement (Koh, 2006) . The government seems to be viewed as hypocritical in their encouragement of youth participation in Singapore politics, since youth activism remains a somewhat taboo subject, and is only allowed in a highly circumscribed and restricted form (Huang, 2006) . Despite this, there is still a small presence of student clubs such as the National University of Singapore (NUS) Students’ Political Association.
The work organization is also extensively postulated as a determinant of political behavior (Pateman, 1970) . An analysis by Elden (1981) of the relationship between how a workplace organizes itself and the political efficacy of the employees found a positive correlation between workplaces that empower workers in their own area of work and the workers’ attitude towards participating in democratic processes. For the working young adult in Singapore, the workplace is perhaps a location where a large amount of peer influence takes place, since Singaporeans on average work some of the longest hours compared to people in other developed countries (Federal Reserve Economic Data, 2013) .
Based on a summary of the above literature review, the hypotheses this study puts forward are:
H3: Political socialization by peers is positively related to offline political participation. H4: Political socialization by peers is positively related to online political participation.
2.6. Offline and Online News Exposure in Singapore
The media is considered an important agent in affecting political education and cultivating participation (Eveland & Scheufele, 2000) . People may even be guided in their voting decision through information on political developments obtained from the media (McCombs, 1994) .
Many studies found that the traditional media use might have impacts on people’s political knowledge and political involvement. For example, Leshner & McKean (1997) showed that watching political news on television improves voters’ knowledge of candidates, and Hoffman & Thomson (2009) found that watching local news and late night news programs on television has a significant effect on the civic participation of high school students.
With the advancement of the Internet in recent years, the potential of online media for engaging young people in civic affairs and changing the general attitude of indifference has become the focus of many political socialization studies (Bachen, Raphael, Lynn, Mckee, & Philippi, 2008) . Opportunities offered by the new online channels of learning about political candidates have enhanced voters’ interest in politics and political discussions (Dalton, 1996) . The media use indeed has an impact on young people’s political attitudes and behavior.
With respect to the situation in Singapore, mass media have traditionally only promoted a pro-governmental stance, avoiding possibly offensive political discussions due to the restrictions of the Newspapers and Printing Presses Act and the Defamation Act (Lee, 2002; Seow, 1998) . The proliferation of new media thus presented an opportunity for a blossoming of the local political sphere, as a platform where alternative views could be transmitted and heard (Zhang, 2013) . Early years however saw the PAP extending their authoritarian rule to the Internet political sphere, locking down against public criticisms of the government (Rodan, 1998) . 2001 brought further tightening of Internet political control, where newly introduced regulations required popular political websites such as the now defunct Sintercom to register and be monitored by the government (Huang, 2006) . These strategies had initially worked in minimizing the potential the Internet has in allowing youth to engage in a wider range of political activities online.
However, a relaxation of political rules online came in 2010 and 2011, which made possible direct contact between political parties and the public and an increased use of new media in GE 2011, especially by the younger generation (Hoe & Saad, 2011) . This could possibly explain the findings of a recent study conducted by the Institute of Policy Studies during GE 2011, which showed that, although
Singaporeans still relied heavily on mainstream media for political news consumption, there was an increase in the consumption of online political news compared to just the previous year (Teo, 2011) .
Online media has undoubtedly become a pertinent source of political information in recent years, particularly for the younger generation in Singapore. Their natural affinity with technology coupled with their response to the restrictions of mainstream media makes them more susceptible to be influenced by new media (Tan, Chung, & Zhang, 2011; Lin & Lim, 2010) . According to Hong, Lin, & Ang, 2015) , more than 56% of Singaporeans―many of them young― have some experience using political (or socio-political) websites or blogs, including the Yawning Bread, Wayang Party, The Online Citizen, Sgpolitics.net, Talking Cock.Com, and mrbrown.com, which has allowed them to be exposed to viewpoints different to the mainstream. In Singapore’s climate of political control by the ruling party and apathy on the part of citizens, the new promises of democratization and increased political participation and activity brought about by the advent of the Internet seem especially eloquent (Yeo & Banerjee, 2003) .
With respect to the influence of offline news exposure (e.g., using print newspapers, radio and TV to contact election news) and online news exposure (e.g., reading foreign news websites or Singapore blogs or news websites) on youth’s political participation, we propose the following hypotheses:
H5-1: Offline news exposure is positively related to offline political participation.
H5-2: Offline news exposure is positively related to online political participation. H6-1: Online news exposure is positively related to offline political participation. H6-2: Online news exposure is positively related to online political participation.
Telephone survey was the research method applied for this study. AC Nielsen Research (Singapore) Pte Ltd was commissioned to conduct the post-election national telephone survey, by which some 2,000 Singaporeans aged 21 and above were interviewed from 24 May (two weeks after Polling Day) to 17 July 2011. All the respondents were informed that their personal information is kept anonymous. Using random stratified sampling, the sample satisfied quotas for race, age, and gender according to the 2010 population census. The Computer-Assisted Telephone Interviews (CATI) was conducted in English, Chinese, and Malay, each lasting about 25 minutes. Among the 2,080 respondents, 447 were aged between 21 and 35, whose responses became the data adopted for our research. Of this data set, 56.4% were male and 43.6% were female; 42.7% were aged 21 - 24, 32% were aged 25 - 29, and 25.3% were aged 30 - 34. The race distribution was 72.3% Chinese, 16.3% Malay, 8.5% Indian, and 2.9% other. Educational level was specified as follows: 2.5% polytechnic diploma and up, 10.7% ITE and other vocational certificate, 10.5% junior college (A Level), 8.7% secondary O or N Level, 32.9% primary school (PSLE), and 34.7% no formal qualification.
4.1. Dependent Variables: Political Participation
4.1.1. Offline Political Participation
Based on the suggestions of Lee et al. (2013), and Barnidge, Macafee, Alvarez, & Rojas (2014) , we created the offline political participation variable by tallying the answers to the four survey items that ask respondents whether in the past six months they had attended a political rally; volunteered to help in a political party; attended a meeting of discussion or dialogue organized by the Residents’ Committee, Community Centre, or the Government; taken part in an event for a good cause, for example, a walkathon, a flag day, or other charity events. The answers were coded 0 = No, 1 = Yes (The results showed that the Cronbach’s alpha value is .62; Mean = .83, SD = .66).
4.1.2. Online Political Participation
Based on the suggestions of Wang (2013) , and Gainous et al. (2013) , the online political participation variable was created by tallying the responses to the four survey items which ask respondents how often during GE 2011 they wrote about the election or matters related to the election on their personal blogs, Facebook pages, or Twitter; wrote or commented on other people’s blog, Facebook page, or responded to a tweet on the election or matters related to the election; took part in online forums to discuss political or social issues related to the election; shared online content on the election or matters related to the election with others by email, Facebook, or Twitter. The answer choices were “never”, “1 - 3 times”, “4 - 6 times”, “7 - 9 times”, “10 or more times”. A 5-point Likert scale was used, where 1 = never, 2 = 1 - 3 times, 3 = 4 - 6 times, 4 = 7 - 9 times, and 5 = 10 or more times. (The results showed that the Cronbach’s alpha value is .84; Mean = 1.44, = .79).
4.2. Independent Variables
4.2.1. Political Socialization by Family
Two questions were applied to measure this variable: 1) how often did you discuss the elections with your family members? 2) How often did you hear your family members talk about political news, public affairs, or government policies? A 5-point Likert scale was used, where 1 = never, 2 =rarely, 3 = sometimes, 4 = often, 5 = very often. The above measurement was based on the suggestions of Westholm (1999) and Diemer (2012) (The results showed that Cronbach’s alpha value is .61; Mean = 2.90, SD = .95).
4.2.2. Political Socialization by Peers
In order to measure this variable, two questions were asked: 1) how often did you discuss the elections with your friends? 2) How often did you hear your friends talk about political news, public affairs, or government policies? Again, a 5-point Likert scale was used where 1 = never, 2 = rarely, 3 = sometimes, 4 = often, 5 = very often. The above measurement was derived from Barnidge et al. (2014) . (The results showed that Cronbach’s alpha value is .64; Mean = 2.83, SD = .92).
4.2.3. Offline News Exposure
In line with Kononova, Alhabash, & Cropp (2011) , the variable was measured by having respondents answer this item: how many minutes you usually spent a day reading/listening/watching election news during the period of the election.
Please provide your answer in five-minute intervals (5, 10, 15, 20 minutes, etc). (open-ended question): 1) reading print newspapers 2) listening to the radio 3) watching television. The results showed that Cronbach’s alpha value is .39. With the value of reliability so low, we decided not to combine the three items as one variable in further statistical analysis, but to treat them separately as three independent variables instead (TV, Mean = 28.66, SD = 42.09; Radio, Mean = 8.34, SD = 38.22; NP, Mean = 19.16, SD = 25.08).
4.2.4. Online News Exposure
With reference to the measurement developed by Warren & Wicks (2011) , the variable was measured by having respondents answer this item: how many minutes you usually spent a day reading/listening/watching election news during the period of the election. Please provide your answer in five-minute intervals (5, 10, 15, 20 minutes, etc) (open-ended question): 1) reading foreign news websites such as the BBC, CNN or the New York Times 2) reading Internet-only Singapore blogs or news websites such as The Online Citizen, Yawning Bread, Temasek Review ( The results showed that Cronbach’s alpha value is .70; Mean = 9.70, SD = 24.05).
4.3. Control Variables
The demographic variables of age, gender, education, income, and race are controlled in this study. As has been pointed out, there is a link between race and political participation (Verba, Schlozman, Brady, & Nie, 1993) , and Singapore has a multiracial population of about 75% Chinese and 25% Malay, Indian, Eurasian, and other minorities. In addition, as researches have noted the effect of parent education level on children’s political participation (McIntosh, Hart, & Youniss, 2007) , and the positive correlations between political interest and online political participation (e.g. Kaye & Johnson, 2002) and between internal efficacy and political participation (Morrell, 2003) , these are all controlled for our exploration of the relationships between the major independent variables and the dependent variable.
Political interest was measured by respondents’ agreement to the questions “I am interested in political issues” and “It’s enough for me that others take care of government and public affairs as I have no wish to be part of it” (reverse-coded) with 5-point Likert scales (1 = strongly disagree, 5 = strongly agree) (the Cronbach’s alpha value is .60; Mean = 3.22, SD = .77). Internal political efficacy was the average of respondent’s agreement with two 5-point Likert scales (1 = strongly disagree, 5 = strongly agree): 1) I have a pretty good understanding of political issues; 2) I feel that I don’t have any say about what the government does (reverse-coded) (the Cronbach’s alpha value is .89; Mean = 2.96, SD = .63).
In order to figure out whether political socialization by family, by peers, and offline and online news exposure are positively related to youth’s offline political participation, hierarchical regression was applied. From Table 1, we see that among the demographic variables, only age (respondents were aged between 21 and 34) is significantly correlated with offline political participation, with younger voters tending to have more offline political participation (Beta = −.143, p < .01) Mother’s education is also positively related to the dependent variable (Beta = .242, p < .01), in that respondents were more likely to have higher offline political participation when mother’s education level was higher.
In addition, Block 2 shows that political interest is significantly associated
Table 1. Demographic variables, parent education level, political interest, news exposure, political efficacy and political socialization on offline political participation.
*Note: *p < .05, **p < .01, ***p < .001; N = 447; Gender coding: Male = 1, female = 0; Race coding: Chinese = 1, Non-Chinese = 0.
with offline political participation (Beta = .231, p < .001), i.e., people with higher political interest were more likely to have higher offline political participation. However, the results in Block 3 indicate that internal political efficacy is not significantly correlated with the dependent variable (Beta = .013, p > .05).
Block 4 indicates the relationship between offline news exposure and offline political participation. It is shown that no aspect of offline news exposure is positively related to the dependent variable. Moreover, there is not a significant relationship between online news exposure and offline political participation either, as indicated by the results in Block 5.
Furthermore, as correlation analysis showed that the two possible factors of political socialization (by family and peers) are significantly correlated (r = .571, p < .001), causing a potential problem of multicollinearity, we chose to run them separately in different regression models.
From the data in Block 6I, we see that political socialization by family is significantly correlated with offline political participation when demographic variables and the variables of parent’s education level, political interest, political efficacy and news exposure are controlled (Beta = .123, p < .05). In other words, offline political participation tended to be higher among youth who more frequently discussed the elections with their families/heard about the elections from their families. Political socialization by peers is also significantly correlated with offline political participation (Beta = .131, p < .05), thus, the more often people discussed the elections with their peers or heard about the elections from their peers, the more likely they were to join in related activities.
Based on the above results, H1 and H3 are supported; H5-1and H6-1 are rejected.
To explore whether the two sources of political socialization (family and peers) are significantly correlated with youth’s online political participation, hierarchical regression was also applied. In Table 2, Block 1 shows that all of the demographic variables and parent’s education are not positively related to online political participation. In Block 2, we can see that political interest is positively related to online political participation (Beta = .301, p < .001), as people with higher political interest tended to have higher online political participation. However, Block 3 shows that there is no significant association between internal political efficacy and the dependent variable (Beta = .073, p > .05).
From Block 4, we see that only newspaper reading is significantly correlated with online political participation (Beta = .218, p < .01): youths who read newspapers more were also more likely to have higher online political participation. On the other hand, online news exposure is significantly correlated with online political participation, as indicated by Block 5 (Beta = .178, p < .01).
Finally, the results in Blocks 6I and 6II show that political socialization by family (Beta = .107, p < .05) and political socialization by peers (Beta = .179, p < .01) are both positively related to the dependent variable of online political participation.
Thus, H2, H4, H6-2 are supported; H5-2 is partially supported.
Table 2. Demographic variables, parent education level, political interest, news exposure , political efficacy and political socialization on online political participation.
Note: *p < .05, **p < .01, ***p < .001; N = 447; Gender coding: Male = 1, female = 0; Race coding: Chinese = 1, Non-Chinese = 0.
5. Discussion and Conclusion
Young people have often been characterized as distrustful of politics, uninterested in political participation. For some scholars, therefore, the process by which the young might develop political attitudes and become engaged with politics makes a worthy and important research subject (Niemi & Klingler, 2012) . This study attempted to discover whether political socialization by family and peers are positively related to the political participation of Singapore youth, beginning from the assumption that since political discussions in Singapore are not encouraged in public, the information exchanges that take place with parents in close-knit families and the sharing of political perspectives with peers at the workplace and in school might possibly have a greater influence on young people’s political behaviors. Our results indeed indicate that political socialization by the two sources of family and peers is significantly correlated with young people’s political participation in both offline and online forms. In keeping with past research (e.g. Humphries et al., 2013 ), we found that youth who more frequently discussed election-related issues with family members and peers were more motivated to participate in offline political activities such as attending a political rally. In addition, their online political participation also tended to be higher, in ways such as taking part in forums or sharing related content with others by email, Facebook, or Twitter. This is again consistent with the finding of extant research (e.g. Hwang & Kiousis, 2010 ). Therefore, we can conclude that a significant correlation exists between Singapore youth’s political participation and their political socialization by family and peers.
The other aspect the study wished to explore whether offline and online news exposure might have a correlation with youth political participation. The results indicate that newspaper reading was associated with youth’s online political participation but not with their offline political participation. The authors believe that, since people who read newspapers more can be said to care more about current affairs and socio-political issues, they would also have greater inclination to access political news and election information online. For young digital natives who are accustomed to using the Internet and have stronger political interests or attitudes, a stronger inclination of using online media to express their views and participate in political and election issues can be accounted for by the ease of access, ease of use, anonymity, and perceived safety of online platforms. Yet why is it that newspaper reading does not have a significant relationship with offline political participation? We would suggest that for youths, the less familiar behavior of offline participation (such as going to a political rally) requires more persuasion to adopt and possibly cannot be adopted without the encouragement of intimate persons like peers and family members.
Furthermore, online news exposure was found to significantly correlate with youth’s online political participation. Young people, who spent more time following election-related news online, such as via Singaporean sources on the Internet like The Online Citizen and Yawning Bread, were more likely to have higher online political participation. This is in line with previous observations (e.g. Kirby & Kawashima-Ginsberg, 2009; Leong, 2011 ) and provides further support to the view that Internet media use can promote political participation (Wen, Hao, & Cherian, 2013) . The advent of new online media has brought about a gradual thawing of attitude towards politics among the young, who are now participating in politics in previously unavailable ways (Vitak, Zube, Smock, Carr, Ellison, & Lampe, 2011), and our results indicate the promise of improving engagement with and participation in politics among young Singaporeans. When Singapore youth thus become more interested in political participation and concerned with civic affairs, true democracy in Singapore might be a real possibility.
The finding is that a significant correlation exists between political interest and both online and offline political participation among Singapore youth parallels observations made by studies conducted in the West. Zhou’s (2010) study at Northwestern University noted political interest to be the most significant factor to predict traditional political participation, and Carlisle & Patton (2013) also found a correlation between political interest and online political participation during election. Young people with an interest in political issues are more inclined to care about government and public affairs, and in turn become engaged to participate in political activities. However, our finding that internal political efficacy does not significantly correlate with political participation is contradictory to previous research (e.g. Verba & Nie, 1972 ). For a possible explanation, the authors would suggest that young people may think they understand political issues without necessarily being interested in politics, and in the unique political environment of Singapore, political participation must be even harder to motivate without enthusiastic interest.
One other result worth mentioning is the positive correlation between mother’s education and respondent’s offline political participation. While this is consistent with the extant findings that parents’ education level has a significant impact on youth’s civic development (e.g. McIntosh, Hart, & Youniss, 2007 ), it is interesting that in Singapore’s predominantly Chinese and therefore patriarchal society, youth political participation is significantly correlated with mother’s education and not father’s education. Although other research has suggested the role-model effect of a politically active mother and the possible impact of mother’s education on children’s level of political engagement (Gidengil, O’Neill, & Young, 2010) , it remains to be explored wherein the root of this disparity in Singapore lies, that is, whether the role played by mother’s education in promoting children’s political participation is due to the strong political involvement of higher-educated mothers, or simply to the situation that mothers have more opportunities to discuss politics and social issues with their children in comparison to the fathers who tend to be the breadwinners and have less time to spend at home.
To conclude, we wish to discuss the limits of this study and propose some directions for further research. First, the so-called political socialization is a longitudinal process involving social interactions and should require long-term observation to be understood in depth. This study only examined the political socialization of Singaporean youth by family and peers and its relationship with political participation within the time frame of GE 2011, which must be a major limitation. Second, this study treated political socialization as offline discussion with family and with peers. However, a further distinction should be made for online discussion with family and with peer in future survey designs, even though young people might not discuss political issues with family members online. Third, strength of partisanship might be another variable that could impact political participation, but was unfortunately not included in the original survey. This oversight should be attended to in the future. Fourth, we should consider implementing a full structural equation model in future analysis, thus allowing ourselves the possibilities of testing the influence of any intervening or antecedent variables in the same model, as well as determining which forms of participation weigh more heavily on a combined participation outcome.
Finally, future research may focus on a comparison of the effects of political socialization on youth political attitudes in different countries, especially ones where the mass media have more controls. In addition, as Singapore’s political environment is undergoing significant changes with the advent of new media, every election may provide different data on young people’s political attitudes and behaviors and how those might be affected by political socialization. A longitudinal study to examine the evolution should be a worthy subject for continued observation and research.