國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
37
Chapter Four Results
The present chapter will be divided into two sections to present the results of the applied survey. The first section of this chapter was based on the descriptive analysis that illustrated the frequency found in the distributions for the demographic variables and other major variables. The second section of this chapter focused on the statistical findings obtained through hierarchical multiple regression analysis.
4.1 Descriptive analysis
Demographics of respondents
As Table 4.1 showed among 703 respondents the male population accounted for a total number of 309 respondents (44.1%), while 392 respondents (55.9%) were females. In terms of the age of the population, those who were between 18-25 years old accounted for the highest percentage with a 72.1 (N=410), followed by the group of those who were between the ages of 26-30 with 27.1% (N=158), and lastly those aged 31-35 years old that accounted for 4.1% (N=27). The respondents' average age is 23.3 years old. In addition, from all the respondents, 70.2% of them received degrees from colleges and universities. The monthly income of the sample was mostly NT$6,000 and below (43.2%, N=303).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
38
Table 4.1 Demographic Information
Demographics Percentage Number
1. Gender
Male 44.1% 309
Female 55.9% 392
2. Age
18-25 72.1% 410
26-30 27.1% 158
31-35 4.1% 27
3. Educational level
Elementary --- 0
Junior High School --- 0
High School 3.3% 23
College 70.2% 492
Post-graduate 26.5% 186
4. Monthly Income
NT$6,000 and below 43.2% 303
NT$60,001- NT$10,000 17.1% 120
NT$10,001- NT$20,000 9.0% 63
NT$20,001- NT$30,000 8.1% 57
NT$30,001- NT$40,000 11.0% 77
NT$40,001- NT$50,000 6.4% 45
NT$50,001- NT$60,000 2.7% 19
Above NT$60,001 2.4% 17
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
39
Mainstream Media Platforms
Table 4.2 showed that only 23.6% of the population read newspaper regularly for news and only 38.2% used TV regularly for news. Both the degree of TV news frequency and attention are higher than newspaper. Regarding time spent on Facebook, respondents reported that they spent an average of 3.38 hours on Facebook on a daily basis.
Table 4.2a Frequency of Getting News from Traditional Media Platforms (N/%) Never Seldom Sometimes Often Always Total
Newspaper 54/7.7 259/36.9 223/31.8 131/18.7 34/4.9 100
76.4% 23.6%
TV 23/3.3 151/21.7 256/36.8 204/29.4 61/8.8 100
61.8% 38.2%
Table 4.2b Attention on News from Traditional Media Platforms (N/%) Very low Low Middle High Very High Total Newspaper 57/8.1 141/20.1 330/46.9 142/20.2 33/4.7 100
28.2% 71.8%
TV 39/5.5 131/18.6 355/50.5 152/21.6 26/3.7 100
24.1% 75.8%
Partisanship and Political interest
As it can be seen in Table 4.3 the majority of participants, 70.8%, were pro-political independence. In other words, most of the population ages 18 to 35 reported no affinity with pan-blue or pan-green alliance. Moreover, among the 703 respondents, 66.1% of the respondents reported that they have no political affiliation to any party.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
40
It was noted that only 18.1% of the respondents felt affinity with political parties, out of which the DDP (N=111) and KMT (N=82) hold the top positions with the largest numbers of followers. Among other parties, Green party (N=12) and Social Democratic Party (N=6) hold the top positions with the largest numbers of followers.
Table 4.3 The Categories of Respondents' Partisanship
N %
Partisanship
Pan-blue camps 85 12.1
Political independence
No Political affiliation 465 66.1%
Other Parties rather than pan-blue and pan-green parties
33 4.7 %
Pan-green camps 120 17.1
Political efficacy
The survey also inquired into respondents' evaluation of his/her political efficacy, the political system, and collective acts of the general public. As it can be seen in Table 4.4, in terms of internal efficacy, the beliefs in one’s own ability to understand and participate in politics, 69% of respondents agreed that “although politics and the government seem so complicated, a person like me still can really understand what’s going on”, and 52.2% of respondents reported that “they have enough knowledge to discuss political issues or join political affairs”. In regards of collective efficacy, the belief in capability of the general public to act collectively to effect social change, 72.2% of them agreed with the sentence “Taiwanese collective action can change the society”, and 79.5% of them agreed “Taiwanese collective action can affect the development of public affairs”. As for external efficacy, the belief in the political system as responsive to public demands, only 14.6% responded that
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
41
“Taiwan politic system can effectively respond to public opinion”, and merely 8.6% of respondents reported that “Taiwan government can effectively react to public opinion”.
‧
Table 4.4 Respondents' Political efficacy
Scale Items
Level of concern Strongly
Disagree
Disagree Neutral Agree Strongly agree
N % N % N % N % N %
Internal efficacy
A person like me still can really understand what’s going on.
9 1.3 88 12.5 121 17.2 423 60.2 62 8.8
13.8% 17.2% 69 %
I have enough knowledge to discuss political issues or join political affairs.
Taiwanese collective action can affect the development of public affairs.
13 1.8 47 6.7 84 11.9 439 62.4 120 17.1
7.5% 11.9% 80.5%
External efficacy
Taiwan political system can effectively react to public opinion.
74 10.5 364 51.8 162 23.0 90 12.8 13 1.8
62.3% 23% 14.6%
Taiwan government can effectively react to public opinion.
181 25.7 331 47.1 131 18.6 51 7.3 9 1.3
73.8% 18.6% 8.6%
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
42
Political discussion and offline political participation
This fourth survey delves into offline political discussion and political participation in different areas. In Table 4.5 it can be appreciated that 31.7% of the respondents regularly talk about politics with friends/family. From the sample it was appreciated that in real life only a few of the respondents between the ages of 18 to 35 regularly participated in activities such as attending a public hearing and policy seminar (1.2%), contacting official or political actors via telephone or e-mail (0.7%), signing a hard copy petition on politics or public affairs (5.7%), attending a public rally and electoral campaign (2.2%), participating in any demonstrations, protests, or marches (6.4%), and being involved in public interest groups, political clubs, or party committees (i.e.
Losheng Youth Alliance and the Anti-Media Monster Youth Alliance) (4.5%). Above all, most of the respondents rarely discussed politics with friends or family while they had almost never taken part in any form of political participation. In terms of political discussion, compared to table 4.5 and 4.6, aged 18 to 35 Taiwanese citizens were more likely to discuss political issues regularly in person (31.7%) rather than on Facebook (14.3%).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
43
Table 4.5 Frequency of Political Discussion and Offline Political Participation (in %)
Item
Never Seldom Some- times
Often Always Total
N % N % N % N % N %
Talk about politics with friend/family 38 5.4 155 22 287 40.8 166 23.6 57 8.1 100
68.3% 31.7%
Attended a public hearing, policy seminar 536 76.2 100 14.2 58 8.3 8 1.1 1 .1 100
98.8% 1.2%
Contact officials or political actors via telephone or e-mail 574 81.7 88 12.5 36 5.1 5 .7 0 0 100
99.3% 0.7%
Signed a hard copy petition on politics or public affairs 405 57.6 130 18.5 128 18.2 28 4.0 12 1.7 100
94.3% 5.7%
Attended a public rally and electoral campaign 493 70.1 126 17.9 68 9.7 13 1.8 3 .4 100
97.8% 2.2%
Participated in any demonstrations, protests, or marches. 389 55.3 154 21.9 115 16.4 32 4.6 13 1.8 100
93.6% 6.4%
Been involved in public interest groups, political clubs, or party committees.
(i.e Losheng Youth Alliance and the Anti-Media Monster Youth Alliance.)
495 70.4 103 14.7 73 10.4 22 3.1 10 1.4 100
95.5% 4.5%
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
44
Political activities on Facebook
This first part of the survey tried to obtain relations among three forms of political activities on SNS in different areas (As Table 4.6 & 4.7). In regards to Facebook use for information, 66.1% of respondents regularly get information about political issues and current events from Facebook Friends' post of news articles or videos, and 38.7% of respondents regularly search for information about political issues and current events from Facebook friends, fan pages, or clubs.
In regards to Facebook for expression, 16.7% of respondents regularly forward the information about politics or current affairs; 12.6% of respondents regularly posted comments about personal experience of participating public affairs or political activities on Facebook, including articles, news, or videos; 12 % of respondents regularly posted comments about politics or current events about articles, news, or videos; 14.3 % of respondents regularly discuss politics or current events with their friends, family, or colleagues via Facebook personal messages. Regarding the connection with public political actors, as shown in Table 4.7, respondents reported that they connected political actors on Facebook such as legislatives (21.2%), media commentators (37.9%), social movement activists (57.9%), and government officials (29%). Also, there are 53.1% of respondents who think over half of social movement activists they connected with on Facebook are political opinion leaders, followed by media commentators (52.6 %), government officials (44.7%) and legislators (40%).
The result shows that the political actors which respondents connected with on Facebook as opinion leaders the most are social movement activists. It seems that after the 318 flower student movement, the youth aged 18-35 has an affinity for social movements more than other political actors.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
45
Table 4.6 Frequency of Facebook information and expression use
Scale Items Never Seldom Some-
times
Often Always
N % N % N % N % N %
Facebook use for information
I normally get information about political issues and current events from Facebook Friends' posts of news, articles, or the videos.
5 .7 69 9.8 164 23.3 348 49.5 117 16.6
33.9% 66.1%
I search for information about political issues and current events from Facebook Friends, Fan pages, or clubs.
38 5.4 175 24.9 218 31 198 28.2 74 10.5
61.3% 38.7%
Facebook use for expression
I forward the information about politics or current affairs 168 23. 243 34.6 175 24.9 87 12.4 30 4.3
83.3% 16.7%
I posted comments about personal experience of participating in public affairs or political activities on Facebook, including articles, news, or videos.
189 26.9 259 36.8 167 23.8 63 9 25 3.6
87.4% 12.6%
I posted comments about politics or current events about articles, news, or videos. 229 32.6 249 35.4 141 20.1 66 9.4 18 2.6
88% 12%
I discuss politics or current events with my friends, family, or colleagues via Facebook personal messages.
138 19.3 261 37.1 206 29.3 70 10 30 4.3
85.7% 14.3%
‧
Table 4.7 Frequency of Connection with Public Political Actors
Items Public Political Actors
Legislator Media
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
47
4.2 Statistical Analysis
Prior to running the multiple regressions for testing the hypothesis, the study contains a table of zero-order Pearson's correlations among the independent variables, dependent variables, and control variables. Table 4.8 indicates that political independence (r=.121, p < .01), internal efficacy (r=.354, p < .001), collective efficacy (r=.224, p < .001), newspaper news use (r=.237, p < .001), Facebook intensity (r=.473, p < .001), political discussion (r=.473, p < .001), Facebook use for information (r=.344, p < .001), Facebook use for expression (r=.566, p < .001), and connection with public political actors on Facebook (r=.445, p < .001) would be inclined to respondents' political participation. Besides, females were more inclined to be involved in offline political participation.
Tests of Hypotheses
Multiple regression was then conducted to examine the main effects on the research question and Hypotheses 1 and 2, which stated that Facebook for information, Facebook for expression, and connections with public political actors affect offline political participation. The independent variables included all the controls and the three aspects of Facebook usage. Offline participation was used as a dependent variable.
RQ1.What is the effect of Facebook use for political information on political participation?
As shown in Table 4.9, the use of Facebook for political information has no significant association with offline political participation. In other words, the variable of Facebook use for political information cannot effectively predict offline political participation.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
48
H1. Political expression on Facebook is positively correlated to political participation.
Regression analysis supported hypothesis 1, the use of Facebook for political expression has a significant and positive association with offline political participation (β = .362, p < .001). Respondents who frequently engaged in expressive action on Facebook were more likely to exhibit high levels of political participation offline.
H2. The connection with public political actors on Facebook is positively correlated to political participation.
As predicted by Hypothesis 2, connection with public political actors has a significant and positive association with offline political participation (β = .178, p < .001). With regard to the control variables, internal efficacy (β = .086, p < .05) and offline political discussion (β = .249, p < .01) has a positive influence on offline political participation. However, TV news use has a negative influence on offline political participation (β = -.111, p < .001), suggesting that television news programs might reduce the intention of some respondents to participate. Based on these findings, people who use Facebook for political expression and have a connection with political actors tend to participate in political participation more actively.
As table 4.9 indicated, all the tested factors contributed to the model took up 46.1% in explaining offline political participation. In prediction of political participation, demographic variables explained 2.5% of the total variance (R2 = .025, p < .01), political antecedent factors explained for 13.1% (R2 = .131, p < .001), media
& discussion factors accounted for 15.1% (R2 = .151, p < .001), and political activities on Facebook (Facebook information use, Facebook expression use, and the connection with political actors) yields an incremental variance explained of 15.4%
(R2 = .154, p < .001). The political activities on Facebook (Facebook information use,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
49
Facebook expression use, and the connection with public political actors) accounted for 15.4% of the variance for offline political participation. This incremental variance was explained by all the detailed controls included in the model. Also, all initial significant beta coefficient in block 1 (demographics) vanished when the remaining blocks (blocks 2, 3, and 4, namely political antecedents, media &discussion, and Facebook use) were entered in the regression analysis.
Above all, users’ internal efficacy, TV news use, Facebook expression use, and connection with public political actors were significant predictors to users’ political participation.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
50
Table 4.8 .Zero order correlations among all independent and dependent variables in the study
Note: Cell entries are two-tailed zero order correlation coefficients. (N = 1,196). * = p < .05, ** = p < .01, *** = p < .001.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1.Gender ____
2.Edu -.093* ____
3.Income -.211*** .237*** ____
4.FB time .013 .099** -.038 ____
5.Panblue .066 -.047 -.012 -.030 ____
6.nonparty -.094* .009 -.024 .006 -.168*** ____
7.Intereffi -.243** .078* .135** .007 .008 .095* ____
8.Colleeffi .021 -.023 .022 -.011 -.098** .154*** .302*** ____
9.extereffi .038 -.056 .047 -.068 .277*** -.115** .019 .017 ____
10.newspaperuse -.164*** .098** .036 -.021 .075* .050 .324*** .166*** .077* ____
11.Tvuse -.062 .058 .083* -.030 .114** .067 .249** .131** .079* .535*** ____
12.Discuss -.063 .072 .005 .070 -.026 .109** .392*** .247*** -.089* .361*** .334*** ____
13.Fbinfo -.050 .072 -.091* .268*** -.098** .089* .246*** .208*** -.131*** .157*** .099** .282*** ____
14.fbexpress -.137** .053 .004 .248*** -.073 .087* .299*** .150*** -.003 .226*** .042 .345*** .474*** ____
15.fbpo -.110** .113** .029 .169*** -.107** .201*** .227*** .231*** -.138*** .137*** .069 .335** .362*** .390*** ____
16.parti -.148*** .062 .050 .129** -.063 .121** .354*** .224*** -.072 .237*** .063 .473*** .344*** .566*** .445*** 1
‧
Table 4.9 Hierarchical multiple regression predicting for offline political participation
Block 2: Political Antecedents Partisanship (pan-green=0)
Pan-blue -.063 -.020 -.018 .008
Block 3: Media & Discussion
Newspaper news use .237*** .121** .059
TV news use .063 -.181*** -.111**
Facebook intensity .129** .095** -.015
Political Discussion .473*** .390*** .253***
Incremental R2 (percent) 30.7 15.1***
Block 4: FB political activities
FB information use .344*** .013
FB expression use .566*** .362***
Connection with political actors .445*** .178***
Incremental R2 (percent) 15.4***
Total R2 (percent) 46.1
N=703. Cell entries for all models are final standardized regression coefficients for Block 1, 2, 3, and 4. **p<0.05; **p<0.01; ***p<0.001.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
52
Chapter Five:
Discussion
5.1 Implications of results Political activities on Facebook
The study has contributed to assess the role of Facebook in fostering offline political participation among young adults aged 18 to 35. While some of the results replicated existing findings, this thesis contributes to the study of the impact of Facebook on participation by differentiating among several dimensions of Facebook use. Overall, the survey findings indicated that political activities on Facebook play important roles in the life of Taiwanese youth. It seems that Facebook has become a main source of news to retrieve information related to political issues among the youth in the age group of 18 to 35 years. Also, this platform is now the new form of political expression and the new connection hub between political actors and followers. While TV news use is also associated with offline political participation among Taiwanese citizens aged 18 to 35, Facebook use for expression and connecting with public political actors for predicting offline political participation is more significant.
The present research empirically tested the influence of the use of political information, political expression, and the connection with political actors. The results showed that the use of Facebook for political information has no significant association with offline political participation. The null finding for the information use on Facebook explanation deserves further examination. Following news and political information on Facebook is not a significant predictor of political participation. The results in this thesis is in line with previous findings by Valenzuela (2013). He suggested that due to the redundancy of hard news content in social media,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
53
the content related to political issues and public affairs were still supplied by mainstream media organizations. Thus, by controlling for news use in traditional media platforms, the variance of political participation explained by Facebook for political information became insignificant. However, although Facebook use for information cannot be predicted for offline political participation. According to the survey, 66.1% of respondents regularly receive information about political issues and current events from Facebook friends' post of news, articles, or the videos. Also, 38.7% of respondents regularly search for information about political issues and current events from Facebook friends. Facebook is a platform that keeps Taiwanese youth informed about political affairs.
In addition, as the hypotheses proposed in this study, the political expression and connection with political actors on Facebook exert significant and positive impacts on individuals’ offline political actions. Thus, results indicated that achieving a better understanding of how citizens use Facebook for participating in political activities may help clarify the new paths that spur political action in the real world. In regards to the connection with public political actors on Facebook, the study showed that the influence of this variable on offline political participation is remarkable.
Similar to the findings from Tang & Lee (2013)'s study on Hong Kong college students, Facebook can influence people politically through facilitating people to build connection with public figures within the political arena. Connection with these public political actors seems to have tremendous influence of young people's political behaviors. In fact, in the survey, there are 53.1% of respondents who think over half of social movement activists they connected with on Facebook are political opinion leaders, followed by media commentators (52.6 %), government officials (44.7%) and legislators (40%). These percentages are substantial. One of the possible interpretations is that these social movement activists might play important roles in
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
54
leading the public opinion among young people after the student-led 318 Sunflower movement. These findings are particularly important for the existing literature related to politics because the connection with political actors is a variable seldom examined in the existing literature.
Political efficacy
In regard to political efficacy, similar to other research’s findings (Gil de Zúñiga
et al., 2012; Lee, 2006), which showed that people who have a higher sense of internal efficacy are more likely to engage in offline political participation. Taiwanese youth have a high level of internal and collective efficacy, while having a low sense of external efficacy. This result further reinforced the findings of previous studies based on the data from Election Studies Center that college students tend to have lower sense of external efficacy but higher internal efficacy (Chen & Huang, 2007; Chen &
Chen, 2013). Young adults have personal political effectiveness but hold negative attitudes toward the responsiveness of the government and political system. They are more critical and retrospective toward the current government and political system.
Also, Lee (2006) suggested how the collective culture would play an important role in collective societies is similar to internal efficacy in individual societies. Both Taiwanese youth and Hong Kong citizens have high level of collective efficacy.
However, the results showed that there is no relationship between collective efficacy and political participation. The findings suggested a different story than Lee (2006), in which research found that Hong Kong citizens who have a higher level of collective efficacy are more likely involved in offline political participation. Although these aged 18 to 35 Taiwanese youth have a remarkable high level of collective efficacy, they didn’t put these beliefs into action in the offline world. The relevance of
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
55
collective efficacy to other collectivist cultures needs to be further examined.
Media use
The relationship between media use and political participation is also worth discussing. The findings showed that people who watch TV news less are more likely to be involved in political action. The finding also supported the claim by Putman (1995) regarding time replacement, which meant TV deprived time that could be spent on civic activities. Putnam suggested that television viewing was negatively linked to social cohesion and it provides us with a pseudo-personal connection to others, in which viewers feel engaged with the community without the effect of actually being engaged. Television as a medium creates a false sense of companionship. Also, because television news favored ‘‘debate’’ formats over investigative reporting, it contains less contextual information. Broadcast media lacked space to provide significant context or mobilizing information; even cable news often favored ‘‘debate’’ formats over investigative reporting (Boyle &
Schmierbach, 2009). The relationship between time spent with Facebook is also worth discussing. In this thesis, time spent on Facebook is not related to offline political participation. This finding is consistent with past studies (Tang &Lee, 2013; Gil de Zúñiga et al., 2012).
Partisanship
In terms of offline political participation, from the total population, 66.1% of respondents possessed no political affiliation and 4.7% respondents supported the parties other than pan-blue and pan-green camps. Compared to the data from Election Study Center in 2015 which found that 43.4 % of the total Taiwanese population were independent voters, the finding showed that the number of independent voters among
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
56
young adults tends to go far beyond that national sample. Over half of Taiwanese aged 18 to 35 don’t have any party attachment. It can be interpreted from Taiwan’s social context: For years politics in Taiwan have been long a fight between the KMT alliance of the pan-blue camp against the DPP and its respective pan-green alliance.
That changed in 2014 when the independent candidate Ko Wen-je took the position of Taipei mayor, attracting unprecedented attention with a real chance to truly introduce
That changed in 2014 when the independent candidate Ko Wen-je took the position of Taipei mayor, attracting unprecedented attention with a real chance to truly introduce