By skyforbes 
Eric Hsu, Researcher, Doublethink Lab
Key findings
- This study compares Taiwanese active and inactive TikTok users and finds important differences in where they get their political information, who they trust as information sources, and how they view Taiwan’s political landscape.
- TikTok active users are more likely to get their political and public affairs news from influencers, fan pages, and certain political figures on social media. In fact, more than twice as many active users rely on these sources compared to inactive users, showing just how important influencers are in shaping their views.
- Active TikTok users also report higher levels of internet addiction, with more saying they “somewhat agree” to addiction-related questions than inactive users.
- When asked who spreads misinformation and cyber army in Taiwan, most respondents point to the Democratic Progressive Party (DPP), regardless of TikTok use. However, inactive users are more likely to also blame the Chinese government (PRC).
- The DPP remains the most popular political party overall, but support is lower among active TikTok users. Notably, more active users favor the Taiwan People’s Party (TPP) compared to inactive users.
- Both groups believe the Kuomintang (KMT) stands to gain the most from the PRC’s strategies to influence Taiwan, but this belief is stronger among inactive users.
- Most people still see democracy as the best political system. However, active users are more likely to believe that authoritarianism might be better in certain situations or or indifferent about political systems compared to inactive users.
- Slightly more than half of respondents say they are satisfied with Taiwan’s current democracy, with no significant difference between active and inactive users.
- A strong majority (about 75%) agree that Taiwan’s judicial system faces serious issues. TikTok active users are more likely to agree (by about 7%), and the proportion who strongly agree is nearly 10 percentage points higher than among inactive users.
- On statements comparing the current government to the Chinese Communist Party (CCP) or questioning free speech in Taiwan, about 60% disagree overall. But active users are more likely to agree with these criticisms than inactive users, and this agreement has risen compared to past surveys.
- Both groups reject the idea that the PRC is a better place to live than Taiwan, but active users are more likely to want to prioritize economic cooperation with the PRC over national security concerns.
- When asked who is responsible for tension between Taiwan and the PRC, most blame the PRC government. However, active users are more likely to also blame Taiwan’s DPP and the U.S. government.
- Nearly 90% of respondents dislike the PRC government, but active users show less intense dislike compared to inactive users. Views on ordinary PRC citizens are generally less negative.
- Compared to data from the 2024 election survey, skepticism toward the U.S. has grown. Active TikTok users are more likely to believe that the U.S. would not come to Taiwan’s defense if the PRC attacked.
- Most respondents (over 70%) reject giving up democracy even if faced with PRC military invasion. Still active users are more likely to believe that unification wouldn’t significantly change everyday life.
- 55% of all respondents say they would be willing to fight to defend Taiwan if the PRC invaded.
- Both active and inactive TikTok users oppose the official narrative that unification with the PRC is inevitable, with 25% strongly disagreeing.
- Most respondents disagree with the claim that the people of Xinjiang are happy and that their culture is well protected by the PRC government. Although both groups mostly reject this narrative, active users express slightly less intense disagreement.
Background
TikTok is an international short-video social media platform owned by the Chinese company ByteDance, and its Chinese counterpart is called Douyin. Since its launch in 2016, the platform has faced significant scrutiny. Concerns have focused on data security, user privacy, the spread of misinformation, and the potential for the People’s Republic of China (PRC) and its affiliates to manipulate content through algorithms. Reports from the U.S. Department of Justice, Taiwan’s Doublethink Lab, and various Taiwanese think tanks have highlighted that the potential threat TikTok poses to public opinion in democratic countries should not be overlooked.
The short-video information format popularized by TikTok has significantly reshaped how the public consumes information. More specifically, the fragmented and rapidly intensive nature of information delivery has enhanced the effectiveness of manipulation techniques, including the spread of conspiracy narratives. In such an environment, cognitive biases shaped by narratives can exert a more subtle and pervasive influence on audiences.
According to a recent report by Doublethink Lab, The Impact of TikTok on Taiwanese Youth and PRC perceptions: A 2025 TikTok User Study, TikTok not only delivers daily entertainment content but also actively pushes political content to users. In the long term, this may continuously shape audiences’ perspectives on certain issues and influence their receptivity to specific political narratives.
This project conducted an online survey targeting TikTok (and Douyin) users in Taiwan to explore potential differences between active and inactive TikTok users. The survey examines various dimensions, including media consumption habits, perspectives on public and social issues, cross-strait issues, international politics, as well as other demographic variables and psychological indicators.
Relevant Projects
The design of the “2025 TikTok/Douyin User Survey Questionnaire” largely builds upon the election survey framework employed by Doublethink Lab during Taiwan’s 2022 local elections and the 2024 presidential and legislative elections. Additionally, it incorporates and references elements from the questionnaire designs of the “Taiwan Communication Survey Database,” the “Taiwan Social Change Survey,” and the Academia Sinica’s thematic studies on “The Happiness Thermometer Project” and “China Impact Survey” to enhance the validity and comparative foundation of this research.
It should be noted that Doublethink Lab’s “2024 Election Survey” was conducted using a dual landline and mobile phone survey method, whereas this survey was implemented through an online questionnaire. Both surveys applied post-stratification weighting based on the population structure published by the Ministry of the Interior to enhance the demographic representativeness of the samples. While the two surveys are comparable to a certain extent and can be used for trend observation and group comparisons, caution is advised when interpreting results due to potential differences in response behavior arising from different survey modes. These include social desirability bias, self-selection characteristics of online samples, and variations in questionnaire interaction. Therefore, potential mode effects (or differences caused by the survey method) should be carefully considered when interpreting the findings.
The successful advancement of this study relies heavily on the accumulation and pioneering efforts of prior research. We hereby express our sincere gratitude to all scholars and research teams who have long contributed to related fields. The raw data and codebook of this survey are also made openly available for use.
We hope that this study will provide valuable empirical data for future research on related issues and further deepen the understanding of both the general public and academic audiences regarding the impact of TikTok on public opinion.
Methodology
The survey covers all counties and cities in Taiwan, Kinmen County and Lienchiang County, targeting individuals aged 15 and above. The formal survey was conducted from March 10, 2025 (Monday) to March 20, 2025 (Thursday), with post-stratification weighting applied.
The survey method utilized an online questionnaire, implemented through the web version of the EZChoice mobile app, exclusively developed by DSIGroup. Additionally, eligible respondents were invited to complete the online questionnaire through online communities (including PTT Bulletin Board System, Dcard, Facebook groups, etc.). A total of 2,612 valid samples were collected, comprising 1,097 TikTok active users[1] and 1,515 inactive TikTok users[2]. After logical error checking, the raw data were subjected to sample representativeness testing based on the population statistics from the Department of Household Registration, Ministry of the Interior (as of the end of December 2024[3]). Multi-variable iterative weighting (raking) was applied to the valid samples based on gender, age, education level, and household registration region.
Research Data
This study aims to explore whether TikTok active users exhibit significant differences in media consumption habits, perspectives on domestic and public issues, cross-strait and international political sentiments, and social and emotional well-being. To effectively distinguish between active and inactive Tiktok/Douyin users, the questionnaire begins with two screening questions targeting respondents’ usage frequency and average session length over the past year. These responses form the basis for subsequent data classification and analysis.
This study operationally defines “TikTok active users” as individuals who, in the past year, have used TikTok with a frequency of “several times a week or more, with an average session duration of 30 minutes or more” or “several times a day, with an average session duration of 10 to 30 minutes.” All other respondents are classified as inactive TikTok users.
The statistical results presented in this data release report have been weighted based on the population structure published by the Department of Household Registration, Ministry of the Interior, to enhance inferential validity.
Media habits and internet addiction
First, through the screening questions, it can be observed that among TikTok active users, as defined by this project, the usage frequency over the past year is predominantly “several times a day” at 42.3%, followed by “several times a week” and “once a day” in descending order. Regarding single-session duration, 54.2% of users predominantly spend between 30 minutes and 1 hour per session. These findings are similar to the observations in the The Impact of TikTok on Taiwanese Youth and PRC perceptions: A 2025 TikTok User Study.
Regarding the social media platforms used daily by respondents, the analysis results show that both TikTok active and inactive users report similar usage rates for Facebook and YouTube, ranging from approximately 55% and 60%. This indicates that these two platforms have comparable popularity across both groups. However, chi-square tests reveal statistically significant differences between the groups in their use of Instagram, Threads, Dcard, PTT, Weibo, TikTok/Douyin, LINE, Facebook chat rooms, Xiaohongshu, and WeChat.
When analyzing the sources of information for political and public issues, it was found that the most frequently used information channels for TikTok active users are, in order, Facebook (39.4%), YouTube (34.8%), TikTok/Douyin (33.3%), and television (32.6%). Further comparison using chi-square tests indicates statistically significant differences between TikTok active and inactive users in their use of multiple information sources, including Facebook, Instagram, Threads, PTT, TikTok/Douyin, Facebook Messenger, Rednote/Xiaohongshu, Telegram, television, and print newspapers and magazines.
As expected, the proportion of TikTok active users obtaining political and public issue information through TikTok/Douyin is significantly higher than that of inactive users, with a gap of nearly 30 percentage points. Additionally, active users show significantly higher rates of obtaining such information from platforms like Facebook, Instagram, Threads, Rednote/Xiaohongshu, Telegram, and WeChat. In contrast, this group has a slightly lower usage rate for PTT, and their reliance on television as an information source is notably lower, with a gap exceeding 12 percentage points.
Both TikTok active and inactive users primarily rely on news media and news channels for perspectives on political and public issues, with over 65% of each group using these sources, slightly higher among inactive users. Further chi-square analysis reveals that, compared to inactive users, TikTok active users are significantly more likely to obtain perspectives from sources such as influencers/key opinion leaders (KOLs), fan pages on social media, specific political figures, and pundits or current affairs commentators. For example, in the case of influencers/KOLs or fan pages on social media, the proportion among active users is more than double that of inactive users, with a gap exceeding 18 percentage points, indicating that influencers and KOLs play a more critical role in how they get their information.
Regarding the choice of “most trusted sources of political and public issue viewpoints,” both groups predominantly select news media and news channels, with proportions exceeding 40%. However, TikTok active users exhibit significantly higher trust in influencers/KOLs or fan pages on social media platforms and specific political figures compared to inactive users.
In the questionnaire section related to internet addiction, TikTok active users show significantly more likely to respond with “somewhat agree” across all items, with differences between the two groups ranging between 12% and 22%. Chi-square tests indicate statistically significant differences in the overall response distribution between the two groups for this section. The results suggest that TikTok active users exhibit a relatively higher tendency toward addiction and dependence in their internet usage.
When asked about their views on the “ Main source of disinformation and cyber army in Taiwan,” respondents, regardless of whether they are TikTok active users, clearly identify the Democratic Progressive Party (DPP) as the source (36.5% for active users and 29.8% for inactive users). Among the next most commonly cited sources, active users name the Kuomintang (KMT) (13.4%) and the PRC (11.3%), while inactive users more often point to the PRC (18.5%) and the KMT (16.9%).
Overall, TikTok active users are significantly more likely to attribute misinformation and online trolls to the DPP compared to inactive users, with a gap exceeding 6.5 percentage points. Conversely, TikTok inactive users are significantly more likely to identify the PRC as the source, with a gap exceeding 7 percentage points.
Political orientation
The largest share of respondents expressed no particular support for or dislike of any specific political party. However, when asked which political party they support most, both active and inactive users chose the DPP most frequently, with support rates of 20.9% and 26.1%, respectively, though the support among active TikTok users was significantly lower. Support for KMT hovered around 20% in both groups, with no statistically significant difference between them. Notably, 11.6% of TikTok active users identified the Taiwan People’s Party (TPP) as their most supported party, compared to 6.4% among inactive users.
The DPP was the party most frequently identified as the least liked, with 35% of active TikTok users selecting it, nearly 6 percentage points more than inactive users. Conversely, TikTok active users were less likely to identify the KMT as their least liked party, with 21% selecting it compared to 29.4% of inactive users, a difference of over 8 percentage points. For the TPP, there was no significant difference between the two groups, with both around 6%.
Overall, approximately 60% of respondents, both active or inactive TikTok users, expressed some level of dislike toward the ruling party (DPP). Among them, TikTok active users are significantly more likely to say that they strongly dislike the DPP, with 27.2% compared to 19.4% of inactive users. In contrast, inactive TikTok users exhibit a milder disfavor, with 28.9% selecting “somewhat dislike”, 6.9 percentage points higher than TikTok active users.
When asked which Taiwanese political party is most likely to benefit from PRC’s escalating infiltration and united front strategies, respondents identified the Kuomintang (KMT) as the leading choice, with 33.6% of TikTok active users and 43.1% of inactive users selecting it — though the proportion among active users was significantly lower. Meanwhile, 28% of TikTok active users believe the DPP would benefit, a figure notably higher than the 17.8% among inactive users.
Public Perceptions of Narratives Across Various Topics
- Relevant to Taiwan democracy and domestic affairs
Democracy is still widely regarded as the best political system, with over 55% of respondents holding this view. A higher proportion of inactive TikTok users agree with this view (65.5%), compared to 59.5% of active TikTok users. Additionally, TikTok active users show slightly higher proportions in agreeing with the statements “In some situations, a dictatorial political system is better than democratic politics” and “For me, any political system makes no difference” compared to inactive users.
Regarding satisfaction with Taiwan’s current democratic system, slightly more than half of respondents expressed satisfaction, while the majority selected moderate opinions, such as “somewhat dissatisfied” or “somewhat satisfied.” Chi-square test analysis indicates no statistically significant difference between TikTok active and inactive users in their satisfaction with Taiwan’s democratic system.
Regarding the statement “Taiwanese society has a very serious issue of judicial injustice,” over 70% of respondents tend to agree (approximately 75% for the full sample). Despite differences in survey methods, this study shows a notable increase compared to the election survey data (68.1%) conducted by Doublethink Lab during 2024. In particular, active TikTok users are significantly more likely (80.1%) to agree with the statement compared to inactive users (73.9%), with the proportion selecting “strongly agree” being nearly 10 percentage points higher than that of inactive users.
Additionally, on the issue of “the Constitutional Court’s interpretations must align with the expectations of the broader Taiwanese public,” over 70% of respondents in both groups tend to agree. Active TikTok users show slightly higher proportions in the “somewhat agree” and “slightly disagree” options compared to inactive users, while inactive users have a slightly higher proportion in the “slightly agree” option.
Regarding the statement “The current ruling Democratic Progressive Party (DPP) is no different from the Chinese Communist Party, and Taiwan lacks freedom of speech,” approximately 60% of respondents expressed disagreement, a trend consistent with the results of the 2024 election survey. However, the proportion of respondents who tend to agree is notably higher among active TikTok users (46.9%) and inactive users (36.5%) than in the overall full-sample proportion for this question in 2024 (31.4%).
Specifically, in the current survey data, active TikTok users have significantly higher percentages in the “somewhat agree” and “strongly agree” options compared to inactive users, with a gap of approximately 4 percentage points. Conversely, inactive users have significantly higher proportions in the “slightly disagree” and “strongly disagree” options, exceeding active users by 5.3% and 7.2%, respectively. These differences reach statistical significance.
On the question of “for ordinary people, PRC is more suitable to live than Taiwan,” over 70% of respondents in both groups overall expressed disagreement. However, active Tiktok users have a significantly higher proportion tending to agree (26.1%) compared to inactive users (19.7%). In contrast, inactive users show a significantly higher proportion in the “somewhat disagree” option, with a gap of nearly 9 percentage points compared to active users.
On the proposition that “in cross-strait communication, Taiwan’s economic interests are more important than national security,” overall, more than 65% of respondents tend to disagree, indicating that the majority prioritize national security over economic interests. However, when excluding more neutral responses such as “slightly agree” and “slightly disagree,” the structure of opinion divergence becomes more pronounced.
Specifically, the combined proportion of active TikTok users selecting “somewhat agree” and “strongly agree” reaches 13.9%, significantly higher than the 8.3% among inactive users. Conversely, inactive users have a significantly higher proportion selecting “strongly disagree,” with a gap of 6.4 percentage points. These results suggest that, on the issue of balancing cross-strait economic cooperation with national security, TikTok active users are relatively more accepting of prioritizing economic interests.
On the statement “Chinese spouses are almost always tools of PRC’s united front strategy, and the Taiwanese government should strictly monitor them,” the majority of respondents overall tend to disagree, with 58.6% of active TikTok users and 56.5% of inactive users expressing disagreement. The percentage gap between the two groups is small, indicating relatively consistent views on this issue.
However, regarding the view that “The current government is the primary producer of misinformation/ disinformation in Taiwan” a clear divergence emerges. Chi-square tests reveal statistically significant differences in the response distributions between active and inactive TikTok users. Specifically, 53.4% of active TikTok users tend to agree with this statement, nearly 10% percentage points higher than the 44.3% of inactive users. This difference is particularly pronounced in the “strongly agree” option, where active users exceed inactive users by over 8 percentage points. Conversely, inactive users have a significantly higher proportion in the “somewhat disagree” option, exceeding active users by over 6 percentage points, indicating a relatively higher proportion holding a negative stance toward this statement.
- International relation and cross strait relations
Regarding the question of who is responsible for cross-strait tension, there is a clear difference in perception between active TikTok users and inactive users. Both groups blame the PRC government, but the proportional gap between them is significant. Among inactive TikTok users, 51.8% hold the PRC government responsible, compared to 38.9% of active users, a statistically significant difference of nearly 13 percentage points.
On the other hand, 35% of active TikTok users believe the DPP government is responsible, a proportion nearly as high as those blaming the PRC government, and significantly higher than the 29.4% among inactive users, indicating a stronger tendency to attribute responsibility to the ruling party. Additionally, the proportion of active TikTok users blaming the U.S. government also reaches statistical significance, exceeding inactive users by more than 5 percentage points.
Overall, active TikTok users tend to attribute responsibility for cross-strait tensions to the DPP, the U.S., and the PRC government, whereas inactive users are more likely to primarily hold the PRC government responsible.
This project also conducted in-depth interviews with teenage Taiwanese TikTok users to explore their perceptions of the PRC government and its citizens. Although respondents generally hold negative impressions of both, there are nuanced differences in the intensity and narrative context of these perceptions.
In the nationwide online survey component, differences in the distribution of perceptions between active and inactive users can also be observed. Regardless of whether they are TikTok active users, approximately 90% of respondents hold a negative attitude toward the PRC government, with responses predominantly concentrated in the “strongly dislike” option. Notably, the proportion of TikTok active users selecting “strongly dislike” (33.1%) is significantly lower than that of inactive users (38.9%), and they are also about 6 percentage points lower in the “somewhat dislike” option. Conversely, TikTok active users have a higher proportion selecting the milder “slightly dislike” option (38%).
In terms of perceptions of PRC citizens, both groups of respondents most frequently selected ‘slightly dislike’ (41% each), while approximately 13% in both groups selected ‘strongly dislike. Chi-square tests indicate that the differences between the two groups on this item do not reach statistical significance. Overall, while respondents’ perceptions of PRC citizens remain predominantly negative, they are relatively less intense compared to attitudes toward the PRC government.
Regarding U.S. Skepticism narratives, it is noteworthy that the response distribution in this study differs slightly from the 2024 election survey data conducted by Doublethink Lab. In the 2024 data, approximately 44% to 54% of respondents selected a disagreement option to questions reflecting skepticism toward the United States. In the current survey, however, the percentage of respondents selecting any of the disagreement options for the three U.S. skepticism narratives is around 40% or lower. The overall distribution pattern has shifted toward agreement.
Further examination of the distribution among different user groups reveals that, for both active and inactive TikTok users, the majority (over 30%) selected “slightly agree” across all three questions. Specifically, in relations to narratives commonly summarized as “the U.S. only wants to exploit Taiwan” and “the U.S. attempts to push Taiwan into war”, inactive TikTok show a significantly higher proportion of responses in the milder “slightly agree” option compared to active users. However, active TikTok users are significantly more likely to select the more extreme “strongly agree” option.
The largest gap of “strongly agree” is observed in the statement “if PRC were to use force to attack Taiwan, the U.S. will definitely not send assistance to Taiwan,” where TikTok active users and inactive users reported rates of 22.3% and 13.5%, respectively, representing a difference of 8.8 percentage points.
The survey results reveal significant differences between TikTok active and inactive users on issues related to the Taiwan Strait conflict and cross-strait unification. Specifically, on the question of “for the sake of cross-strait peace, it’s okay to give up the democratic system,” 30.3% of active TikTok users tend to agree, significantly higher than the 23.4% observed among inactive users. In contrast, inactive users have a significantly higher proportion in the “slightly disagree” option, exceeding active users by approximately 7.5 percentage points. Overall, more than 70% of respondents in both groups tend to disagree with the statement “it’s okay to give up the democratic system.”
Additionally, when asked about the statement “even if cross-strait unification occurs, it will not significantly change ordinary people’s lives”, the percentage of active TikTok users in the “tend to agree” range (45.2%) is significantly higher than that of inactive users (33.2%), a gap of approximately 12%. The primary differences lie in the “slightly agree” and “somewhat agree” options. In contrast, inactive users have significantly higher proportions in the stronger “strongly disagree” and “somewhat disagree” options, with an overall gap exceeding 11 percentage points.
When asked, “If PRC invades Taiwan militarily, would you be willing to fight to defend the country?” Although the distribution between active and inactive users shows significant differences, over 55% of respondents in both groups express willingness to fight for their country. Notably, active TikTok users also exhibit a slightly higher proportion selecting the stronger “very willing” response compared to inactive users.
Less surprisingly, on the question of “whether TikTok/Douyin should be completely banned due to national security concerns,” active and inactive TikTok users exhibit markedly contrasting distribution patterns. In the relatively milder response options, TikTok active users have a significantly higher proportion selecting “slightly disagree” compared to “slightly agree,” with a gap exceeding 9 percentage points. In contrast, inactive users predominantly select “slightly agree” as their main option, but this is only slightly higher than the next most common option, “slightly disagree,” by less than 3 percentage points. Overall, active TikTok users show a higher concentration in the “disagree” range, while inactive users display a more dispersed distribution of opinions and a more uncertain stance.
In response options reflecting stronger stances, the differences between the two groups are also statistically significant and more pronounced. Inactive TikTok users exceed active users by nearly 10 percentage points in selecting “strongly agree” to completely banning TikTok/Douyin due to national security concerns. Conversely, active users overwhelmingly surpass inactive users by approximately 16 percentage points in the combined “strongly disagree” and “somewhat disagree” options.
- PRC political propoganda
Regarding the identity and cross-strait unification narratives long promoted by the PRC government, although nearly 40% of both active and inactive TikTok users tend to disagree with the statement “Taiwanese people are part of the Chinese nation,” but also approximately 40% in both groups select “slightly agree.” Chi-square tests indicate that the response distributions for this question do not differ significantly between the two groups. In options reflecting stronger stances, the proportions for both “tend to agree” and “tend to disagree” are similar, each hovering around 20%. This result suggests that the issue of Taiwanese identity and subjectivity remains a critical topic requiring ongoing dialogue and collective construction in contemporary society.
As for statements like “both sides of the Strait are one family, and regardless of the form, unification is inevitable in the foreseeable future,” although active TikTok users have a significantly higher proportion across all levels of agreement, exceeding inactive users by more than 10 percentage points in total, both groups overwhelmingly selected disagreement options (69.7% for active users and 79.7% for inactive users). Notably, in the strongly opposing “strongly disagree” option, both groups exceed 25%. This result reflects that, regardless of whether they are active TikTok users, Taiwanese people generally maintain a highly reserved and negative attitude toward the official narrative of “inevitable cross-strait unification.”
Regarding the common external narrative about Xinjiang that “the local people are happy and content, and the Chinese government has made significant efforts to preserve local culture,” over 65% to nearly 70% of respondents tend to disagree with this statement. Further examination of the options reflecting stronger stances (“somewhat disagree” and “strongly disagree”) shows that the proportion of active TikTok users is slightly lower than that of inactive users, indicating that, while both groups predominantly reject the narrative, the intensity of opposition among active users is relatively lower.
Life satisfaction and happiness level
This project also aims to examine differences in life satisfaction and mental well-being between active and inactive TikTok users, based on responses to four related questions. Regarding the level of happiness in life, both groups of respondents predominantly selected the option “neutral”, with proportions exceeding 40% and approaching 50%. However, inactive users have a significantly higher proportion in this option, while TikTok active users show a higher proportion in the “strongly unhappy” option, with a gap of approximately 5 percentage points. In terms of life satisfaction, both groups similarly have a majority selecting “neutral”, with about 40% leaning toward satisfaction.
Regarding satisfaction with social life, both groups’ responses are primarily concentrated in the “neutral”, with about 40% leaning toward satisfaction. Notably, while approximately 40% of respondents indicated they are “not lonely,” Active TikTok users are significantly more likely to report some degree of loneliness, with a gap exceeding 9 percentage points compared to inactive users.
Regarding the social trust indicator, over 50% of respondents in both groups tend to believe that “you usually need to be cautious in dealings with others,” and approximately 15% indicate that “you almost always need to be very careful when interacting with others.” Chi-square test analysis shows that the distribution of responses between active and inactive TikTok users on this issue does not reach statistical differences.
This study also seeks to compare the subjective evaluations of active and inactive TikTok users regarding economic and lifestyle-related issues. The overall satisfaction with the current economic situation is similar between the two groups, with approximately 40% of respondents in both groups selecting “average,” followed by “dissatisfied” and “satisfied.” TikTok active users have slightly higher proportions for both “satisfied” and “dissatisfied,” while inactive users are more concentrated in the neutral “average” evaluation.
Regarding perceptions of fairness in personal living standards and individual effort, the majority of respondents (nearly 50%) consider it “fair enough,” with little difference between the two groups. However, active TikTok users have slightly higher proportions in the “somewhat unfair” and “very unfair” options, and the distribution between the two groups shows statistically significant differences based on chi-square test analysis.
Concerning perceptions of relative deprivation, the majority in both groups select “about the same,” with proportions around 50%, and inactive users have a slightly higher proportion by approximately 3 percentage points. Additionally, the percentages for other options are similar between active and inactive TikTok users, and chi-square tests indicate no statistically significant differences.
Demographics Data
Initial Conclusion
This section draws general conclusions based on cross-comparisons and relevant statistical tests using weighted data from active and inactive TikTok users. Additional cross-tabulations, correlation analyses, counterfactual analyses, and latent class analyses will be presented in the subsequent comprehensive analysis report.
The survey data indicate that TikTok active users differ significantly from inactive users in the sources they rely on for political and public affairs information. Active users are more inclined to obtain perspectives from influencers/key opinion leaders (KOLs), fan pages, and specific political figures on social media platforms, with KOLs and fan pages having the most pronounced influence. The proportion of active users relying on them is more than double that of inactive users. Moreover, their trust in these sources is significantly higher, highlighting the critical role of influencers/KOLs in shaping their information environment and perspectives.
Regardless of TikTok usage, the DPP is the most supported party among respondents, while also nearly tying with the KMT as the “most disliked” party. Notably, TikTok active users show significantly lower support for the DPP, while their support for the TPP is significantly higher compared to inactive users.
On cross-strait issues, clear divergences emerge between active and inactive users. Although most respondents attribute the primary responsibility for cross-strait tensions to the PRC government, TikTok active users are more likely to blame the DPP and the U.S. government, with the proportion blaming the DPP being nearly as high as that for the PRC government. Regarding the PRC’s united front strategies toward Taiwan, both groups believe the KMT is most likely to benefit, though active users are slightly less likely to hold this view. When asked about willingness to fight to defend the country, over half of both groups express willingness, reflecting a fundamental consensus on defense.
Democracy is still regarded as the best political system, but the proportion among active users is slightly lower, with a higher proportion expressing openness to authoritarianism or indifference to the type of system. Most respondents have a somewhat positive evaluation of Taiwan’s current democratic situation. However, when asked whether Taiwanese society faces severe judicial injustice, active users show significantly higher agreement, particularly in the “strongly agree” option, exceeding inactive users by nearly 10 percentage points.
Regarding strongly critical narratives about the current government, such as “the current government is no different from the Communist Party” or “Taiwan lacks freedom of speech,” although most respondents disagree, the proportion agreeing has slightly increased compared to past surveys. On comparisons with life in the PRC, both groups tend to believe Taiwan is more suitable to live in, but in weighing economic versus national security issues, active users are more inclined to prioritize economic interests.
Regarding skepticism toward the United States, both groups show subtle shifts in attitudes compared to the 2024 election survey. Although disagreement remains the majority, the overall opinion distribution has shifted closer to “tend to agree”, particularly on the statement “the U.S. will not assist Taiwan,” where active users’ proportion of strong agreement is significantly higher than that of inactive users. These shifts in attitudes warrant further analysis or additional research for deeper exploration.
Finally, regarding views on cross-strait unification and lifestyle changes, while most respondents oppose the idea of abandoning democracy, active TikTok users are significantly more likely to accept the notion that “unification would not significantly change lifestyles.” On official narratives like “cross-strait unification is inevitable,” Taiwanese respondents, regardless of TikTok usage, overwhelmingly remain highly skeptical and negative. In summary, these results reveal the distinct profile of TikTok active users in terms of information acquisition, trust structures, and political attitudes, with significant differences from inactive users.
Limitations
Sampling Limitations
This survey was conducted using an online questionnaire. To ensure data quality and validity, in addition to the preliminary error-checking mechanisms built into the EZChoice system, the research team designed a logic verification table (covering logical consistency and relational checks) based on the survey content and performed multiple logical validations on the data. Furthermore, the survey data were weighted and adjusted based on the latest demographic statistics from the Ministry of the Interior, using variables such as respondents’ gender, age group, education level, and geographic region to enhance the sample’s representativeness of the population. However, as this survey was conducted online, certain sampling limitations may still exist. For instance, respondents’ internet usage habits may influence their willingness to participate, or certain demographic groups may be underrepresented in the online sample.
Sensitivity of Public Political issues, Cross-Strait Issues, and International Relations Perspectives
Some questions in this questionnaire address public political issues, cross-strait issues, and international relations, topics that are highly polarized in Taiwanese society, where differing political ideologies may lead respondents to hold sharply divergent views. In this context, some respondents may feel distrust toward the questionnaire content or the survey itself. Although this online survey was commissioned to a politically neutral polling company to mitigate this effect, such sensitivities may still pose potential challenges to the survey results.
Prejudices and Social Desirability Bias Among TikTok/Douyin Users
At the outset of the questionnaire, respondents were screened based on their weekly usage frequency and single-session duration of TikTok/Douyin to distinguish between active and inactive users. This design may make respondents aware of the survey’s targeted focus, potentially prompting them to associate it with negative societal perceptions of TikTok. In this scenario, some respondents may adjust their responses to avoid being labeled as “pro-PRC” or “irrational,” leading to social desirability bias. Such bias may affect the authenticity of the data, particularly for questions involving political attitudes or values.
Limitations in Comparisons with Previous Projects
Although this report references previous projects for comparative analysis in some results, it should be noted that the 2024 election survey employed a mixed-mode approach combining landline and mobile phone interviews. While that survey was also weighted according to the Ministry of the Interior’s demographic structure, its methodology differs from the online questionnaire format used in the 2025 TikTok user survey. Therefore, comparisons between the two in this report are only suitable for observing distribution patterns and overall trends, and direct comparisons of specific percentages or detailed statistical data are not feasible.
Acknowledgments
This project expresses special gratitude to Dr. Zonghong Lin, a full-time research fellow at the Institute of Sociology, Academia Sinica, for providing numerous suggestions and insights during the research planning stage regarding questionnaire design, research logic, and subsequent analysis. We also thank Yun-ju Chen and research assistant Chun-ying Li from the Institute of Sociology, Academia Sinica, for their significant assistance in questionnaire design and statistical analysis of the online survey data.
Additionally, we extend our gratitude to the Public Opinion Program team at the Academia Sinica Research Center for Humanities and Social Sciences and the China Impact Survey team for providing valuable data from the Well-Being Survey and telephone surveys, respectively, which served as references for this project’s questionnaire. The research design of this project also partly drew upon the Taiwan Communication Survey Database and the Taiwan Social Change Survey. We express our sincere respect and gratitude to the many scholars and teams who have contributed to these fields in the past.
Furthermore, we thank DSIGroup Co., Ltd., for conducting the online survey on behalf of Doublethink Lab in accordance with the planned specifications. We also appreciate the friends who provided assistance and suggestions for this project; without everyone’s support, this survey would not have been possible. All matters arising from this project and its report are the sole responsibility of Doublethink Lab.
Footnotes
- Definition of TikTok Active Users: Individuals who, in the past year, have used TikTok with a frequency of “several times a week or more with an average session duration of 30 minutes or more” or with a frequency of “several times a day with an average session duration of 10–30 minutes” are classified as TikTok active users.
- Definition of TikTok inactive Users: Individuals who do not meet the criteria of using TikTok in the past year with a frequency of “several times a week or more with an average session duration of 30 minutes or more” or with a frequency of “several times a day with an average session duration of 10–30 minutes” are classified as TikTok inactive users.
- The data is sourced from the population statistics of individuals aged 15 and above as of the end of December 2024, provided by the Department of Household Registration, Ministry of the Interior. https://www.ris.gov.tw/app/portal/346
Contact us
- Doublethink Lab: [email protected]
- Social Engagement Researcher- Eric Hsu: [email protected]
Attachments
◎ 2025 TikTok 抖音使用者調查＿全國網路調查 原始數據/Raw data
◎ 2025 TikTok 抖音使用者調查＿全國網路調查 譯碼簿/ Code book