Research on the Influence of Media Use on Government Trust -- Empirical Analysis based on CGSS2010 Data

. The development of science and technology has promoted the diversification of media forms, and the emergence of new media has enriched people’s access to information. At the same time, this situation has also led to the rapid spread of some negative news and gradually dispelled people’s political trust. In order to explore the specific relationship between media use and government trust, it is necessary to conduct more in-depth research. Based on the CGSS2010 survey data, this paper focuses on the influence of media use on political trust and explores the relationship between media use and government trust in detail by using regression analysis, and factor analysis. It is found that the use of traditional media has a positive and significant impact on political trust, while the use of new media has just the opposite effect on it in general.


Introduction
The development of science and technology has brought about changes in communication methods and entertainment channels, and to a certain extent, it has also changed the way people acquire information. In the past, people could only get information through newspapers, radio, television, and other traditional ways, with low timeliness and poor interaction. Nowadays, thanks to technological innovation, new media has spread to thousands of households, and the speed at which people get news and express their personal views is far faster than in the past. At the same time, with the continuous improvement of people's awareness of right protection and supervision, people pay more attention to the projection of administrative organs. This series of changes could have helped the government to improve its administrative ability but led to unexpected adverse effects. As a matter of fact, the public space for expressing opinions has become larger, and their awareness of exercising supervisory power has risen. However, the news of ineffective government administration has been exposed one after another. Therefore, people's voices of questioning will gradually appear, and people's trust in the government will also show a downward trend.
As early as the last century, some scholars suggested that news reports in the media would affect public trust in the government . Although our government has always attached great importance to the management of news media and actively participated in the construction of citizens' political trust, it has to be acknowledged that the public trust in the government in China is in a declining trend . In particular, the case of "The Mother of Eight Children" in Xuzhou, which happened during the Spring Festival in 2022, made us realize the huge negative impact on the credibility of local governments after information was spread on new media. If citizens' political trust continues to decline, the government may no longer have the authority in management, and the social order will be greatly impacted. Therefore, in the grand context of the development of science and technology, exploring the influence of media use on citizens' political trust has multiple practical significance. The government can not only take this opportunity to strengthen its administrative ability and improve its image among the people, but also find out the difference in the influence of different media forms on the trust between local government and central government, so as to better guide the development of media, and achieve a benign closed loop in which the feedback of the masses and the improvement of government administration are coordinated.
This study will discuss the relationship between media use and political trust. It is believed that the use of different types of media will have different effects on people's political trust. Based on the understanding and analysis of political trust, this study uses the option settings in the CGSS2010 survey to subdivide it into political trust for the central government and local governments. At the same time, it will also generalize the scope of political trust, including people's trust in judicial departments, public security departments, central media, and local media, thus expanding the extension of research. As for the research design, it will also make a factor analysis of media use, and extract two factors, that is, new media use and traditional media use, so as to further explore the influence of two different media ways on political trust.

Literature Review
At present, the research on political trust is mainly divided into three categories, which includes history and culture, political system, and social capital.
Originated from the cultural foundation, religious beliefs, and social concepts established in the early stage, the historical and cultural perspective holds that political trust has exogenous characteristics . It is not difficult to understand that trust is not only a social consensus but also has distinct cultural characteristics . Studies by domestic scholars showed that Chinese traditional political culture and Confucianism have played a role in maintaining national political trust [5,6]. Meanwhile, the comparative study of different value cultural orientations also confirmed this point. In a comparative study of the Chinese mainland and Taiwan province, scholars found that the government trust in Taiwan province is more obviously influenced by the government's administrative performance, while the people's trust in the mainland government is more constrained by the traditional cultural values . The research of foreign scholars has similar findings. Using the data of the "World Values Survey", scholars found that in different countries, people's political trust will be significantly affected by differences in religious beliefs . Simultaneously, Fukuyama's research also pointed out that the degree of trust of national people in the government is largely influenced by national culture .
As mentioned above, the cultural perspective focuses on the external section of political trust. This leads to a question, of whether we can return the perspective to the inside of government trust and analyze its internal characteristics. In fact, the same country will show different levels of trust at different times, even in different regions of the same time. This is the research content from the institutional perspective. Institution refers to a social structure that regulates individual actions by rules or operation modes, and it is a kind of purposeful existence. At the same time, the institution also carries the current people's value judgment, thus standardizing and influencing people's behavior within the institution. According to the institutional theory, the degree of government trust depends on people's feelings about government performance . When the government's performance matches the public expectation of the government, the public trust in the government will rise, or vice versa. The government's behaviors, work achievements, and abilities to provide services are usually regarded as the government's performance. Undoubtedly, the performance of the government will affect public trust in the government. Research shows that public satisfaction with government affairs has a significant impact on political trust. That is to say, the higher public satisfaction with the government's achievements (government work and transparency, administration according to law, improvement of the gap between the rich and the poor, etc.), the higher their trust in the government .
Another explanation of political trust is the social capital theory. At present, mainstream studies all hold that social capital includes information communication, trust, and public participation. In groups, interpersonal trust can promote political trust . In an organization, the more contacts the participants have, the more comprehensive the information they have and the deeper they trust in each other. Similarly, government trust is also transmitted like that, that is, when there are people in the government who they know well and trust, they tend to show their trust in the government. The microinterpersonal relationship is transformed into macro-political trust, which has a far-reaching impact on the process of government administration.
In a word, the cultural perspective focuses on the shaping of public cognition by the political culture in the process of socialization, while the institutional theory pays attention to the principle of the influence of public performance perception on political trust and the social capital theory reflects the great role of micro interpersonal trust in transferring to political trust. In the complex background and social form, the three theories can be compatible with each other to explain a specific phenomenon of political trust.
However, none of the above three theories pays attention to the influence of the use of information media on political trust. Nowadays, the mobile Internet has become the general trend in the world, and various forms of Internet technologies have penetrated into people's life and work, forming an influential force that cannot be ignored.
First of all, in terms of the cultural theory, people mainly form their own subjective understanding of culture through school study and perception of social information. Nowadays, together with the mainstream traditional media like newspapers, new media, such as Sina Weibo and QQ, have powerful information transmission functions, which have played a very important role in shaping people's cultural values. In addition, from the perspective of institutional theory, the degree of government trust depends on people's feelings about government performance, and people have two main ways to perceive government performance. One is to judge the administrative effect of the government directly through one's own feelings in life. The other is to obtain information through various media channels and make judgments based on one's own actual feelings. Moreover, regarding the social capital theory, the information communication part included is mostly based on interpersonal communication, ignoring the massive information provided by the use of media. Especially in the Internet age at present, the information flow brought by the employment of media is far beyond the reach of interpersonal communication.
In recent decades, there have been three mainstream theoretical views on the influence of media use and government trust in western academia. The first view is the theory of media depression, which holds that the use of media will lead to the decline of people's trust in the government, and the media reports negative news about the government, resulting in people's distrust and negative evaluation of the government . Contrary to the depression theory, the second one holds that by using various media, the government can conduct political propaganda to the people and increase their understanding of the government, so as to promote their participation in politics, and eventually achieve the goal of maintaining and improving people's political trust . Different from the views aforementioned, the third one is in an intermediate state, holding that the media is only a means of transmitting information, and what really affects political trust is the content diffused by the media.
With respect to the relationship between media use and political trust, domestic scholars' research has relatively consistent results. Lu and Quan (2015) found that the traditional media, including newspapers, magazines, and television, can promote the trust of the central and local governments after analyzing the data of CGSS2010. However, the new media with the Internet as the carrier has a negative correlation with the political trust of the central and local governments. Using the same data, Hu and Zhuang (2011) also found the same result, that is, the use of new media will significantly reduce people's political trust in the government, but traditional media can enhance residents' political trust. In addition, using the national public survey data from 2010 to 2011, Zhang et al. (2014) found that the traditional media has a weak role in enhancing people's political trust, but the new media can squeeze the spread of traditional media, that is, the more people use the Internet to understand or participate in public practice, the lower their level of political trust.
To sum up, the use of new media such as the Internet is not conducive to the promotion of political trust, while the use of traditional media like newspapers and television has a certain promotion effect on political trust. However, there are some shortcomings in the existing research, mainly in that the division of political trust is unified and solidified, only focusing on the central government and local governments, but ignoring other important government functional departments. In addition, in the analysis of new media and traditional media, there are many variables to investigate, and the problem of multicollinearity is likely to occur, which will affect the regression results and lead to an unreliable conclusion. In view of the above problems, this paper intends to further subdivide the investigation factor of political trust, expand the scope of political trust to the courts and judicial systems, public security departments, central media, and local media, and investigate the relationship between media use and broad political trust in an all-round and multi-field way. In the meantime, the use of media is divided into traditional media and new media by factor analysis, which avoids multicollinearity among variables and improves the reliability of regression analysis results.

Research Hypotheses
As a traditional way of propaganda, TV, newspapers, radio and other traditional media are actually subject to multiple reviews by the government press and publication administration and the propaganda department of the party committee, which has the effect of mainstreaming in a connotative sense. Its content is presented after being screened and filtered , which can strengthen the publicity effect of positive news, weaken the influence of negative news, and help people have a positive impression of the government. Therefore, the greater and higher the frequency of people's contact with traditional media, the more opportunities they have to receive positive signals from relevant government departments. And the stronger their perception of government performance, the higher their political trust in the government. Based on the above analysis, this paper puts forward the following research hypotheses: H1a: There is a significantly positive relationship between the use of traditional media and the trust of the central government.
H1b: There is a significantly positive relationship between the use of traditional media and the trust of local government (rural refers to township government).
H1c: There is a significantly positive relationship between the use of traditional media and the trust of courts and judicial systems.
H1d: There is a significantly positive relationship between the use of traditional media and the trust of public security departments.
H1e: There is a significantly positive relationship between traditional media use and central media trust.
H1f: There is a significantly positive relationship between traditional media use and local media trust.
Compared with the traditional media's "party-controlled media" feature, the new media has higher communication efficiency and lower monitoring degree, which is a low-threshold information communication channel. In the context of new media, the sources, themes, and contents of information are more extensive, and the speed of information diffusion is faster. The development of new media enables all kinds of information to spread earlier than traditional media, which leads people to form a mindset first. Meanwhile, due to the transparency of new media, all kinds of negative news from the government are more difficult to cover up, which can cause the formation of low political trust among the people. In addition, owing to the push algorithm characteristics of new media, some netizens and unscrupulous media make up false news about the government publicly on the Internet, so as to achieve the purpose of wide dissemination and profit. However, the government's rumors are often delayed, which brings people's difficulty in receiving them, thus eroding the trust in the government. This is not conducive to the maintenance and promotion of political trust. Based on the above analysis, this paper puts forward the following research hypotheses: H2a: There is a significantly negative relationship between the use of new media and the trust of the central government.
H2b: There is a significantly negative relationship between the use of new media and the trust of local government (rural refers to township government).
H2c: There is a significantly negative relationship between the use of new media and the trust of courts and judicial systems.
H2d: There is a significantly negative relationship between the use of new media and the trust of public security departments.

EMEHSS 2022
Volume 25 (2022) 409 H2e: There is a significantly negative relationship between the use of new media and the trust of central media.
H2f: There is a significantly negative relationship between the use of new media and local media trust.

Data Sources
In this paper, a total of 11,826 valid questionnaires from the Chinese General Social Survey (2010) are used as data sources. The Chinese General Social Survey (CGSS, henceforth) project was started in 2003 and jointly implemented by the Renmin University of China and academic institutions all over the country. It is the first national, comprehensive, and continuous large-scale social academic survey project in China. CGSS regularly collects multi-level data from society and individuals. Its purpose is to summarize the long-term trend of social changes, explore social topics of great theoretical and practical significance, provide data for government work planning, and promote the opening and sharing of domestic social science research. Because the sampling of CGSS is strict and scientific, the data is representative, which can truthfully reflect the situation of the respondents. Therefore, CGSS data has become the most important data source to analyze Chinese society at present, which has been widely applied in research.

Variable Measurement
This paper takes political trust as the dependent variable, media use as the independent variable, and age, gender, income, education level, and religious belief as the control variables. This study adopts the OLS regression model to test the research hypothesis. The measurement and operation process are as follows.

Dependent Variable: Political Trust
Because CGSS2010 directly investigated the political trust degree of respondents, this paper directly selected the relevant data in CGSS2010 for analysis. In the survey data, D301, D302, D303, D305, D306, and D307 represent the degree of political trust (how do you trust the following institutions? ) D301 represents court and judicial system; D302 represents central government; D303 represents local government (rural refers to township government), D305 represents public security department, D306 represents central media, D307 represents local media. And the corresponding options are, 1-totally untrustworthy, 2-comparatively untrustworthy, 3-somewhere between trustworthy and untrustworthy, 4-comparatively trustworthy, and 5-totally trustworthy. Because the positive and negative terms of the scale have the same semantic meaning, and out of the convenience of the analysis, this study directly selects the factor score of CGSS for regression analysis.

Independent Variable: Media Use
The independent variable of this paper is media use, and according to the data results in CGSS2010, some indexes are selected and factor analysis is carried out. The selected item in the CGSS2010 questionnaire is A28, that is, "in the past year, your use of the following media are newspapers, radio, television, the Internet (including mobile phone surfing the Internet) and customized messages on mobile phones." The corresponding options are, 1-never, 2-rarely, 3-sometimes, 4-often, 5-always.
In order to reduce the multicollinearity among independent variables, this paper uses factor analysis to analyze the five variables used by media (see Table 1). The KMO measured by factor analysis is 0.618, and Bartlett's spherical test is significantly correlated at the level of 0.000, which indicates that the factor analysis logic is self-consistent. This factor analysis divides these five variables into two categories, namely, new media use and traditional media use. Among them, the use of traditional media includes the use of newspapers, radio and television, while the use of new media includes the use of the Internet and customized messages on mobile phones.

Control Variables
Considering that the trust of the government is easily disturbed, this paper specially controls the indicators such as gender, age, income, political outlook, household registration type, nationality, marital status, education level, religious belief, and social trust. In order to keep the simplicity and comprehensibility of the analysis results, some indexes are processed.
With respect to the variable of age, this study uses the number 2010 minus the year of birth for processing. For education level, 0 represents no education at all, 3 represents literacy class, 6 represents elementary school, 9 represents junior high school, 13 represents vocational high school, 12 represents ordinary high school, 13 represents technical secondary school, 14 represents college adult higher education, 15 represents college formal higher education, 16 represents undergraduate adult higher education, 17 represents undergraduate formal higher education, and 20 represent postgraduate education or above. For income, the number of the actual data divided by 10000 is used for processing. Simultaneously, we take gender (0 = female), political outlook (0 = non-Communist Party of China member), household registration type (0 = rural household registration), nationality (0 = minority), marital status (0 = not married) and religious belief (0 = no religious) as binary variables to facilitate the analysis.
At the same time, in order to reduce the multicollinearity among the control variables, this paper uses the factor analysis method to analyze the five indicators of social trust in the control variables (as shown in Table 2). The KMO measured by factor analysis is 0.839, and Bartlett spherical test is significantly correlated at the level of 0.000, which indicates that the factor analysis logic is selfconsistent.

Descriptive Statistics
Descriptive statistics of all variables designed in this study are shown in Table 3. It can be seen that in the selected samples, the number of males and females is basically the same. The subjects of the survey are mainly middle-aged people, with the youngest participant being 17 years old and the BCP Business & Management

EMEHSS 2022
Volume 25 (2022) 411 oldest being 94 years old. The average education level is junior high school. From the descriptive analysis, among the indicators to measure political trust, the average value of the central government trust is 4.39, which is significantly higher than the other four indicators, indicating that the respondents' trust in the central government is generally high, and it is greater than their trust in local governments, courts, and public security organs. The trust of local governments is generally low, ranking the last among the five indicators.

Regression result 4.1 Regression Analysis of the Influence of Traditional Media Use on Political Trust
In order to verify the research hypotheses H1a to H1f, the study establishes a regression model and analyze it. The results are shown in Table 4.
The analysis shows that the use of traditional media has a significantly positive relationship with political trust in general. In the regression data, it can be found that the influence of traditional media on the central and local governments is stronger than that on the central and local media. This is because the masses pay more attention to social news when using traditional media to obtain information, which is easy to generate positive perception of government performance. Meanwhile, social news also occupies the main page of traditional media, making it easier to find and read. However, for the judicial system and public security departments, the traditional media does not have a significant impact. This is mainly due to the fact that in the traditional media, there are few reports about judicial collections and public security departments, and the layout in them is scattered and small, which cannot better stimulate people's sense of political efficacy. Therefore, it is assumed that H1a, H1b, H1e, and H1f are confirmed.   Note: * represents p<0.1, ** represents p<0.05, *** represents p<0.01, and the number in brackets is the value of t.

Regression Analysis of the Influence of the Use of New Media on Political Trust
In order to verify the research hypotheses H2a to H2f, we establish a regression model to analyze. The results are shown in Table 5.
On the whole, the use of new media has a significantly negative relationship with political trust, while he uses of new media has a significantly negative correlation with the central government, judicial departments, public security departments, central media, and local media, which just confirms our previous analysis. Due to the timeliness, transparency, intelligent push, and lack of effective supervision of new media, negative news will erode people's sense of political efficacy, thus reducing people's political trust. However, the use of new media has no significantly negative impact on the trust of local governments. This may be because local negative news cannot always reach the people. Facts have also proved that, according to the big data algorithm, news with high attention is widely circulated in new media, while some negative news about local governments is gradually forgotten because they do not get enough attention, which will not have a fatal blow to the political trust of local governments. Therefore, hypothesis H2a, H2c, H2d, H2e, and H2f are confirmed.  Note: * represents p<0.1, ** represents p<0.05, *** represents p<0.01, and the number in brackets is the value of t.

Conclusion and Discussion
This study uses the methods of factor analysis and regression analysis to investigate the influence of media use on political trust. First of all, the use of traditional media has a significantly positive impact on the central government, local governments, central media, and local media. Among them, the influence of the central and local governments is more significant than that of the central and local media. Simultaneously, the influence of traditional media on judicial departments and public security departments is not significant. Secondly, the use of new media has a significantly negative impact on the central government, judicial departments, public security departments, central media, and local media. Besides, the impact of new media on local governments is not significant.
The development of science and technology has brought great changes to our lives, and to some extent, it has changed the way we get information. At present, when the new media popularizes, access to information will undoubtedly be more greatly affected. However, there is no definite answer to the question of whether new media will further erode political trust. As long as the relevant laws and regulations are implemented and the appropriate supervision mechanism is introduced, it can provide a benign development space for the new media. As a platform for rapid information communication, new media provides an excellent platform for information communication between the people and the government. Nowadays, governments at all levels have successively launched the feedback function on new media platforms. In the short run, the use of new media will indeed damage the trust of the government. But in the long run, as long as it is used properly, the use of new media can promote public feedback and government improvement to achieve a benign closed loop.