Is There a Role of Religion? The Moderation Role of Religious Identity and Religious Practice between Traditional Media Usage and Moral Evaluation

: Although the relationship between traditional media usage and moral evaluation has been studied in China, it is not clear what role religion plays in this relationship. The 2013 Chinese General Social Survey was used to examine the moderation role of religious identity and religious practice in this correlation. The STATA 15.1 and PROCESS macro for SPSS (Model 2) was employed. This research conﬁrms that religion has a moderating role in the correlation between traditional media usage and moral evaluation. Speciﬁcally, religious identity, no matter whether it is polytheistic or monotheistic, will strengthen the correlation between traditional media usage and moral evaluation. However, religious practice will weaken the correlation between traditional media usage and moral evaluation, except the religious practice of monotheism in China. Furthermore, our ﬁndings prove that religion is an important situational factor in the correlation between traditional media usage and moral evaluation. We should take religious identity and religious practice as independent factors to conduct a richer study in the future. Most importantly, our ﬁndings further conﬁrm that the rationalization of society does not necessarily lead to the secularization of religion.


Introduction
Since China's reform and opening up, under the influence of the socialist market economy, social morality has changed a lot. Whether it is the moral "uphill theory", "landslide theory", or "stagnation theory", it has caused a lot of discussions, among which the core issues are related to the public's moral evaluation towards society (Li 2017). The definition of moral evaluation in academic circles is relatively unified, which is the process that an individual gives moral value to something or someone according to his own moral standards, and the process of expressing his moral opinion and selfconsciousness by evaluating other people's moral behavior (Greene et al. 2001;Szekely and Miu 2015). It can be seen that there is no simple unified standard for moral evaluation (Wright and Baril 2011). Since moral evaluation is an irrational process, individuals always find suitable explanations and reasons for their moral judgments (Haidt 2001). Although moral evaluation is the embodiment of an individual's thinking and values, the moral evaluation towards society is not closely related to the individual itself most of the time, which makes us wonder whether external factors affect moral evaluation.
Many studies have shown that in addition to individual self-awareness, social context factors are also important sources of variation that affect individual moral judgment and evaluation (Pizarro 2010;Sinnott-Armstrong 2006;Wynn 1990;Zarinpoush et al. 2000). Relevant scholars put forward the concept of "mediazation", emphasizing the shaping role of media as a social context factor in social culture (Hjarvard 2013). As an important part of social culture, the relationship between the moral field and media has also been widely studied. In recent years, the role of the media in the field of ethics has been gradually recognized. Some studies have pointed out that the moral content in media reports can Religions 2021, 12, 137 2 of 17 attract people's attention, that is to say, among moral vocabulary and non-moral vocabulary, people will pay attention to moral vocabulary first (Brady et al. 2020). Chinese scholars have defined traditional media and non-traditional media in China today. Non-traditional media is a communication mode based on digital technology and network technology, which integrates various new media technologies. Traditional media are mainly electronic media such as television, radio, film, and printed media such as newspapers, magazines, and books. Traditional media has unique advantages in the aspects of public cognition, reputation, brand, and authority (Ling and Deng 2015). With the rapid development of non-traditional media, traditional media has become gradually marginalized in people's lives. According to the data of the annual press and publication industry analysis report of the State Administration of press, publication, radio, film, and television of China, the publishing scale of traditional media such as newspapers and periodicals reached its peak in 2013 and then began to decline gradually. However, as a kind of formal and informal control institution (Hall 2005), traditional media has the authority that any non-traditional media cannot replace (Xiao and Han 2019). In daily life, people are always willing to rely on television and other authoritative traditional media reports to evaluate social and public affairs. The interpretation and transmission of information by the media has guidance, which will greatly affect people's concept and judgment of things (Coleman 1994;Misra et al. 2003). Since the first day of the broadcast, traditional media is the channel for the government to disseminate information, carrying the important role of policy propaganda and economic mobilization, having a positive impact on the people, making their moral values relatively unified (Scheufele 2002). It can be seen that the nature and function of traditional media determine that the content it spreads is always positive. When the society is faced with crisis events and the network public opinion is out of control, the traditional media will actively intervene to play the role of maintaining stability and deliver positive energy to the public, and reduce the social dissatisfaction caused by crisis events (Lin 2013). Some studies have pointed out that when people perceive social justice, life satisfaction is high, and the overall evaluation of society will also be improved (Shi and Yu 2006). Chinese scholar Zeng (2019) found that there was a positive correlation between the frequency of traditional media usage and media trust, and media trust was the key factor to explain media usage and satisfaction (Katz and Gurevitch 1973). German scholars Mundorf et al. (1994) through a survey of German families, found that the more frequent the use of traditional media such as television, the higher people's evaluation of the quality of life. Therefore, the use of traditional media will improve people's life satisfaction and evaluation of the quality of social life, which is also one of the tasks of moral evaluation. That is to judge the actions, intentions, good and evil of the whole society or social managers (Han 2018). It can be seen that the use of traditional media will improve people's life satisfaction, and then enhance their moral evaluation towards the whole society. Therefore, we propose: Hypothesis 1 (H1). The use of traditional media will have a positive impact on people's moral evaluation towards society.
However, in the macro similar media environment, the individual's subjective moral evaluation of society is not the same. What kind of mechanism is playing a role in this needs to be further explored. As a powerful social force (McCullough and Willoughby 2009;Mónico and Alferes 2018), religion plays an important role in maintaining social order and ensuring social survival (Nanda 1991). Therefore, the role of religion between traditional media and moral evaluation is worth studying.
Religion is everywhere in people's lives. Durkheim once said that religion, as the initial common consciousness, contains all the performance and practice in society, and then each function gradually differentiates (Durkheim 2000). Religion is related to life satisfaction, prosocial behavior, moral behavior, health behavior, and risk perception (Buck et al. 2009;Charles 2001;Hoover and Coats 2011;Mónico et al. 2016;Mónico 2012;Sun and Zhang 2019;Zeng et al. 2020). Most of the previous studies in the academic circle take western religion as the research background. The belief and practice of western religion are often symbiotic, and most of them regard religion as religious belief and identity by default (Ano and Vasconcelles 2010). Chinese religion is very different from western religion (Liu and Yang 1989). Under the Chinese social system, there is a lack of important religions with a formal organization system, and the boundaries between religions are not very clear. Many unorganized folk belief activities have become the main source of religious belief of Chinese people (Yang 1961). Different from the traditional religious practice in the west, Chinese religious practice is mostly atypical religious activities, such as fortune-telling, worshiping gods, offering sacrifices, and praying for Buddha, which are nested in daily life (Cao 2010). In such a religious context, religion has the characteristics of disunity of identity and practice (Yang 1961). Many individuals who participate in religious practice have no religious identity. Among China's religious populations, there are not only a small number of Christians, Orthodox and Protestant originated from the west, whose religious identity and religious practice are unified; but also a large number of Buddhists, Taoists, and other followers of religions originated in China, and their religious identity and religious practice are not unified. Therefore, China's religion is complex and diverse, which can be divided into different situations. It cannot be simply studied as an independent variable. Religious practice and religious identity should be listed separately for analysis. Through combing the existing literature, we find that the role of religious identity and religious practice as a moderator in the field of Chinese people's subjective well-being has been confirmed (Sun and Zhang 2019). It also enlightens us that we can use religious identity and religious practice as moderating variables to analyze the different roles they play in the relationship between traditional media and moral evaluation.
The aim of religion spreading its doctrine is to achieve social harmony. For example, Christianity emphasizes love for oneself, not only for one's relatives and friends but also for the enemy; Buddhism emphasizes compassion for all things and creates a harmonious life according to the "right way" (Yu 1988). Therefore, religious doctrines will make people more self-regulated, identify with society and achieve social harmony (Qiu 2011). The dissemination of religious doctrines cannot do without media. Before the birth of printing, religion preached morality and justice through popular and vivid spoken language. Bowman called the missionaries in the Catholic Church "mass media" (Burke 2005). With the advent of printing, religion no longer became the privilege of monks and nobles, making religious doctrines widely spread among people in different regions and languages, forming a broad community of identity (Liu 2009). With the emergence of television, newspapers, and other media, the ideological and spiritual communication advocated by religion has a new model. Take the United States as an example. Since 1977, the religious cable television network has been established. It has gradually developed into a scale of 530 broadcasting companies and spread all over the world. Contemporary religious television is often called "TV Evangelism" (Korpi and Kim 1986). When religious doctrines are transmitted through traditional media such as television, the audience will experience a great sense of dignity, pleasure, and satisfaction from the use of the media (Liu and Chen 2009). Therefore, it can be seen that authoritative media such as TV and newspapers will make people happier by spreading religious doctrines, increase their sense of social identity and moral evaluation, and help to achieve social harmony. Under the complex background of the disunity of religious identity and religious practice in China, it is necessary to study whether the relationship between traditional media and moral evaluation in different religious situations is consistent. First, when people have a religious identity. Studies have shown that people with a religious identity subconsciously present positive religious information that will make them show better self-regulation effects (Weisbuch-Remington et al. 2005). This ability will make individuals more easily satisfied (McCullough et al. 2000). McCullough and Willoughby (2009) put forward in their research that religious identity will have an impact on moral evaluation. Individuals with religious identity are more likely to be evaluated by moral objectivism and less likely to be evaluated by moral relativism (Yilmaz and Bahcekapili 2015). Therefore, under the influence of religious identity, the more an individual's contact with traditional media, the higher the moral evaluation of society. Second, when people have a religious practice. In Chinese society, the most widespread group is the group with religious practice. The common Chinese religious practice includes fortune-telling, incense burning, ancestor worship, and other informal religious activities. Generally, people will employ divination practitioners (fortune tellers, geomancy, witches, etc.) for the purpose of solving the difficulties they encounter in life (Lu 2019). Most of these people have encountered difficulties in real life and come to seek the protection of gods. So their life satisfaction is low. Therefore, under the influence of religious practices, the more frequent the traditional media usage is, the lower their moral evaluation towards society. To sum up, we propose the following two hypotheses: Hypothesis 2 (H2). Religious identity and religious practice can moderate the impact of traditional media usage on people's moral evaluation towards society. Compared with people without a religious identity, the influence of traditional media usage on moral evaluation towards society is enhanced when people have a religious identity.

Hypothesis 3 (H3).
Religious identity and religious practice can moderate the impact of traditional media usage on people's moral evaluation towards society. Compared with people without a religious practice, the influence of traditional media usage on moral evaluation towards society is weakened when people have a religious practice.
In summary, the hypothetical model of the study is shown in Figure 1.
sciously present positive religious information that will make them show better self-regulation effects (Weisbuch-Remington et al. 2005). This ability will make individuals more easily satisfied (Mccullough et al. 2000). Mccullough and Willoughby (2009) put forward in their research that religious identity will have an impact on moral evaluation. Individuals with religious identity are more likely to be evaluated by moral objectivism and less likely to be evaluated by moral relativism (Yilmaz and Bahcekapili 2015). Therefore, under the influence of religious identity, the more an individual's contact with traditional media, the higher the moral evaluation of society. Second, when people have a religious practice. In Chinese society, the most widespread group is the group with religious practice. The common Chinese religious practice includes fortune-telling, incense burning, ancestor worship, and other informal religious activities. Generally, people will employ divination practitioners (fortune tellers, geomancy, witches, etc.) for the purpose of solving the difficulties they encounter in life (Lu 2019). Most of these people have encountered difficulties in real life and come to seek the protection of gods. So their life satisfaction is low. Therefore, under the influence of religious practices, the more frequent the traditional media usage is, the lower their moral evaluation towards society. To sum up, we propose the following two hypotheses: Hypothesis 2 (H2). Religious identity and religious practice can moderate the impact of traditional media usage on people's moral evaluation towards society. Compared with people without a religious identity, the influence of traditional media usage on moral evaluation towards society is enhanced when people have a religious identity.

Hypothesis 3 (H3). Religious identity and religious practice can moderate the impact of traditional media usage on people's moral evaluation towards society. Compared with people without a religious practice, the influence of traditional media usage on moral evaluation towards society is weakened when people have a religious practice.
In summary, the hypothetical model of the study is shown in Figure 1.

Data Collection
The data used in this research were from the 2013 Chinese General Social Survey (2013 CGSS), which is the earliest national, comprehensive, and continuous academic survey project in China. The main reason we used the 2013 CGSS in this research is that the publishing scale of traditional media peaked in 2013 and then gradually declined, which indicated that it is more effective to examine the impact of traditional media usage on moral evaluation. In accordance with international standards, the CGSS has conducted a continuous cross-sectional survey on more than 10,000 households in various provinces, municipalities, and autonomous regions in mainland China by The National

Data Collection
The data used in this research were from the 2013 Chinese General Social Survey (2013 CGSS), which is the earliest national, comprehensive, and continuous academic survey project in China. The main reason we used the 2013 CGSS in this research is that the publishing scale of traditional media peaked in 2013 and then gradually declined, which indicated that it is more effective to examine the impact of traditional media usage on moral evaluation. In accordance with international standards, the CGSS has conducted a continuous cross-sectional survey on more than 10,000 households in various provinces, municipalities, and autonomous regions in mainland China by The National Survey Research Center at Renmin University of China (NSRC). The 2013 CGSS covered 487 communities in 28 provinces, autonomous regions, and municipalities, excluding Taiwan, Hong Kong, and Macau. In order to ensure the authenticity, objectivity, and validity of the data, phased sampling, stratified sampling, and cluster sampling were employed, with face-to-face interviews and supervised inspection methods. According to the overall investigation arrangement, the 2013 CGSS questionnaire includes four parts, which mainly includes a core module, a ten-year retrospective module, social morality module, and a public service satisfaction module. The variables used in this paper were mainly selected from the core and social morality modules. Specifically, these variables include both respondents' demographic characteristics, such as gender, age, ethnicity, religion, education, income, hukou, political status, housing status, marital status and health status, and other social Religions 2021, 12, 137 5 of 17 characteristics, such as the frequency of traditional media usage and respondents' moral evaluation towards society. In the 2013 CGSS, a total of 11,438 samples were collected. Since the section on social morality in the 2013 CGSS is an optional module, not all respondents answered these items. Hence, we ultimately obtained 4201 samples in this research after deleting missing and invalid samples, such as "Refused answer", "Unanswered", and "Singular value".

Measurements
Respondents' moral evaluation towards society was used as a dependent variable in this research. The relevant question presented in the 2013 CGSS was as follows: "How satisfied are you in the current moral situation of our society?" This item can effectively measure the people's moral evaluation towards society based on their own moral standards or guidelines. Since the 2013 CGSS is a national social survey, this item can represent the basic opinions of most Chinese people on moral evaluation towards society. In this item, the Likert scale was employed to measure the answer, which ranges from 1 to 5, respectively indicating "Very satisfied" to "Very dissatisfied". In order to ensure consistency in the process of data analysis. We recoded the answer "Very satisfied" to "Very dissatisfied" as "5" to "1".
Religious identity (RI) and religious practice (RP) were used as moderation variables in this research. In the 2013 CGSS, respondents were asked to report their own religious identity. Based on the previous measurement of religion by other scholars, we assigned "0" to "no religious identity" and "1" to those who had religious identities, no matter what religious identity they belonged to. The religious identity involved in the item not only includes polytheism in China, such as Buddhism, Taoism, and Folk religion but also includes monotheism in China, such as Islam, Catholicism, Christianity, Orthodox, Hinduism, Judaism, and other Christianity. Given the potential differences between polytheism and monotheism in China, we generated a new variable to represent the different religious types in China. To facilitate data analysis, we coded "No religious identity" as "0", "Believing in polytheism" as "1", and "Believing in Chinese monotheism" as "2". In terms of religious practice, respondents were asked to report their frequency of participating in religious activities, such as fortune-telling, burning incense at home, worshiping, etc. The relevant answer was as follows: "Never", "Less than once a year", "1-2 times a year", "Once a month", "2-3 times a month", "Almost every week", "Every week", "Several times a week". We found the frequency of people's participation in religious activities was relatively low and there was little difference in it. For example, only 0.76% of respondents were found to participate in religious activities once a month and only 0.5% of respondents participated in religious activities almost every week. Hence, it is difficult to further distinguish the frequency of participating in religious activities and explore the differences in it. Based on this empirical evidence, we simply divided religious practice into "No religious practice" and "Having religious practice", and coded "No religious practice" as "0" and "Having religious practice" as "1".
Traditional media usage was the main explanatory variable in our research. In the 2013 CGSS, respondents were asked to report the frequency of using media. This item contained six media types, which included newspaper, magazine, radio, television, the internet (including mobile internet access), and mobile customized news. Based on the results of factor analysis and reference to previous research, we treated newspaper, magazine, radio, and television as traditional media, and treated the internet (including mobile internet access) and mobile customized news as non-traditional media. In this paper, we only study the usage of traditional media. The relevant answer was also measured by a Likert scale ranging from 1 to 5 and respectively indicated "Never" to "Very frequent". The higher the Likert score is, the more frequent the usage of traditional media is.
Besides, we controlled other variables that could affect our outcomes. These variables in our research included gender, age, ethnicity, hukou, education, annual income, political status, housing status, marital status, health, and non-traditional media usage. In the Religions 2021, 12, 137 6 of 17 2013 CGSS, respondents were asked to report this basic information. To facilitate the data analysis, we recorded all these covariables. First, we coded "Female" as "0" and "Male" as "1". Then, we calculated the age of respondents based on the birth year and divided respondents into three age groups. We coded "Youth (18-45)" as "0", "Middle-aged (46-69)" as "1", and "Elderly (over 70)" as "2". In terms of ethnicity, we coded "Minority groups" as "0" and "Han people" as "1". Next, we divided education into six stages, where 1 to 6 presented "Illiterate", "Primary school", "Middle school", "High school or vocational high school", "Undergraduate", and "Graduate and above", respectively. In addition, we took annual income, which respondents were inquired about in 2013 in CNY (Chinese Yuan) as a logarithm in this research. In terms of marital status, we coded "divorced and widowed" as "0" and "Non-divorced and non-widowed" as "1", regardless of whether they married or not. In addition, we also set housing status and political status as dummy variables, among which we coded "Own housing property" as "1" and "Other's housing property" as "0", and coded "Non-communist party members" as "0" and "Communist party members" as "1". Last, we coded "Very unhealthy" to "Very healthy" as "1" to "5". Non-traditional media usage has been encoded before and we will not repeat it again.

Data Analysis Strategy
In this research, our main purpose was to examine the critical role of religion between traditional media usage and moral evaluation in modern China. Based on this purpose, we firstly report the summary statistics of samples and present the distribution of religions in this research. Then, we explored the moderation role of religious identity and religious practice between traditional media usage and moral evaluation. On this basis, we divided religious identity into polytheism, monotheism, and non-religious identity, and further compared the role of different religious identities in the relationship between traditional media usage and moral evaluation. Coefficients, standard error, and the significance level of the impact of traditional media usage on moral evaluation were obtained. In this research, STATA 15.1 and the PROCESS macro for SPSS (Model 2) were employed to conduct the data analysis.

Results
In this paper, we ultimately retained 4201 samples after deleting "Unanswered" "Refused answer", and "Singular value" samples. Since we collected and analyzed data scientifically and objectively, the representativeness of samples and the reliability of results can also be guaranteed. In this section, we present the results of data analysis, which included the results of descriptive analysis, moderation analysis, and heterogeneity analysis. Table 1 presents the distribution of all variables used in this research. As shown in Table 1, the distribution of moral evaluation showed that respondents' moral evaluation towards society is relatively high (M = 3.124), which indicated that respondents were quite satisfied with the moral status of society. Besides, we found that the frequency of traditional media usage (M = 2.502) was obviously higher than non-traditional media usage (M = 1.907). In addition, a total of 11.43% of respondents reported they had a religious identity and 15.04% of respondents reported they had a religious practice. In addition, a total of 55.11% of respondents were male and 44.89% of respondents were female. The average age of respondents was 49.18. Only 5.43% of respondents were "Minority people", more than 90% of respondents were "Han people". On the whole, the education level of respondents was relatively low (M = 3.086). 64.03% of respondents reported they only had middle school and below education and only 17.02% of respondents reported they had undergraduate and above education. In terms of personal annual income, the results showed that the average personal annual income was 27,464.81 CNY in 2013, while the lowest personal income was 200 CNY and the highest annual income was 80,000 CNY. From the political status reported by respondents, more than 80% of respondents were non-communist party members and only 11.57% of respondents were communist party members. In addition, almost 64% of respondents had their own housing property in 2013. In addition, the results showed that almost 90% of respondents had never been divorced or widowed. Notably, almost 10% of respondents had been divorced or widowed. Last, the distribution of health in Table 1 shows that respondents' self-estimated health was relatively positive. Almost 65.15% of respondents reported they were healthy.

Results of Descriptive Analysis
In addition, we conducted the cross table analysis to clarify the relationship between different types of religious identity and religious practice in China. As shown in Table 2, a total of 480 respondents reported they had a religious identity in the 4201 valid samples, which accounted for 11.43% of the total sample. This is roughly consistent with the statistical results (10%) of CFPS (2012). The results showed that more than 85% of respondents reported that they had no religious identity, which suggested that most people had no religious identity in China. As described before, a total of 632 respondents reported that they had a religious practice, which accounted for 15.04% of the total sample. Specifically, the results showed that 2.26% of respondents had no religious practice but had a religious identity. Almost 9.16% of respondents reported that they had both religious identity and religious practice. Almost 5.88% of respondents had no religious identity but had a religious practice. These results indicated that more than 80% of respondents had neither a religious identity nor religious practice. Based on the potential differences between monotheism and polytheism in China, we divided religious identity into polytheism and monotheism, of which polytheism included Buddhism, Taoism, and Folk religions and monotheism included Islam, Christianity, Catholicism, and other religions. The results in Table 2 show that a total of 7.16% of respondents had a polytheist identity, of which 1.83% of respondents had no religious practice but 5.33% of respondents had a religious practice. In terms of monotheism, the results indicated that a total of 4.38% of respondents had a monotheist identity, of which only 0.43% of respondents had no religious practice yet 3.95% of respondents had a religious practice. Although we found a few respondents with monotheist identity had no religious practice, but this was always a small probability event. However, it was a relatively common phenomenon to have a polytheist identity but no religious practice. Besides, we found that there were five people who had both polytheist and monotheist identities, which accounted for 1% of the total sample. Based on the characteristics of polytheism and monotheism, we believed that these people who claimed to have both polytheist and monotheist identities were mostly polytheist followers. Besides, we found that a total of 5.28% of respondents were Buddhists, which accounted for the largest proportion in China. 2.29% of respondents were Muslims and 1.86% of respondents were folk religion believers. Obviously, the proportion of folk belief was seriously underestimated in China. By asking people specifically about the folk belief practices they had participated in, this number surged to 52% (Yang and Hu 2012). In a word, there was a significant separation between religious identity and religious practice in Chinese religion, while this separation had not yet occurred in western religions.

The Results of Moderation Analysis
The moderation analysis was examined using the PROCESS macro for SPSS (Model 2), As shown in Table 3, when we controlled the covariables that affected our results, the direct effect of traditional media usage on moral evaluation was positively significant (β = 0.0463, p < 0.05) which indicated that the more frequent the traditional media usage was, the higher people's moral evaluation towards society was. Besides, the direct effect of religious practice on moral evaluation was marginally significant (β = −0.0919, p < 0.1), and the effect values were small and negative. However, the direct effect of religious identity on moral evaluation was not significant (β = 0.0704, p > 0.1). These two results indicated that it was difficult for religion to have a direct and significant impact on moral evaluation. But it is worth noting that the interaction of traditional media usage and religion was significant. Specifically, the interaction of traditional media usage and religious identity was positive and significantly affected people's moral evaluation towards society (β = 0.3195, p < 0.01), while the interaction of traditional media usage and religious practice was negative and significantly influenced people's moral evaluation towards society (β = −0.1491, p < 0.05). These results indicated that religion played a moderation role in the relationship between traditional media usage and moral evaluation towards society. To further explore the moderation role of religion in this relationship, we conducted conditional effect analysis to reveal the impact of traditional media usage on moral evaluation under different religious identity and religious practice levels. On this basis, we performed a simple slopes pattern. These results are shown in Table 4 and Figure 2.  Notes: "RI = 0" means people have no religious identity; "RI = 1" means people have a religious identity; "RP = 0" means people have no religious practice; "RP = 1" means people have a religious practice.  Table 4 displays the conditional effect of traditional media usage on moral evaluation under different religious identity and religious practice levels. As shown in Table 4, when people had a religious identity but had no religious practice, we found a positive significant relationship between traditional media usage and moral evaluation (95 percent CI = (0.2008, 0.5026)). When people had both religious identity and religious practice, traditional media usage was also positively significant to affect people's moral evaluation towards society (95 percent CI = (0.0944, 0.3108)). However, when people had no religious identity but had a religious practice, the effect of traditional media usage on moral evaluation is small, negative, and only marginally statistically significant (95 percent CI = (−0.2492, 0.0157)). Last but not least, we surprisingly found that when people had neither a religious identity nor religious practice, there was no significant relationship between traditional media usage and moral evaluation (95 percent CI = (−0.0113, 0.0758)).  Table 4 displays the conditional effect of traditional media usage on moral evaluation under different religious identity and religious practice levels. As shown in Table 4, when people had a religious identity but had no religious practice, we found a positive significant relationship between traditional media usage and moral evaluation (95 percent CI = (0.2008, 0.5026)). When people had both religious identity and religious practice, traditional media usage was also positively significant to affect people's moral evaluation towards society (95 percent CI = (0.0944, 0.3108)). However, when people had no religious identity but had a religious practice, the effect of traditional media usage on moral evaluation is small, negative, and only marginally statistically significant (95 percent CI = (−0.2492, 0.0157)). Last but not least, we surprisingly found that when people had neither a religious identity nor religious practice, there was no significant relationship between traditional media usage and moral evaluation (95 percent CI = (−0.0113, 0.0758)). Figure 2 displays the results of a simple slope pattern. As shown in Figure 2, there was a significant moderation role of religion in the relationship between traditional media usage and moral evaluation. The increasing slope of the line of traditional media usage to moral evaluation with the level of different religious identity and religious practice demonstrates the moderation role of religion. In a word, religious identity strengthened the effect of traditional media usage on moral evaluation, while religious practice weakened the impact of traditional media usage on moral evaluation. Specifically, when people had a religious identity but had no religious practice, the effect was significantly strengthened. When people had both religious identity and religious practice, the impact has also been enhanced to some degree. When people had neither a religious identity nor religious practice, the effect was weakened. When people had no religious identity but had a religious practice, the effect changed from positive to negative.

The Results of Heterogeneity Analysis of Polytheism and Monotheism
In this section, we further conducted heterogeneity analysis to test whether monotheism and polytheism had different impacts on the correlation between traditional media usage and moral evaluation considering the potentially significant differences. Ordered probit regression was used in this section. Table 5 displays the results of heterogeneity analysis of polytheism and monotheism. Based on the difference between religious identity and religious practice, we divided people into six groups. The first group referred to people with monotheist identity but no religious practice. Although these people did not participate in religious practice for some reason, the monotheist identity significantly enhanced the impact of traditional media usage on moral evaluation (see Model 1). However, in most cases, religious identity and religious practice were not separate in monotheism. In other words, monotheism had both religious identity and religious practice. The second group referred to people with a monotheist identity and religious practice, and the results showed that monotheism also strengthened the correlation between traditional media usage and moral evaluation (see Model 3). Compared with Chinese monotheism, there was an obvious separation between religious identity and religious practice in Chinese polytheism. The third group referred to the case that people had a polytheist identity but had no religious practice. In this case, the effect of traditional media usage on moral evaluation was also strengthened (see in Model 2). However, this effect was weakened compared with Model 2 when people had both polytheist identity and religious practice (see Model 4) and the effect was only marginally significant. Consistent with the previous results, when people only had a religious practice but no religious identity, traditional media usage had a significant negative impact on moral evaluation (see Model 5). Similarly, the correlation between traditional media usage and moral evaluation disappeared when people had neither a religious identity nor a religious practice (see Model 6).  N  18  77  161  224  247 3474 Notes: All covariables were controlled. Independent variable, traditional media usage; Dependent variable, moral evaluation; Model 1, people have a monotheist identity but have no religious practice; Model 2, people have a polytheist identity but have no religious practice; Model 3, people have both a monotheist identity and religious practice; Model 4, people have both a polytheist identity and religious practice; Model 5, people have no religious identity but have a religious practice; Model 6, people have neither religious identity nor religious practice. *** denotes significance at 1% level, ** at 5% level, * at 10% level. Standard errors in parentheses.

Discussion
Moral evaluation is happening all the time in the process of social transformation in China. When moral events occur, it will arouse widespread moral concern and discussion in society. These moral events include not only negative moral events, such as longevity vaccine, Didi hitchhike, surrogacy incident, etc., but also moral examples, such as the most beautiful villain, public welfare kitchen, and the most beautiful students. Previous research has confirmed the impact of traditional media usage on moral evaluation. However, it is not clear whether religion plays a critical role in the relationship between traditional media usage and moral evaluation and how it works in this relationship.
Peter Berger, who is the pioneer of secularization theory, believed that most countries in the world today are indeed religious rather than secularized (Berger 1997). Although we acknowledged that religion plays an important role in the past, we also cannot ignore that the role of religion seems to be diminishing or less obvious. With the development of social rationalization, the mystery of religion gradually faded and the original function and meaning also gradually faded. While some scholars believed that religion still plays a unique role in some specific fields, such as environment, happiness, mental health, and trust, etc. (Chung et al. 2019;Dilmaghani 2018;Preston and Baimel 2020;Sander 2017;Xu et al. 2018). This phenomenon makes us wonder if it is true that we are in a secularized society today? As well, to what extent does religion influence the correlation between traditional media usage and moral evaluation? In other words, this paper further examines whether we are in a secularized society today. In the past 20 years, academic circles have produces a large number of religious studies. However, they paid more attention to western religions, and this phenomenon still has not been obviously improved in more than 10 years (Lang 2004;Lu 2008). Religion is deeply embedded in local political and social culture (Zeng et al. 2020). Specific cultural background is one of the factors religious studies cannot ignore. Compared with western religion, Chinese religion is diffused religion, which indicates that religious identity and religious practice could exist independently. However, only a few scholars regarded religious practice as an independent factor in the past research (Cao 2010;McGuire 1990;Sun and Zhang 2019). Thus, there is no doubt that this research contributes to an in-depth understanding of Chinese religion in the relationship between traditional media usage and moral evaluation. In addition, since Chinese religion includes both polytheism and monotheism, it is helpful to expand the current study of religion when we explore whether there is a difference between Chinese polytheism and Chinese monotheism in the relationship between traditional media usage and moral evaluation.
Firstly, we further confirmed the positive correlation between traditional media usage and moral evaluation. The more frequent the traditional media usage is, the higher the people's moral evaluation towards society is. The results are consistent with previous studies (Scheufele 2002). Traditional media is the main information source of life satisfaction and political trust. As the frequency of traditional media usage increases, the people's inner efficacy, social identity, and quality of life will also improve accordingly, which greatly improves people's moral evaluation towards society. Therefore, these results confirm hypothesis one. In addition, we find it is difficult for religion to have a direct and significant impact on moral evaluation, which further confirms that religion is a complex situational factor and it hardly works directly in some specific fields. Furthermore, the different direct roles of religious identity and religious practice demonstrated that religious identity and religious practice are separate in Chinese society. This finding proved once again that Chinese religion is more about diffused religion. To a certain extent, although western religion (also represented by monotheism in this research) is the unity of knowledge and action. In other words, those who have a religious identity will participate in religious practice. Similarly, those who participate in religious practices will also have a religious identity. It is worth noting that these groups mainly participate in organized religious activities. However, Chinese religious believers account for the majority in China. These people may not necessarily participate in Chinese religious practice if they have a religious identity. Accordingly, these people may not necessarily have a religious identity if they participate in religious practice. In China, a large number of people mainly participate informal religious practices, such as fortune-telling, Feng Shui, burning incense at home, worshiping ancestors, and worshiping Buddha. Most of these people have no formal religious identity, but they widely participate in religious practice. Thus, it is not surprising that religious practice is more widely distributed and more influential than religious identity in the unique religious background of China. Why religious identity and religious practice have different impacts on moral evaluation will be interpreted in the following section.
Secondly, we find religions play an important moderation role in the correlation between traditional media usage and moral evaluation. Specifically, religious identity can strengthen the effect of traditional media usage on moral evaluation. On the contrary, religious practice can weaken the impact of traditional media usage on moral evaluation. This finding is unexpected but reasonable. This conclusion may be difficult to understand in a western context, while it may be reasonable in a Chinese context. Since the characteristic of Chinese religion, religious identity and religious practice are independent and separate from each other. Most people who have religious identity are clergy or believers who are in charge of religious affairs, or are interested in religious doctrines, or have an inheritance or have pious beliefs. No matter what the situation is, it is true that they have a sense of identity and belonging and identify the doctrines and rules of their religion. This religious information puts forward some specific moral standards and living standards for each individual who has a religious identity. These religious contents will be gradually internalized in themselves. On this basis, these people form a higher sense of self-efficacy. As the traditional media are used more frequently, people's self-satisfaction and selfefficacy will gradually externalize into social justice and moral satisfaction towards society. Therefore, it is not difficult to understand that these people who have a religious identity have higher moral evaluations towards society when they use traditional media more frequently. However, religious practice is significantly different from religious identity in China. Most people often participate in informal religious practice to pursue good fortune and avoid bad luck, illness, and dissatisfaction with current real life. Only a few people participate in religious practice based on doctrine and belief, such as Christianity, Islam, and Judaism. Compared with monotheism in China, polytheism in China always lives for the present. These people want to be blessed by God and the deceased relatives through sacrifice, praying to God, burning incense at home, worshiping ancestors, worshiping Buddha, etc. Most people who participate in religious practice are more likely to turn to seek help from illusory God as they are troubled by diseases or dissatisfied with current real-life (Zhou and Sun 2017). Most people participate in religious practice for the purpose of "practicality and utility", and have nothing to do with religious identity. Our data analysis found that these people who have religious practice but have no religious identity have low levels of education and income. When self-needs cannot be met by society, these people will ask the gods for help through participating in religious practice. In this situation, the information conveyed by traditional media and self-experience in society is often contradictory. Thus, the more frequent the traditional media usage is, the lower the moral evaluation towards society is.
On this basis, there are four different situations in the impact of traditional media usage on moral evaluation towards society. The first situation is that when people have a religious identity but do not participate in religious practice, traditional media usage positively and significantly affects moral evaluation towards society and the effect is largest.This situation is not hard to understand. The self-satisfaction and self-efficacy of religious information transmitted to the individual successfully transformed into social satisfaction and social justice under the impact of traditional media usage. Accordingly, people's moral evaluation towards society will also improve. The second situation is that when people participate in religious practice but have no religious identity, traditional media usage negatively and significantly influences moral evaluation towards society. Why is there a discovery that is completely opposite to the first situation above? We think that this finding just explains the difference between religious identity and religious practice. Since these people are dissatisfied with reality, the positive values conveyed by traditional media may aggravate their negative attitude towards their moral evaluation towards society. The third situation is that when people have both religious identity and religious practice, the effect of traditional media usage on moral evaluation is obviously weakened. Religious identity and religious practice are separated in China. It is not necessary for these people who have a religious identity to participate in religious practices, such as fortune-telling, sacrifice, Buddha worship, etc. Once people participate in religious practice, they may have some expectations and needs that cannot be met in society. When they face some difficulties and have some disappointments in real life, they will ask God and the dead for help. Although western religions also have some religious activities, their role is obviously different from the informal religious practices of Chinese religion. However, the religious practice of monotheism is only a minority in China. In most circumstances, people mainly participate in the religious practice of polytheism in China. This phenomenon indicates that even though people have a religious identity, they participate in religious practice mainly because some of their needs cannot be met in the current society. In this situation, it is difficult for people to extend the self-satisfaction and self-efficacy brought about by religious identity to the social realm through traditional media. Thus, although traditional media usage has a positive effect on moral evaluation, the effect is small. The fourth situation is that when people have neither religious identity nor religious practice, there is no correlation between traditional media usage and moral evaluation. This finding indicates that only in the context of religion can traditional media usage affect moral evaluation towards society. In other words, religion is an important situational factor in the relationship between traditional media usage and moral evaluation towards society.
Some people may wonder whether there is a difference in the role of monotheism and polytheism in the relationship between traditional media usage and moral evaluation. In this research, we use heterogeneity analysis to reveal whether there is a difference between monotheism and polytheism in China. On this basis, we compare the role of polytheism and monotheism in the correlation between traditional media usage and moral evaluation. Eventually, we found that the religious identity and religious practice of monotheism have the same impact on the relationship between traditional media usage and moral evaluation, while the religious identity and religious practice of polytheism have the opposite effect in the correlation between traditional media usage and moral evaluation. Besides, the religious identity of both polytheism and monotheism have positive guiding and disciplinary functions. This finding indicates that it is effective to divide religious identity into whether there is a religious identity in our research. Besides, this finding also indicates that we should regard religious identity and religious practice as independent factors when we study Chinese religion. Similarly, we can merge different religious identities together when we do some related research.
This research is very meaningful and enlightening. First, this paper indirectly proves that religion plays an important role in modern society. Although some people claim that we are in a secularized society, our research shows that religion is playing a unique role in specific fields in its own form. In a word, the rationalization of society does not necessarily lead to the secularization of religion. Second, the moderation role of religion is also confirmed in this research. Third, the different role of religious identity and religious practice is also confirmed. This research shows that we should emphasize the role of religion in modern society. To be sure, this research also has some limitations. Firstly, because of the limitation of data, we did not further distinguish the role of different religions, such as Buddhism, Islam, Christianity, etc. Secondly, we cannot provide more information about religion because there are only two items in the questionnaire to measure religion. Thirdly, considering the unique role of religious practice, we should further distinguish religious practice in further research. Lastly, perhaps the religious field has been gradually changing in the past ten years. We should re-examine the role of religion in the relationship between traditional media usage and moral evaluation in further research.

Conclusions
The current study examines the moderation role of religious identity and religious practice in the relationship between traditional media usage and moral evaluation. The current study can not only deepen the research on the relationship between traditional media usage and moral evaluation by adopting religion into this correlation but also expand the research of religion by dividing religion into religious identity and religious practice. The results of moderation analysis and heterogeneity analysis demonstrated the necessity to divide religion into religious identity and religious practice. Traditional media usage was proven to positively influence moral evaluation, which further confirms the unique role of traditional media in moral evaluation. However, the critical role of traditional media in moral evaluation cannot be separated from a religious context. Furthermore, religious identity was proven to strengthen the effect of traditional media usage on moral evaluation. Yet, religious practice was proven to weaken the influence of traditional media usage on moral evaluation. Besides, we found that both polytheist identity and monotheist identities play a positive role in this relationship, which indicates that the religious identities of polytheism and monotheism have the same role in the relationship between traditional media usage and moral evaluation. Last but not least, a large number of people in China mainly participate in informal religious practices, such as fortune-telling, Feng Shui, burning incense at home, worshiping ancestors, and worshiping Buddha, which negatively influence the impact of traditional media usage on moral evaluation. This paper helps to distinguish the different role of religious identity and religious practice in the correlation between traditional media usage and moral evaluation and clarify whether there is a different role of polytheist and monotheist religions in this relationship. Most importantly, this paper confirms that the rationalization of society does not necessarily lead to the secularization of religion.