Front. Environ. Sci. Frontiers in Environmental Science Front. Environ. Sci. 2296-665X Frontiers Media S.A. 1088895 10.3389/fenvs.2022.1088895 Environmental Science Original Research Influencing factors of air-quality perception in China: What is constructed? What is hidden? Zeng and Yang 10.3389/fenvs.2022.1088895 Zeng Xiangming 1 * Yang Gengjiao 2 1 Department of Administration Management, Law School, Wenzhou University, Wenzhou, China 2 College of Mathematics and Physics, Wenzhou University, Wenzhou, China

Edited by: James Kevin Summers, United States Environmental Protection Agency (EPA), United States

Reviewed by: Henk Folmer, University of Groningen, Netherlands

Qiang Wang, China University of Petroleum, Huadong, China

 *Correspondence: Xiangming Zeng, 20200193@wzu.edu.cn

This article was submitted to Environmental Citizen Science, a section of the journal Frontiers in Environmental Science

 04 01 2023 2022 10 1088895 03 11 2022 07 12 2022 Copyright © 2023 Zeng and Yang. 2023 Zeng and Yang

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

China is known to be widely concerned about the air quality of its major cities. Nevertheless, the factors that influence air-quality perception (AQP) are still a matter of discussion. On the basis of one idea that AQP is constructed by media contact, this study assesses the effects of quantitative approach to compare the effect of constructionism and realism on AQP, that is media contact, individualism, and health condition on AQP. Based on data of the Chinese General Social Survey (CGSS) in 2010, we build an ordered logistic regression model and introduce urban/rural dual structure and urbanization as control variables. The results show that media contact, health condition, and individualization affect AQP. The effect of media contact, however, varies depending on the media type, and the relationships between media contact and AQP change according to region when adding control variables. Health condition is significant only when it affects work and activities as well as when mental health (MH) occurs. Additionally, the function of health effect (HE) and MH on AQP only becomes obvious in some areas after adding control variables. Moreover, in most places, the effect of some individualization variables on AQP disappear after adding control variables, whereas they continue to be obvious in others. Therefore, the effect of control variables, which are urbanization level and urban/rural dual structure, is hidden. Furthermore, there is a need to pursue economic and environmental justice between developed and undeveloped areas. The study replies with the effect of constructionism and realism on AQP. And it also sheds light on both the constructed and hidden functions of media contact.

 air-quality perception media contact individualism social constructionism urbanization social realism 1 Introduction

AQP is the process occurring in people whereby sensory stimulation based on air quality is translated into organized experience. Previous study classified AQP into perception of the intensity and the harmfulness, and discovered both were influenced by environmental knowledge, socio-economic status, health experience, among others (Li et al., 2016). In this paper, AQP is defined as the joint product of the stimulation from air quality and of the process itself. Research on AQP is of importance in China, when the COVID-19 pandemic has profoundly affected and changed the global and regional environmental issues, and strengthened individuals’ awareness of air-quality indoor and outdoor (Wang and Su, 2020; Abdel-Salam, 2022). Moreover, the energy consumption and the industrial structure, which result from economic activity, traffic transportation, and urbanization, are still major concerns worldwide (Wang and Su, 2020; Zhang L. et al., 2022; Chen et al., 2022). Furthermore, the effects of China’s coal consumption control policy are diverse across regions (Zhao et al., 2022).

During the past 10 years, many studies have concerned AQP in China (Shi, 2015; Guo et al., 2016; Brimblecombe and Zong, 2019; Zhang and Zhao, 2019). Numerous factors are found to be correlated with AQP, including multiple stakeholders, industrial facilities, vehicles and city constructions, among others (Liu, 2022). Moreover, AQP has influences on destination image, tourists, urban settlement intentions, and health effects and concerns (Lan et al., 2016; Becken et al., 2017; Reames and Bravo, 2019; Yao et al., 2022) and, moreover, influences self-protection demands and migration, among other factors (Li and Tilt, 2018; Tu and Zhao, 2021). However, little research has investigated the effect from the perspective of social constructionism and social reality, which was one of the important concern in environmental sociology (Shi et al., 2022). In this study, we seek to compare the effects of social constructionism and social reality on AQP.

From the theoretical framework of this paper, AQP is influenced by two aspect of factors. One is the particles harmful to the health in the air, which influence human experience. Particulate materials, such as PM10 and SO2 are the main causes of rates of hospital admission due to asthma in Northeast China (Zhang and Zhao, 2019). The influence of PM2.5 on the mortality related to urogenital diseases is greater than that of cardiovascular and respiratory diseases, and the influence of NO2 and black carbon (BC) on the mortality related to cancer is greater than that of respiratory diseases (Ho et al., 2020). Individuals acquit with environmental knowledge would evaluate the air condition and take action like restricting outdoor activities and increasing expenditure for risks (Li and Folmer, 2022). Therefore, both the exposure to polluted air and the hazard of pollutants negatively and significantly influences individuals’ happiness (Li et al., 2014). And the deteriorating air condition has caused people’s AQP to worsen (Tu and Zhao, 2021). Moreover, it is found that AQP influences tourist arrival into China (Becken et al., 2017) and migration intention and residential satisfaction in urban areas (Yao et al., 2022). Another factor originated from the need of individuals, which is a social constructed phenomenon. With the purpose of protecting public health, the Chinese government has expended great efforts toward improving air quality. And the air pollution is gradually controlled, consequently, although it is heterogeneous in different regions (Lv et al., 2022; Yan et al., 2022). In September 2021, there were 7.9% days with mild pollution, 0.5% days with moderate pollution, and above 0.1% days with severe pollution. However, the public remain skeptical, believing that the solutions to air-pollution problems are only short term (Brimblecombe and Zong, 2019). The reason is that the energy use and greenhouse gas exceed after the event of the coronavirus disease COVID-19, although the outbreak of the coronavirus disease COVID-19 has greatly decreased the economic activities of China and restricted the traffic (Wang and Zhang, 2021). It also leaded to changes of the energy consumption of China, and blocked the environment deteriorate (Wang and Su, 2020). Therefore, the perception of air quality remains low particularly in some areas (Shi, 2015; Guo et al., 2016).

The process of haze presentation has shown that the process of stimulation, which is influenced by environmental policies and media contact, among others, also affects AQP. Most of the Chinese people are not supposed to be aware of PM2.5 because lacked of knowledge and it was not incorporated in the Air-Quality Index of China until 2012, or not freely available (Liu and Anbumozhi, 2009; Li et al., 2016; Hong et al., 2021). Moreover, there are few reports on haze in China before 2010, with data from China National Knowledge Internet (CNKI) showing that the number of reports in 2010 was 64 (cumulative number 228). In 2011, the number was 111 (cumulative number 339), and 99 in 2012 (cumulative number 448). It then increased abruptly to 1861 in 2013 (cumulative number 2309). Thus, PM2.5 has been gradually attracting attention after the US Embassy announced the hourly data of PM2.5 on 8 April 2008, which caused dispute between the US Embassy and the Environmental Protection Bureau of Beijing (Hong et al., 2021). On 22 October 2011, Pan Shiyi, one of the most successful real estate developers in China, forwarded the PM2.5 data of the US Embassy on his microblog, and attracted the attention of some microblog users. In 2012, the Ministry of Environmental Protection of China revised and formulated a new standard, which added PM2.5. At 19:00 to 19:30 on 18 January 2013, the national news broadcast of China Central Television (CCTV1) consumed 3 min and 30 s to report the serious haze problem in central and eastern China for the first time. Later in 2015, the documentary Under the Dome had more than 200 million views in 48 h in China, and has a long-term effect (Tu et al., 2020). Due to increasing debate about haze in media, such as on Microblog and in news media reports, researchers have concluded that media contact plays an important role in the appearance of PM2.5 and caused people to become familiar with the haze problem.

The process of haze presentation has triggered a dispute between social constructionism and social realism. Due to the effect of media, there is an idea in the literature that AQP is constructed by media contact and other social cultural factors rather than physiological feelings on air pollution (Huang and Yang, 2020; Tu et al., 2020; Hong et al., 2021; Shi et al., 2022). As a result, the social characteristics and social impacts of environmental problems of China are ignored. In addition, the connection between environment and society, which is the object of environmental sociology, is neglected. However, media content may be hiding something that is more important to AQP. It is found that the smog in London has been a reference point when journalists talk about smog problem in China, which justifies air pollution and naturalizes the problems (Li and Svarverud, 2018). When talking about the data of PM2.5 announced by the US Embassy, it is found the media preferred to compare the action of the US Embassy in China with silence of the US Embassy in India, Pakistan, and Bangladesh. Thus, some Chinese researches lead to the illusion that transforming the haze problem in China into a conflicting politicized framework (Xiaosu, 2014; Minghua et al., 2020). Therefore, the realism of the haze problem becomes hidden, although there is an opportunity to use a global perspective to combat the haze problem (Zhang and Samet, 2015).

The purpose of this article is to compare the effects of social constructionism and social reality on AQP. Media contact is used to measure the effect of social constructionism. Health condition is used to measure the effect of social reality. Individualism, which is constructed by the culture and social condition, is used to measure both the effects of construcionism and social reality. Nationally representative data are used to understand relationship between AQP and influencing factors. The data were collected in 2010, when most Chinese residents were not aware of PM2.5. Otherwise, it is harder for us to compare which factor is more important, particularly when knowledge about PM2.5 spread in China since 2011 and media contact play a more important role. Media contact, health condition, individualism, urbanization, and urban/rural dual structure are selected as possible influencing factors. The subjective perception of air quality near residences, which is selected as the dependent variable. By analyzing the relations of AQP and influencing factors, our study aims to compare the effects of social constructionism and social reality on AQP. This objective is achieved by exploring the following questions: (A) How does media contact affect AQP? (B) How does health affect AQP? (C) How does individualism affect AQP? (D) What is the role of urbanization and urban/rural dual structure?

 2 Data source and methods 2.1 Data source

This article uses data from the 2010 version of the Chinese General Social Survey (CGSS), which was officially implemented by the National Survey Research Center (NSRC) of China Renmin University in 2003. The CGSS is a national representative survey conducted annually on more than 10,000 households in 31 province-level administrative units in China. The 2010 CGSS adopted a three-stage stratified sampling design. The first step is selecting the primary sampling unit (PSU), which is divided into two categories. In category 1, the PSU is the sub-district office in five cities. Otherwise, the sample would be overly concentrated in distribution and lose representation. Included are the households in the selected five large and developed cities, which are Shanghai, Beijing, Guangzhou, Shenzhen, and Tianjin, which were selected from 36 possible cities (including municipalities directly under the central government, provincial capitals, and sub-province cities of China) based on the gross domestic product (GDP), total number of teachers, and foreign direct investment (FDI). Category 2 includes all families in the country except the five cities. The PSU is county-level units and districts. In the second step, community level units are selected, which are urban and rural communities. The last step is the selection of households.

Due to worldwide cooperation, the 2010 CGSS includes core content, a class consciousness module, a social stratification module, an income and consumption module, a religion module, an environment module, which is also conducted as part of the International Social Survey Program (ISSP), and a health module, which is conducted as part of the East Asian Social Survey (EASS). The number of valid respondents of the 2010 CGSS was 11,783, and 3,866 of these who were born in January, April, June, and August were selected to answer questions of the health module.

 2.2 Data exploration

Before employing descriptive and comparative statistics, the root mean square error of approximation (RMSEA) was introduced (McDonald and Ho, 2002), due to the large sample size, for the purpose of assessing the level of significance between AQP and variables related with media contact, health condition, individualism, urbanization, and urban/rural dual structure (see Table 1). This was conducted using the following equation. It is suggested that an RMSEA value of about 0.08 or less is a reasonable error of approximation, whereas a model with RMSEA greater than 0.1 should not be employed (Browne and Cudeck, 1992). R M S E A = F ̂ 0 d f test where F ̂ 0 = χ test 2 − d f test N

Chi-square results and RMSEA value.

Pearson chi-square Df Sig. N RMSEA
AQP-newspaper 241.69 12 0.000 3,838 0.07
AQP-magazine 231.04 12 0.000 3,836 0.07
AQP-radio 158.07 12 0.000 3,827 0.06
AQP-TV 33.90 12 0.001 3,846 0.02
AQP-Internet 248.78 12 0.000 3,835 0.07
AQP-SMS 120.52 12 0.000 3,835 0.05
AQP-health 62.69 12 0.000 3,844 0.03
AQP-HE 58.06 12 0.000 3,831 0.03
AQP-MH 40.63 12 0.000 3,834 0.02
AQP-OASRBSS 45.89 12 0.000 3,813 0.03
AQP-OASROTMP 40.50 12 0.000 3,823 0.02
AQP-OASRBM 72.47 12 0.000 3,821 0.04
AQP-ELDCOST 34.06 9 0.000 3,829 0.03
AQP-GENDIV 151.07 12 0.000 3,844 0.05
AQP-HOUSWKDV 28.51 12 0.005 3,840 0.02
AQP-FREESPECH 95.23 12 0.000 3,824 0.04
AQP-FREEBIRTH 70.99 12 0.000 3,835 0.04
AQP-FREEWRK 42.72 12 0.000 3,831 0.03
AQP-urbanization 166.93 6 0.000 3,849 0.08
AQP-Urban/rural 558.44 3 0.000 3,849 0.22

HE, Health effect; MH, Mental health; OASRBSS, Opinion about sexual realtions between adults of the same sex; OASROTMP, Opinion about married person having sexual realtions with someone other than the marriage partner; OASRBM, Opinion about sexual relation between two adults of the same sex), among others, are abbreviation. The variable name, measurement level, variable value could be seen in the following part 2.3.

Based on the RMSEA value, we classify urbanization and urban/rural dual structure as control variables, and media contact, health condition, individualism as independent variables.

 2.3 Variables and measures 2.3.1 Dependent variable 2.3.1.1 AQP

AQP concerns the judgment and assessment of the risks when we are faced with ambiguous and contradictory information about the air conditions. Air-pollution exposure has both physical and psychological effects. It is suggested that psychological effects might have a more significant influence than physical effects (Meertens and Swaen, 1997). Moreover, because the psychological effect is difficult to measure, researchers have introduced subjective indicators as an alternative (Deguen et al., 2012; Chiarini et al., 2020; Zhang Y. et al., 2022). We do not attempt to develop a new tool to measure AQP but instead use an existing question in the questionnaire, “How serious is the air pollution problem in your area?”, to which there are four responses, namely 1) most severe, 2) more severe, 3) less severe, and 4) not severe.

 2.3.2 Independent variables 2.3.2.1 Media contact

Previous study found that social networks transmit channel rather than media contact is more important for enhance awareness and perception (Tang et al., 2013). The CGSS proposes six questions measuring media contact that concern frequency. They are (A) “How often do you read newspapers?” (B) “How often do you read magazines?” (C) “How often do you listen to radio?” (D) “How often do you watch TV?” (E) “How often do you surf the Internet?” (F) “How often do you use short message service (SMS)?” There are five responses, namely 1) never, 2) several times a year or less often, 3) several times a month, 4) several times a week, and 5) daily.

 2.3.2.2 Individualism

The relationship of individual and society influences environmental action and perception. Measurement of the relationship between individual and society has previously involved the concept of cultural worldview, which includes individualism, hierarchy, egalitarian, and fatalism (Kim and Kim, 2019). Individualization is also another important concept. This means, firstly, that individuals have discarded all embracing social categories in industrial society, such as family, kinship, gender, and class. Individualization indicates the increasing awareness about emotions and desires, particularly sexual knowledge and rights, which have been controlled or stigmatized in traditional and socialist Chinese culture (Yan, 2010). It also indicates an attempt by individuals to remove the estrangement between the two sexes that was once characteristic of the basic traditional relationship in China (Qiu Zeqi, 2014). Additionally, studies have shown that the more acceptance there is of individualization, the more likely there is to be acceptace of premarital sex (Lo et al., 2010; Twenge et al., 2015).

Based on the measurements of a previous study (Qiu Zeqi, 2014), we use nine variables to measure individualism. (A) “What is your opinion about a man and a woman have sex relations before marriage (OASRBM)?” (B) “What is your opinion about a married person having sexual relations with someone other than the marriage partner (OASROTMP)?” (C) “What is your opinion about sexual relations between two adults of the same sex (OASRBSS)?” The responses are 1) always wrong (2) almost always wrong, 3) not sure whether it is wrong, 4) wrong only sometimes, and 5) not wrong at all. (D) “Do you think who should primarily be responsible for supporting elderly people who with children? 1) The government, 2) the children,3) the elderly themselves, or 4) both (ELDSPT). (E) “To what extent do you agree or disagree, a man’s job is to earn money; a woman’s job is to look after the home and family (HUBBYWK).” (F) “Both the man and woman should contribute to the housework (HOUSWKDV).” (G) “If a person makes a speech criticize government, the government should not interfere (FREESPECH).” (H) “The number of children one wants to give birth to is his or her own business. The government should not interfere (FREEBIRTH).” (I) “It is personal freedom to choose working and living place, the government should not interfere (FREEWRK).” The responses of (E-I) are 1) absolutely disagree, 2) disagree, 3) not sure whether I disagree or agree, 4) agree, and 5) absolutely agree.

The variables OASRBM, OASROTMP, and OASRBSS, which are related to questions (A-C), are used to measure whether individuals have cast off the shackles of traditional gender relations. The variable ELDSPT, which is related to question (D), is used to answer whether individuals have cast off the shackles of traditional generation relations. The variables HUBBYWK and HOUSWKDV, which are related to questions (E-F), are used to answer whether individuals have cast off the shackles of traditional family division of labor. The variables FREESPECH, FREEBIRTH, and FREEWRK, which are related to questions (G-I), are used to answer whether individuals have cast off the shackles of the government.

 2.3.2.3 Health condition

Air-pollution exposure has both physical and psychological effects (Deguen et al., 2012). However, the relationship between air pollution and health mainly depends on the perception, which is related to health condition, other individual and environmental factors (Elliott et al., 1999; Veloz et al., 2020). AQP reflects human perception of odor and dustiness, although there is no sensory organ for air-quality index in humans (Wolkoff, 2018). In this study, we consider both physical and psychological health conditions based on three questions, (A) “Would you say your own health, in general, is excellent, good, fair, or poor (health)?” The responses are 1) poor, 2) fair, 3) neither good nor fair, 4) good, and 5) excellent B) “Would you say your own health affect your work and activity during the past 30 days (HE)?” (C) “Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good (MH)?” The responses are 1) always, 2) almost always, 3) sometimes, 4) seldom, and 5) never.

 2.3.3 Control variables 2.3.3.1 Urbanization

Urbanization as a fundamental feature of economic development has major impacts on the environment, although the impact is relatively small when compared with economic growth and the energy structure (Li et al., 2021). Some study shows that air quality was related to the urbanization stages and policies in China (Liu et al., 2022). Urban land use is particularly negative for air quality unless the energy consumption, industrial, transportation structure and education levels are upgraded (Narducci et al., 2019; Chiarini et al., 2020; Gan et al., 2020). Based on the economic conditions of the 31 province-level administrative units, we classified those units into three types of urbanization. They are 1) developed regions, which include Beijing, Shanghai, Tianjin, Shandong province, Jiangsu province, Zhejiang province, Fujian province, and Guangdong province; 2) developing regions, which include Chongqing, Inner Mongolia, Shanxi province, Hainan province, Hunan province, Liaoning province, Anhui province, Henan province, Hubei province, and Shaanxi province; and 3) undeveloped regions, which include the remaining province-level administrative units.

 2.3.3.2 Urban/rural dual structure

Urban/rural dual structure originates from the household registration (Hannum, 1999). The household registration regulations were promulgated by the PRC in 1958 (Wu and Treiman, 2004) for the purpose of controlling movement. However, this led to inequality regarding resources supply and differences in development opportunities between urban and rural residents (Cheng and Selden, 1994; Chan and Buckingham, 2008). It is found that a household registration in the cities have impacts on residential satisfaction (Ren and Folmer, 2017). Moreover, PM2.5 is more highly concentrated in dense urban areas than in rural areas (Lu et al., 2019), although air pollution in urban areas contributes to disease both in urban and rural areas (Gong et al., 2012). Due to these reasons, this study introduces the survey location as a variable to measure the impact urban/rural dual structure. The responses are 1) urban area, 2) rural area.

 2.4 Descriptive and comparative method

In the descriptive part, the frequency statistics is used. It helps to describe the general condition of the respondent properties and their attitudes on AQP, media contact, health condition and individualism. In the comparative part, the ordered logistic regression (OLR) is used. The reason is that the purpose of this study is finding and comparing the influencing factors. Moreover, the dependent and the independent variables are ordinal measurement, which means that higher numbers mean either “more” or “less”, and control variables are nominal measurement. OLR is good at comparing the more important influencing factors, although it limits in detecting the casual links when compared with structure equation model (SEM). Meanwhile, SEM can handle latent and observable variables simultaneously (Tang et al., 2020). With the purpose comparing the effect of social constructionism and realism, we introduced the control variable to overcome the limits of OLR. We compare the functions of different types of media, health condition, and individualism on AQP, and consider it based on urban/rural dual structure and urbanization.

Relationships between AQP and media contact, AQP and health condition, AQP and individualism are compared by STATA. First, AQP is selected as a dependent variable, and media contact, health condition, and individualism are selected as independent variables batch by batch. Secondly, the OLR is repeated by introducing urbanization and dual structure in grouping variables separately. Finally, β and p are integrated into one table.

 2.5 Descriptive statistics of responses and variables

Table 2 presents the percentages of respondents regarding their responses related to AQP. It shows that around 33.70% of the respondents are in developed areas, 32.79% in developing areas, and 33.52% in undeveloped areas. 59.89% of respondents are in urban areas and 40.11% are in rural areas. Around 48.51% of the respondents are male and 51.49% female. The average age of the respondents is 47.12, and 24.43% of them are aged 17–35 years old, 54.86% are aged 36–60 years old, and 20.71% are more than 60 years old. The proportion of respondents who hold an AQP of air pollution as most severe is lowest, at 7.72%, whereas 21.38% of the respondents hold the AQP view that air pollution is more severe, 44.76% as less severe, and 26.14% as not severe. It can also be seen in Table 2 that the AQP of respondents varies substantially depending on differences in urbanization levels, regions, and ages. Respondents living in developed and urban areas are more likely to have an AQP that air pollution is most severe or more severe. Older respondents are more likely to hold an AQP that air pollution is not severe.

Responses recorded for AQP and their corresponding percentages according to various categories.

Urbanization Dual structure Gender Total Age Total
Deve- loped Deve-loping Undeve-loped Urban Rural Male Female 17–35 36–60 >60
Most severe 9.33 6.89 6.90 9.67 4.79 7.66 7.77 7.72 7.66 8.57 5.52 7.72
More severe 29.30 17.04 17.67 28.50 10.75 20.73 22.00 21.38 23.09 21.46 19.07 21.36
Less severe 46.72 45.40 42.17 48.72 38.86 45.42 44.15 44.76 49.79 43.25 42.91 44.78
Not severe 14.65 30.67 33.26 13.10 45.60 26.19 26.08 26.14 19.47 26.72 32.50 26.14
% 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Count 1,297 1,262 1,290 2305 1,544 1867 1982 3,849 940 2111 797 3,848
Percent 33.70 32.79 33.52 59.89 40.11 48.51 51.49 100.00 24.43 54.86 20.71 100.00

Table 3 shows data for the respondents’ media contact, health condition, and individualism. It shows that TV is the most popular media type, with newspaper second and Internet third. The proportion of respondents who always had contact through newspapers, magazines, radio, TV, Internet, and SMS was 6.88%, 1.27%, 4.64%, 40.56%, 9.51%, and 3.35%. 14.77%, 8.80%, 8.52%, 39.32%, 11.02%, and 4.83%. The self-evaluation of health condition shows that 33.15% of respondents answered that their health was good and 24.79% excellent. The self-reported of health effect shows that 43.69% never experienced effects on work or activity, 26.56% seldom, 3.98% always, and 10.34% almost always. The mental health (MH) self-evaluation shows that 2.05% always feel effects on MH, 9.35% almost always, 32.94% seldom, and 31.43% never.

Percentages of media contact, health condition, and individualism.

Never Seldom Sometimes Often Always N
Newspaper 41.97 20.48 15.91 14.77 6.88 3,853
Magazine 49.81 24.88 15.24 8.80 1.27 3,851
Radio 56.86 19.04 10.94 8.52 4.64 3,839
TV 2.23 6.37 11.62 39.22 40.56 3,863
Internet 66.20 6.50 6.78 11.02 9.51 3,849
SMS 75.60 10.11 6.11 4.83 3.35 3,849
poor Fair Neither good nor fair Good Excellent N
Health 4.04 14.22 23.80 33.15 24.79 3,861
Always Almost always Sometimes Seldom Never N
HE 3.98 10.34 15.44 26.56 43.69 3,848
MH 2.05 9.35 24.23 32.94 31.43 3,850
Always wrong Almost always wrong Not sure Wrong sometimes Not wrong at all N
OASRBM 56.34 14.27 19.69 6.94 2.76 3,834
OASROTMP 83.08 8.63 6.39 1.41 0.50 3,835
OASRBSS 83.51 4.08 10.77 1.02 0.63 3,826
Government Children Elder themselves Both —- N
ELDSPT 7.46 57.30 4.24 31.00 —- 3,845
Absolutely disagree Disagree Not disagree or agree Agree Absolutely agree N
HUBBYWK 6.45 17.20 8.55 41.75 26.06 3,861
HOUSWKDV 3.11 10.87 12.50 41.44 32.08 3,856
FREESPECH 14.22 33.14 20.57 23.41 8.67 3,841
FREEBIRTH 29.01 41.13 9.87 14.31 5.69 3,851
FREEWRK 6.78 15.86 10.87 38.97 27.53 3,847

Table 3 also shows the extent of individualism. The statistics show that 56.34% of respondents believe OASRBM is always wrong, 14.27% almost always wrong, 19.69% not sure, 6.94% wrong only sometimes, and 2.76% not wrong at all. Attitudes regarding OASRBSS and OASROTMP are similar, with respectively 0.63% and 0.50% considering it is not wrong at all, 1.02% and 1.41% wrong only sometimes, 10.77% and 6.39% not sure if it is wrong or not, 4.08% and 8.63% that it is almost always wrong, and 83.51% and 83.08% that it is always wrong.

Statistics for ELDSPT shows that 57.30% of individuals regard that children should be primarily responsible for supporting elderly people, 31.00% that it should be shared between the government, children, and elder themselves. When talking about the division of main duties for the family and responsibilities on housework, 67.81% of individuals agree or absolutely agree with the idea that it is a man’s job to earn money, whereas it is a woman’s job to look after the home and family, and 73.52% agree or absolute agree with the idea that both the man and woman should contribute to housework. It is shown that 32.08% of individuals agree or absolutely agree that the government should not interfere with a criticizer. Around 20.00% agree or absolutely agree that the government should not interfere with how many children one want to give birth, and 66.50% of individuals agree or absolutely agree that the government should not interfere with working and living place.

 3 Results and discussion 3.1 The effect of media contact

Media communication plays an important role in the public perception of air problems, which is related with the risk source and consequence of perception (Zhang Y. et al., 2022). However, the channel of information transmission is more important. Our study confirms the effects of media contact on AQP, but the level of media influence differs according to the type. Table 4 presents the relationship between AQP and media contact. Contact with newspaper (β =−0.18, p < 0.01), magazine (β =−0.05, p < 0.10), TV (β =−0.06, p < 0.05), and Internet (β =−0.20, p < 0.01) significantly affect AQP of air pollution in the living area. The β shows that AQP decreases or perception of air pollution is worse as media contact increases.

Coef. of AQP and media contact.

AQP Coef. By (repeated command by groups)
Urban Rural Developed Developing Undeveloped
Newspaper −0.18*** −0.06 −0.10 −0.07 −0.14* −0.21***
Radio −0.02 0.01 −0.07 −0.04 −0.03 0.01
Magazine −0.05* 0.02 −0.11 −0.06 −0.23*** 0.12*
TV −0.06** 0.01 −0.12** −0.02 0.02 −0.14***
Internet −0.20*** −0.09*** −0.22*** −0.13 −0.11* −0.31***
SMS 0.01 −0.01 0.01 0.04 −0.16** 0.10*
/cut1 −3.72 −2.55 −4.22 −3.04 −3.79 −3.79
/cut2 −2.09 −0.79 −2.91 −1.22 −2.31 −2.26
/cut3 −0.05 1.60 −0.99 1.05 −0.24 −0.31
Log likelihood −4634.54 −2755.52 −1,689.06 −1,563.90 −1,357.22 −1,653.50
LR chi26) 261.27 22.55 51.94 31.37 83.13 91.89
p 0.000 0.001 0.000 0.000 0.000 0.000

*, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively.

The function of each type of media varies by location. Moreover, in regions with different levels of development, the roles of various media are not identical. Previous research has not examined regional differences (Brody et al., 2004; Tu et al., 2020; Hong et al., 2021). And found individuals living in cities with limited AQP media coverage are more likely to be unaware of the haze problem (Hong et al., 2021). The results of this study show that urbanization, urban/rural dual structure, and AQP are significantly associated. Moreover, in contrast with previous evidence that people in cities tend to assign AQP as worse or poor than rural areas (Brimblecombe and Zong, 2019), our study found that TV plays an important role in the difference between urban and rural areas. It is also found that the effect of TV is more significant in rural and undeveloped areas. In Table 4, some of the significance disappears or changes when classifying by the variable urban/rural dual structure and urbanization. It shows that contact with newspaper has significantly effects in developing areas (β =−0.14, p < 0.10) and undeveloped areas (β =−0.21, p < 0.01). It is shown that newspapers have a reverse effect on AQP in undeveloped and developing areas. Furthermore, contact with magazines also affects those in developing areas (β =−0.23, p < 0.01) and undeveloped areas (β = 0.12, p < 0.10), which means the effect of magazine on AQP differs according to location. Thus, magazines have a reverse effect in developing areas and a promoting effect in undeveloped areas. Contact with TV significantly affects those in rural areas (β =−0.12, p < 0.05) and in undeveloped areas (β =−0.14, p < 0.01). Surfing the Internet significantly affects those in all the areas. TV has reverse effect both in rural areas and undeveloped areas. Internet has a reverse effect in all areas. Thus, the impact of newspaper, magazine, and SMS is more obvious in developing and undeveloped areas.

This result is in line with the findings that controlled media coverage of polluted air would limit the chance of concerns developing about the air quality (Hong et al., 2021). Nevertheless, individuals living in developing and undeveloped areas could also subscribe to magazines and receive customized SMS messages, which are vital for knowing the environmental condition. Moreover, the effect of Internet on AQP is more significant in rural and undeveloped areas than in urban and developing or developed areas. Individuals could instead access to information through the Internet, which is also critical for determining AQP. Therefore, the prosperity of mobile We Media may add some opportunities to bridge the gap.

From the consideration of different places, we conclude that: 1) Not all media have the function of construction. In general, the effect of radio and SMS on AQP is not significant. 2) The media types with a constructive function have different functions and degrees of influence in different areas. 3) Internet is one of the main sources involved in the construction of AQP. However, the effect differs according to area, being stronger in rural and undeveloped areas. 4) There is a reduction in the relationship between AQP and media contact resulting from grouping variables. 5) Control variables, including urban/rural dual structure and urbanization, have significant effects during the regulation process.

 3.2 The effect of health condition

The effect of health condition on AQP is a realistic factor that varies according to regional differences. Our study is in line with previous study that perceived hazardousness is influenced by health experience (Li et al., 2016). The result suggests that an effect on AQP will only be observed when health also affects work and activity. This relationship is significant in developing and undeveloped areas. Table 5 shows that when dividing by the variable urban/rural dual structure or urbanization, the significance of HE disappears in rural and urban areas and strengthens in developing (β =−0.14, p < 0.10) and undeveloped areas (β =−0.17, p < 0.01). Besides the effect of HE (health effect), we found MH (mental health) also has impact on AQP. Table 5 shows that HE (β =−0.17, p < 0.01) and MH (β = 0.10, p < 0.01) significantly affect AQP of air pollution in living areas. Thus, HE is significant in developing and undeveloped areas. It is consistent with the suggestion that respondents’ AQP worsens as they become more concerned about the impact of pollution exposure on health (Reames and Bravo, 2019). Furthermore, the MH (mental health) is found to be significantly associated with AQP, except for in rural areas. Table 5 shows that MH significantly affects those in urban areas (β = 0.18, p < 0.01) and also in developed (β = 0.16, p < 0.01), developing (β = 0.14, p < 0.05), and undeveloped (β = 0.10, p < 0.10) areas. That is, on the one hand, AQP bring concerns about physical health and, on the other hand, MH problems will also affect people’s evaluation of AQP.

Coef. of AQP and health condition.

AQP Coef. By (repeated command by groups)
Urban Rural Developed Developing Undeveloped
Health 0.01 0.04 −0.05 0.08 −0.06 −0.04
HE −0.17*** −0.07 −0.06 −0.06 −0.14 −0.17
MH 0.10*** 0.18*** 0.04 0.16*** 0.14 0.10
/cut1 −2.77 −1.68 −3.36 −1.61 −2.87 −3.01
/cut2 −1.18 −0.07 −2.06 0.21 −1.43 −1.53
/cut3 0.76 2.46 −0.17 2.45 0.54 0.34
Log likelihood −4717.22 −2742.49 −1,694.79 −1,565.08 -1,389.94 −1,673.18
LR chi23) 29.43 19.06 7.78 11.40 9.34 15.96
p 0.000 0.000 0.051 0.010 0.025 0.001

From the coefficients and discussion, we conclude that 1) Health evaluation does not affect AQP. 2) When there is an HE happens or MH is poor, there will be a significant effect on AQP. 3) The effect of HE on AQP is more obvious in developing and undeveloped area. 4) MH effects are more frequent and serious in urban and developed areas. 5) Control variables, including urban/rural dual structure and urbanization level, have significant effects during the regulation process.

 3.3 The effect of individualism

When asked how individualism affects AQP, our results indicate an obvious effect of individualism on AQP. Table 6 presents the relations between AQP and individualism. OASRBM (opinion about sexual relations before marriage) (β = −0.07, p < 0.05) significantly affects AQP. The data indicate a negative influence, that the higher the acceptance of OASRBM, the worse the AQP. ELDSPT (β =−0.05, p < 0.10) significantly affects AQP. GENDIV (β = 0.19, p < 0.01) significantly affects AQP. The significance of GENDIV is strengthened in rural (β = 0.20, p < 0.01) and developing (β = 0.22, p < 0.01) areas. It is also significant in developed (β = 0.08, p < 0.05) and undeveloped (β = 0.19, p < 0.01) areas. FREESPECH (β = −0.16, p < 0.01) significantly affects AQP. It is shown that FREESPECH significantly affects those in both urban (β = −0.5, p < 0.01) and rural (β = −0.13, p < 0.01) areas and in developed (β = –0.18, p < 0.01), developing (β = −0.10, p < 0.10), and undeveloped (β = –0.19, p < 0.01) areas. The FREEWRK (β = −0.05, p < 0.05) significantly affects AQP. FREEWRK significantly affects in undeveloped areas (β = –0.09, p < 0.05). OASROTMP, OASRBSS, HOUSWKDV, and FREEBIRTH are not significant.

Coef. of AQP and individualism.

AQP Coef. By (repeated command by groups)
Urban Rural Developed Developing Undeveloped
OASRBM −0.07** 0.02 −0.08 −0.01 −0.02 −0.08
OASROTMP −0.07 −0.06 −0.01 −0.17** 0.00 0.07
OASRBSS −0.03 −0.07 0.04 −0.02 −0.16** 0.03
ELDSPT −0.05* −0.04 0.06 −0.07 −0.02 −0.02
GENDIV 0.19*** 0.03 0.20*** 0.08** 0.22*** 0.19***
HOUSWKDV 0.00 −0.05 0.03 0.00 0.02 −0.01
FREESPECH −0.16*** −0.15*** −0.13*** −0.18*** −0.10* −0.19***
FREEBIRTH 0.01 −0.01 −0.05 0.00 −0.04 0.02
FREEWRK −0.05** −0.02 −0.04 0.07 −0.02 −0.09**
/cut1 −2.86 −3.14 −2.64 −2.74 −2.44 −2.81
/cut2 −1.26 −1.39 −1.33 −0.91 −1.00 −1.34
/cut3 0.75 1.02 0.59 1.37 1.04 0.54
Log likelihood −4576.08 −2688.60 −1,651.40 −1,533.73 -1,479.66 −1,490.65
LR chi29) 146.24 35.68 37.38 35.03 35.40 51.20
p 0.000 0.000 0.000 0.000 0.000 0.000

This result is in line with the idea that social networks transmit channel is more important for enhance awareness and perception (Tang et al.,2013). Due to the access to information, instant messaging (IM) has gained popularity in the past 2 decades. It is attractive to younger generations because of its ease sharing information (Correa et al., 2010). QQ (an IM and social platform developed initially for PC and then extended to mobile users) was once the most widely used IM platform in China before WeChat (an IM and social platform for mobile users). Besides QQ, the Chinese government has invested in a project that aims to extend TV coverage, and also the use of Internet and smartphones, to remote villages. Nevertheless, friendlier interface designs are needed to facilitate access for the elderly and disabled (Blažun et al., 2014; Guner and Acarturk, 2020; Sankhi and Sandnes, 2020; Shagerdi et al., 2022). Table 3 shows a high degree of individualism because of the effect due to OASRBM, ELDSPT, GENDIV, FREESPECH, and FREEWRK. It is shown that 31.00% of individuals considered multiparty responsibility most appropriate for supporting an elder, 73.52% agreed with a man and woman completing housework together, and 66.50% agreed or absolutely agreed to free selection of work and living place. Thus, the increase in access to information would contribute to the extent of individualism.

From the coefficients and discussion, we conclude that 1) The effect of OASRBM, ELDSPT, GENDIV, FREESPECH, and FREEWRK on AQP are significant. 2) GENDIV has a promoting effect on AQP, which indicates there is higher acceptance of GENDIV and more positive the judgment of AQP. 3) OASRBM, ELDSPT, FREESPECH, and FREEWRK have a reverse effect on AQP, the higher acceptance OASRBM, FREESPECH, and FREEWRK, the worse the judgment of AQP. Furthermore, the more accepting one is of supporting an elder by themselves or through multiparty responsibility, the worse the judgment of AQP.

 4 Conclusion

This study explores AQP in terms of the air-pollution problem in living areas, examining its relationship with media contact, health condition, and individualism. The OLR method is used. Two control variables, urban/rural dual structure and urbanization, are introduced.

The analysis replies with the dispute between social constructionism and social realism. It reveals the basic relationship between media contact, health condition, individualism, and AQP. In addition, the study sheds light on both the constructed and hidden functions of media contact. It is suggested that the difference in AQP is constructed due to the selective development of certain types of media in different regions. Furthermore, MH and HE are realistic factors that affect AQP in different areas. These results indicate AQP of air pollution is a concern among people living in all regions, although it is constructed in some areas and hidden in others. This indicates the demand for justice on air quality. Therefore, there is a need to reduce regional socioeconomic gaps, to popularize the application of Internet and smartphones, and to prevent the transfer of pollution from developed to undeveloped areas.

There are two methodological innovations in this study. One is the adoption of a quantitative approach in comparing constructionism and realism. The other is that the functions of different types of media, health condition, and individualism are considered based on urban/rural dual structure and urbanization. Compared with previous research, which found that media contact, health condition, as well as discussions in social groups are vital for the public perception of fog turning into haze in China, this study shows another shielding function. That is, the kind of media adopted in different regions can be crucial to further studying the regional differences in AQP. It is also crucial to protect people who live in these areas from air pollution. Furthermore, this study is limited regarding consideration of the contribution of education in different areas, although education may be a practical way to improve AQP.

The results can help individuals and policymakers better understand the function of each type of media aimed at air quality disclosure, evaluate the relationship between health risks and air quality accurately, and search for protection measures properly.

 Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: http://www.cnsda.org/.

 Author contributions

GY contributes by providing technical support. The corresponding author is responsible for all aspects of the study and its publication.

 Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

 Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

 References Abdel-Salam M. M. M. (2022). Assessment of children’s exposure to air pollutants in urban residences during the Covid-19 pandemic. Front. Environ. Sci. 10, 2176. 10.3389/fenvs.2022.1050623 Becken S. Jin X. Zhang C. Gao J. (2017). Urban air pollution in China: Destination image and risk perceptions. J. Sustain. Tour. 25, 130–147. 10.1080/09669582.2016.1177067 Blažun H. Vošner J. Kokol P. Saranto K. Rissanen S. (2014). “Elderly people’s interaction with advanced technology,” in Nursing informatics 2014 (IOS Press), 1–10. Brimblecombe P. Zong H. (2019). Citizen perception of apec blue and air pollution management. Atmos. Environ. 214, 116853. 10.1016/j.atmosenv.2019.116853 Brody S. D. Peck B. M. Highfield W. E. (2004). Examining localized patterns of air quality perception in Texas: A spatial and statistical analysis. Risk Anal. 24, 1561–1574. 10.1111/j.0272-4332.2004.00550.x Browne M. W. Cudeck R. (1992). Alternative ways of assessing model fit. Sociol. methods & Res. 21, 230–258. 10.1177/0049124192021002005 Chan K. W. Buckingham W. (2008). Is China abolishing the hukou system? China Q. 195, 582–606. 10.1017/s0305741008000787 Chen W.-T. Chang I. Li Z. (2022). Weighting analysis on air pollution among recent years in China via grey theory models. Front. Environ. Sci. 10, 1568. 10.3389/fenvs.2022.871839 Cheng T. Selden M. (1994). The origins and social consequences of China’s hukou system. China Q. 139, 644–668. 10.1017/s0305741000043083 Chiarini B. D’Agostino A. Marzano E. Regoli A. (2020). The perception of air pollution and noise in urban environments: A subjective indicator across European countries. J. Environ. Manag. 263, 110272. 10.1016/j.jenvman.2020.110272 Correa T. Hinsley A. W. De Zuniga H. G. (2010). Who interacts on the web?: The intersection of users’ personality and social media use. Comput. Hum. Behav. 26, 247–253. 10.1016/j.chb.2009.09.003 Deguen S. Ségala C. Pédrono G. Mesbah M. (2012). A new air quality perception scale for global assessment of air pollution health effects. Risk Anal. 32, 2043–2054. 10.1111/j.1539-6924.2012.01862.x Elliott S. J. Cole D. C. Krueger P. Voorberg N. Wakefield S. (1999). The power of perception: Health risk attributed to air pollution in an urban industrial neighbourhood. Risk Anal. 19, 621–634. 10.1111/j.1539-6924.1999.tb00433.x Gan T. Liang W. Yang H. Liao X. (2020). The effect of economic development on haze pollution (pm2. 5) based on a spatial perspective: Urbanization as a mediating variable. J. Clean. Prod. 266, 121880. 10.1016/j.jclepro.2020.121880 Gong P. Liang S. Carlton E. J. Jiang Q. Wu J. Wang L. (2012). Urbanisation and health in China. Lancet 379, 843–852. 10.1016/s0140-6736(11)61878-3 Guner H. Acarturk C. (2020). The use and acceptance of ict by senior citizens: A comparison of technology acceptance model (tam) for elderly and young adults. Univers. Access Inf. Soc. 19, 311–330. 10.1007/s10209-018-0642-4 Guo Y. Liu F. Lu Y. Mao Z. Lu H. Wu Y. (2016). Factors affecting parent’s perception on air quality–from the individual to the community level. Int. J. Environ. Res. Public Health 13, 493. 10.3390/ijerph13050493 Hannum E. (1999). Political change and the urban-rural gap in basic education in China, 1949-1990. Comp. Educ. Rev. 43, 193–211. 10.1086/447554 Ho H. C. Wong M. S. Chan T.-C. (2020). Spatially differentiating the effects of long-term air pollution on specific causes of death from cardiovascular and respiratory mortality in Hong Kong: A territory-wide register-based study. Air Qual. Atmos. Health 13, 721–730. 10.1007/s11869-020-00828-4 Hong W. Wei Y. Wang S. (2021). Left behind in perception of air pollution? A hidden form of spatial injustice in China. Environ. Plan. C Polit. Space 40, 666–684. 10.1177/23996544211036145 Huang J. Yang J. Z. (2020). Beyond under the dome: An environmental documentary amplified public risk perception about air pollution in China. J. Risk Res. 23, 227–241. 10.1080/13669877.2019.1569090 Kim H. K. Kim Y. (2019). Risk information seeking and processing about particulate air pollution in South Korea: The roles of cultural worldview. Risk Anal. 39, 1071–1087. 10.1111/risa.13231 Lan G.-l. Yuan Z.-k. Maddock J. E. Cook A. Chu Y.-y. Pan B.-b. (2016). Public perception of air pollution and health effects in nanchang, China. Air Qual. Atmos. Health 9, 951–959. 10.1007/s11869-016-0397-0 Li H. Svarverud R. (2018). When London hit the headlines: Historical analogy and the Chinese media discourse on air pollution. China Q. 234, 357–376. 10.1017/s0305741017001400 Li R. Wang Q. Liu Y. Jiang R. (2021). Per-capita carbon emissions in 147 countries: The effect of economic, energy, social, and trade structural changes. Sustain. Prod. Consum. 27, 1149–1164. 10.1016/j.spc.2021.02.031 Li X. Tilt B. (2018). Perceptions of quality of life and pollution among China’s urban middle class: The case of smog in tangshan. China Q. 234, 340–356. 10.1017/s0305741017001382 Li Z. Folmer H. (2022). Air pollution and perception-based averting behaviour in the jinchuan mining area, China. Ann. Reg. Sci., 1–29. 10.1007/s00168-022-01157-3 Li Z. Folmer H. Xue J. (2016). Perception of air pollution in the jinchuan mining area, China: A structural equation modeling approach. Int. J. Environ. Res. Public Health 13, 735. 10.3390/ijerph13070735 Li Z. Folmer H. Xue J. (2014). To what extent does air pollution affect happiness? The case of the jinchuan mining area, China. Ecol. Econ. 99, 88–99. 10.1016/j.ecolecon.2013.12.014 Liu H. Cui W. Zhang M. (2022). Exploring the causal relationship between urbanization and air pollution: Evidence from China. Sustain. Cities Soc. 80, 103783. 10.1016/j.scs.2022.103783 Liu H. (2022). How to improve environment performance? The comparison of stakeholder perceptions on how to improve air quality in China. Front. Environ. Sci. 478. 10.3389/fenvs.2022.889588 Liu X. Anbumozhi V. (2009). Determinant factors of corporate environmental information disclosure: An empirical study of Chinese listed companies. J. Clean. Prod. 17, 593–600. 10.1016/j.jclepro.2008.10.001 Lo V.-h. So C. Y. Zhang G. (2010). The influence of individualism and collectivism on internet pornography exposure, sexual attitudes, and sexual behavior among college students. Chin. J. Commun. 3, 10–27. 10.1080/17544750903528724 Lu X. Lin C. Li W. Chen Y. Huang Y. Fung J. C. (2019). Analysis of the adverse health effects of pm2. 5 from 2001 to 2017 in China and the role of urbanization in aggravating the health burden. Sci. Total Environ. 652, 683–695. 10.1016/j.scitotenv.2018.10.140 Lv Y. Fan T. Zhao B. Zhang J. Zheng Y. Zhang Z. (2022). How do government environmental concerns affect haze pollution? Front. Environ. Sci. 10. 10.3389/fenvs.2022.945226 McDonald R. P. Ho M.-H. R. (2002). Principles and practice in reporting structural equation analyses. Psychol. methods 7, 64–82. 10.1037/1082-989x.7.1.64 Meertens R. Swaen G. (1997). Psychological factors in air pollution epidemiology. Socioecon. Cult. Factors Air Pollut. Epidemiol. EC Air Pollut. Epidemiol. Rep. Ser. 8, 105–112. Minghua X. Danni L. Zhongzi W. (2020). Different others”: Differences in the construction of environmental images of eastern countries in the wWesternperspective——comparative perspectives on haze issues in CChinaand IIndiabased on gGooglenews. Journalism Commun. 27, 68–85. Narducci J. Quintas-Soriano C. Castro A. Som-Castellano R. Brandt J. S. (2019). Implications of urban growth and farmland loss for ecosystem services in the Western United States. Land Use Policy 86, 1–11. 10.1016/j.landusepol.2019.04.029 Qiu Zeqi B. Z. (2014). Who are the Chinese striving for? Forum of famous sociologists, 3. Beijing: China Social Sciences Press, 103–132. Reames T. G. Bravo M. A. (2019). People, place and pollution: Investigating relationships between air quality perceptions, health concerns, exposure, and individual-and area-level characteristics. Environ. Int. 122, 244–255. 10.1016/j.envint.2018.11.013 Ren H. Folmer H. (2017). Determinants of residential satisfaction in urban China: A multi-group structural equation analysis. Urban Stud. 54, 1407–1425. 10.1177/0042098015627112 Sankhi P. Sandnes F. E. (2020). A glimpse into smartphone screen reader use among blind teenagers in rural Nepal. Disabil. Rehabilitation Assistive Technol. 1–7, 875–881. 10.1080/17483107.2020.1818298 Shagerdi G. Ayatollahi H. Hemmat M. (2022). Opportunities for using health information technology for elderly care in the emergency departments: A qualitative study. Perspect. Health Inf. Manag. 19, 1h. Shi X. (2015). Factors influencing the environmental satisfaction of local residents in the coal mining area, China. Soc. Indic. Res. 120, 67–77. 10.1007/s11205-014-0584-z Shi X. Li X. Chen X. Zhang L. (2022). Objective air quality index versus subjective perception: Which has a greater impact on life satisfaction? Environ. Dev. Sustain. 24, 6860–6877. 10.1007/s10668-021-01730-4 Tang J. Folmer H. Xue J. (2013). Estimation of awareness and perception of water scarcity among farmers in the guanzhong plain, China, by means of a structural equation model. J. Environ. Manag. 126, 55–62. 10.1016/j.jenvman.2013.03.051 Tang J. Ren H. Folmer H. (2020). “Subjective wellbeing as valuation system of environmental quality: An environmental social sciences approach,” in Handbook on wellbeing, happiness and the environment (Edward Elgar Publishing), 85–103. Tu L. Zhao Y. (2021). Air pollution and the public perception level and self-protection demand in three cities of China: Fractional grey modelling analysis. J. Math. 2021, 1–15. 10.1155/2021/4951714 Tu M. Zhang B. Xu J. Lu F. (2020). Mass media, information and demand for environmental quality: Evidence from the “under the dome”. J. Dev. Econ. 143, 102402. 10.1016/j.jdeveco.2019.102402 Twenge J. M. Sherman R. A. Wells B. E. (2015). Changes in American adults’ sexual behavior and attitudes. Arch. Sex. Behav. 44, 2273–2285. 10.1007/s10508-015-0540-2 Veloz D. Gonzalez M. Brown P. Gharibi H. Cisneros R. (2020). Perceptions about air quality of individuals who work outdoors in the san joaquin valley, California. Atmos. Pollut. Res. 11, 825–830. 10.1016/j.apr.2020.01.010 Wang Q. Su M. (2020). A preliminary assessment of the impact of Covid-19 on environment–a case study of China. Sci. total Environ. 728, 138915. 10.1016/j.scitotenv.2020.138915 Wang Q. Zhang F. (2021). What does the China’s economic recovery after Covid-19 pandemic mean for the economic growth and energy consumption of other countries? J. Clean. Prod. 295, 126265. 10.1016/j.jclepro.2021.126265 Wolkoff P. (2018). Indoor air humidity, air quality, and health–an overview. Int. J. Hyg. Environ. health 221, 376–390. 10.1016/j.ijheh.2018.01.015 Wu X. Treiman D. J. (2004). The household registration system and social stratification in China: 1955–1996. Demography 41, 363–384. 10.1353/dem.2004.0010 Xiaosu H. (2014). The characteristics of the New York times’ report on China’s haze and its response to publicity. Extern. Commun. 213, 37–38. Yan D. Zhou M. Diao Y. Yang M. (2022). Air pollution in China: Spatial patterns and spatial coupling with population and economy. Front. Environ. Sci. 10, 2000. 10.3389/fenvs.2022.1040131 Yan Y. (2010). The Chinese path to individualization. Br. J. Sociol. 61, 489–512. 10.1111/j.1468-4446.2010.01323.x Yao L. Li X. Zheng R. Zhang Y. (2022). The impact of air pollution perception on urban settlement intentions of young talent in China. Int. J. Environ. Res. Public Health 19, 1080. 10.3390/ijerph19031080 Zhang H. Zhao Y. (2019). Land use regression for spatial distribution of urban particulate matter (pm10) and sulfur dioxide (so2) in a heavily polluted city in northeast China. Environ. Monit. Assess. 191, 712–714. 10.1007/s10661-019-7905-2 Zhang J. J. Samet J. M. (2015). Chinese haze versus Western smog: Lessons learned. J. Thorac. Dis. 7, 3–13. 10.3978/j.issn.2072-1439.2014.12.06 Zhang L. You S. Zhang S. Yi S. Zhou B. (2022a). The effects of urbanization on air pollution based on a spatial perspective: Evidence from China. Front. Environ. Sci. 10, 2280. 10.3389/fenvs.2022.1058009 Zhang Y. Chen J. Wei X. Wu X. (2022b). Development and validation of the haze risk perception scale and influencing factor scale–a study based on college students in beijing. Int. J. Environ. Res. Public Health 19, 4510. 10.3390/ijerph19084510 Zhao W. Chang M. Huangfu J. Yu L. (2022). Revealing effectiveness and heterogeneity of the impact of China’s coal consumption control policy on air quality. Front. Environ. Sci. 10, 2073. 10.3389/fenvs.2022.1050736