Factors influencing the energy saving potential of suburban dwellings in Jiangsu Province, China

Abstract

As environmental concerns continue to rise, scholars worldwide are increasingly focusing on energy efficiency in rural energy use. This study explores the energy efficiency potential of rural households in China, aiming to identify the key factors that influence this potential through correlation analysis. By utilizing multiple regression analysis, the study investigates the relationship between these factors and household characteristics. The research is based on rural areas in central Jiangsu Province, China, specifically targeting near and far residents to assess variations in energy-saving potential across different household types. The research findings indicate that: (1) The residential energy-saving potential of near suburban residents is primarily influenced by PC, ER, PR and education, with PR having the most significant impact. (2) The residential energy-saving potential of far suburban residents is mainly influenced by ER, PR, and EC, with EC having the most significant impact. (3) For all residents, PR, EC, SE, and GP significantly influence their residential energy-saving potential, with PR and EC having the most substantial impact. Based on these observations, targeted policy recommendations are proposed.

Introduction

Energy is the foundation of national development. Since the beginning of the twenty-first century, global energy consumption has surged. However, the inefficient use of energy has led to severe environmental pollution, posing a major threat to sustainable development¹. In response, China has outlined a comprehensive energy conservation and emission reduction plan in its 14th Five-Year Plan, positioning it as a key national strategy for the coming years². As a result, addressing energy consumption, improving energy efficiency, and reducing energy usage have become urgent priorities for China.

The use of traditional energy sources in rural areas has had a significant impact on the environment. The negative consequences of these energy sources, including air pollution, climate change acceleration, and health risks, are well-documented³. In 2015, the widespread reliance on traditional energy in rural China contributed to 17% of rural residents experiencing premature deaths⁴. Reducing the use of traditional energy in rural areas can help optimize the energy supply structure, improve public health, and mitigate indoor air pollution⁵. By 2020, 172 million rural households were still dependent on biomass fuels. Focusing on the adoption of renewable energy in rural areas could potentially reduce total CO2 emissions by up to 97% by 2050⁶. Therefore, accelerating the transition from traditional solid fuels to clean and efficient energy is essential for the sustainable development of rural China⁷.

There is a significant urban–rural disparity in carbon emissions across China, with Jiangsu Province standing out in this regard⁸. In 2014, the per capita energy consumption in rural areas was 541.35 kgce, significantly higher than the 364.10 kgce in urban areas, with rural residents accounting for 33.65% of total residential energy consumption⁹. Jiangsu Province exhibits a distinct urban–rural dual structure, characterized by considerable differences in cultural traditions, economic development, and energy consumption habits, all of which contribute to the higher per capita energy consumption in rural areas¹⁰ Residents’ electricity consumption behaviors are influenced by a range of factors, including environmental conditions, economic costs, policies, and social psychology, resulting in differing energy consumption patterns between urban and rural populations¹¹. Research indicates that integrating heat pumps with photovoltaic power generation and other clean energy sources in rural Jiangsu could facilitate the transition to green energy and reduce air pollution¹². Therefore, understanding the energy consumption behavior of rural residents in Jiangsu Province is essential for reducing carbon emissions and transitioning away from traditional energy sources.

Geographic location plays a significant role in shaping energy use patterns. In the Netherlands, for example, the distance from residents’ homes to the city center has a substantial impact on their lifestyle choices. Typically, residents living closer to the city center are more likely to adopt low-carbon transportation modes such as walking or cycling, which greatly reduces energy consumption¹³. Studies in the United States have shown that energy consumption increases significantly as the distance from homes to the city center grows¹⁴. Urban residents in the U.S. use approximately 80% less transportation energy than those living in rural areas¹⁵. In China, energy efficiency in eastern cities is notably higher than in western cities, with a disparity of 0.27¹⁶. These findings underscore the influence of residential proximity to the city center on energy consumption behaviors.

The Yangtze River Delta is one of China’s most economically dynamic regions, and in 2022, Jiangsu Province had the highest carbon emissions within the region¹⁷. For regional studies, Jiangsu is typically divided into Northern, Central, and Southern areas, with Nantong, located in Central Jiangsu, recording the highest carbon emissions in this region, making it a key area of interest for research. This study focuses on the energy-saving potential of rural households in Nantong, which is situated at varying distances from the city center. The findings may offer valuable insights that can be applied to other regions, including Northern and Southern Jiangsu. It is important to note that there is limited literature analyzing the regional disparities within Jiangsu, especially regarding how rural geographic location influences energy-saving potential. Therefore, this study not only addresses this research gap but also introduces a new perspective and methodology for understanding the energy-saving potential of rural households across Jiangsu.

This study employs correlation analysis and multiple regression analysis to conduct an in-depth comparative examination of the differences in rural energy consumption and the factors influencing it in Central Jiangsu. The findings offer valuable insights and guidance for developing low-carbon energy policies in rural China. The key contributions of this study are as follows: (1) A comparative analysis of the factors affecting the residential energy-saving potential of rural residents in Central Jiangsu. (2) A comparison of the differences in residential energy-saving potential between far suburban and near suburban rural residents. (3) The proposal of tailored strategies for residents based on their geographic location.

The structure of the paper is organized as follows: Section “Literature review and hypotheses” reviews the relevant literature and methodology. Section “Materials and methods” details the materials and methods used in the study. Section “Results and discussion” presents the results and provides a comprehensive discussion. Finally, Section ‘Conclusion and policy implications” concludes the paper and discusses the policy implications.

Literature review and hypotheses

Residential energy-saving potential and influencing factors

The energy-saving potential of residential buildings is influenced by both the energy consumption behavior of the residents and the inherent characteristics of the buildings themselves. Some scholars have predicted the energy-saving potential of homes by analyzing factors such as building type, residents’ energy consumption patterns, fuel types, and equipment configurations. These studies often evaluate the impact of these factors on residential energy consumption by developing models, providing a scientific basis for energy-saving renovations and policy development¹⁸. Research on energy-saving renovations for existing buildings has shown that the adoption of efficient technologies, optimization of building structures, and enhancement of equipment performance can significantly reduce energy consumption¹⁹. Based on these insights, this study categorizes the factors influencing the energy-saving potential of rural residences into two main types: traditional energy consumption behavior and the willingness to implement energy-saving renovations.

The energy-saving potential of residential buildings is significantly influenced by residents’ traditional energy consumption behavior. Existing literature, both from China and abroad, often focuses on concepts such as “residential energy behavior,” “household energy behavior,” and “residential energy-saving behavior,” which primarily describe individuals’ patterns of energy consumption and conservation²⁰. In addition to these behaviors, the residential energy-saving potential is also affected by the promotion of energy-saving technologies, community interactions, and energy-saving subsidies. Moreover, individuals’ perceptions of energy conservation play a crucial role in shaping household energy-saving outcomes²⁰. Regulating residents’ energy consumption behaviors can lead to substantial reductions in overall energy use. By raising awareness of energy conservation, encouraging the adoption of energy-efficient devices, and optimizing energy management policies, unnecessary energy waste can be effectively minimized²¹. Therefore, this study identifies traditional energy consumption behavior as a key factor influencing the energy-saving potential of residential buildings.

The energy-saving potential of residential buildings is significantly influenced by residents’ willingness to undertake energy-saving renovations. Research suggests that the success of energy-saving renovations in traditional homes is closely linked to this willingness, which is shaped by a range of factors, including economic conditions, living needs, government policies, corporate incentives, and socio-cultural influences²². In areas with deteriorating housing, such as urban slums, where high energy consumption and poor living conditions are common, residents typically exhibit a strong desire to implement energy-saving renovations. However, economic constraints often serve as a major barrier to carrying out these improvements²³. Other factors, such as education level, income, and the size of the dwelling, also influence residents’ willingness to undertake renovations²⁴. In the UK, some homeowners are interested in policies aimed at improving energy efficiency, which encourages them to undertake energy-saving renovations²⁵. Therefore, this study considers the willingness to undertake energy-saving renovations as one of the factors influencing residential energy-saving potential.

Traditional energy consumption behavior and the willingness to undertake energy-saving renovations are mutually influential. In some elderly communities, residents are often open to basic energy-saving improvements, such as enhancing building insulation and updating outdated facilities, to improve living comfort and reduce energy consumption. However, they are typically reluctant to completely abandon traditional energy sources²⁶. In Germany, private households demonstrate strong interest in clean energy, particularly in renovating areas with clean energy solutions, which have become their preferred choice²⁷. In the Netherlands, providing residents with clean energy options, such as natural gas, often serves as an incentive for them to undertake energy-saving renovations²⁸. In Shandong Province, China, the shift of rural residents from traditional energy sources to clean technologies like solar photovoltaic panels has effectively driven residential energy-saving renovations²⁹.

Individual, situational and sociodemographic factors

Existing literature indicates that the factors influencing residents’ energy use behavior are complex, diverse, and dynamic. For instance, in Jiangsu, residents in Central Jiangsu exhibit the highest level of energy-saving awareness, followed by those in Northern and Southern Jiangsu³⁰. In Central Jiangsu, residents are particularly influenced by policies, regulations, and energy-saving knowledge, while in rural areas, socio-technical factors also play a significant role in shaping energy use behavior³¹. Much of the existing research on energy-saving behavior and its influencing factors is grounded in Ajzen’s (1991) Theory of Planned Behavior. Scholars have broadly categorized these influencing factors into three main groups: individual factors, social factors, and sociodemographic variables^10,32. Individual factors include comfort preferences, energy-saving awareness, energy-saving knowledge, habits, and lifestyles. Social factors involve promotions, educational information, policies and regulations, economic costs, and socio-technical influences. Sociodemographic variables include gender, age, education level, occupation, income, family structure, and housing type³³. This study adopts a similar categorization, grouping the factors influencing energy consumption behavior into three categories: individual, contextual, and sociodemographic factors. It also reviews the latest research developments in this field to provide a comprehensive understanding of the factors at play.

Individual factors

Individual factors refer to the experiences, beliefs, and habits that residents accumulate over time, and these tend to be more stable compared to other influencing factors.

Some studies have identified a direct link between residents’ perception of climate change and their willingness to save energy, making it a key determinant of energy use behavior³⁴. Specifically, residents’ awareness of climate change can significantly influence their energy consumption patterns, and in turn, their energy use behavior can indirectly shape their energy-saving actions¹⁰. For instance, if residents recognize the negative environmental impact of climate change or understand the importance of energy-saving measures in mitigating its effects, they are more likely to adopt energy-saving practices, such as purchasing energy-efficient appliances and adjusting their energy usage habits³⁵.

H1a(b)

There is a positive correlation between residents’ awareness of climate change and their willingness to undertake energy-saving renovations (traditional energy use).

Residents of Myanmar demonstrate a strong sense of environmental responsibility, which significantly influences their energy-saving behavior. Those with a higher sense of environmental responsibility are more likely to adopt energy-saving measures, such as using clean energy technologies, which in turn reduces household energy consumption³⁶. Similarly, in China, residents who prioritize environmental responsibility are more inclined to accept clean energy devices like solar photovoltaic panels and air-source heat pumps. These technologies not only help reduce household energy consumption but also contribute to lowering carbon emissions³⁷.

H2a(b)

There is a positive correlation between environmental responsibility and the willingness to undertake energy-saving renovations (traditional energy use).

The relationship between spatial environment quality and space occupancy is both strong and reliable. Studies have shown that high-quality spatial environments—featuring good lighting, ventilation, and comfortable temperature and humidity—can significantly increase the frequency with which spaces are used³⁸. Indoor comfort, in particular, has a substantial impact on residents’ occupancy patterns, often encouraging them to spend more time indoors, which can lead to increased energy consumption in buildings³⁹. Conversely, research suggests that well-designed courtyards with high comfort levels can reduce the time residents spend indoors, thus indirectly lowering building energy consumption⁴⁰. As people age, their sensitivity to factors such as temperature, air quality, and noise increases. Older homeowners, in particular, tend to have higher demands for living comfort, which can drive their motivation to undertake energy-saving renovations⁴¹.

H3a(b)

There is a positive correlation between indoor comfort and the willingness to undertake energy-saving renovations (traditional energy use).

The adoption of clean energy by rural residents is crucial for long-term sustainable development. The widespread use of clean energy not only helps reduce carbon emissions but also improves the energy structure, enhances energy efficiency, and promotes green development⁴². Studies have shown that the willingness of rural residents to adopt clean energy is a key factor in its widespread implementation. Residents who are open to adopting clean energy tend to be more receptive to clean energy solutions in general⁴³. Additionally, factors such as traditional energy consumption habits and accumulated fuel usage significantly influence residents’ energy choices and their willingness to invest in cleaner alternatives⁴⁴.

H4a(b)

There is a positive correlation between the willingness to use clean energy and the willingness to undertake energy-saving renovations (traditional energy use).

The willingness to adopt low-carbon energy-saving measures plays a crucial role in influencing residents’ decision to undertake energy-saving renovations⁴⁵. Household energy-saving behavior is often shaped by a combination of personal values and the willingness to adopt low-carbon measures, with the latter being a more significant driver for promoting energy-saving renovations⁴⁶. When residents are strongly aware of the benefits of low-carbon energy-saving practices, they are more likely to implement energy-saving technologies and improve the energy efficiency of their homes, leading to reductions in both energy consumption and carbon emissions. In rural areas, particularly, increasing farmers’ awareness of the importance and necessity of energy-saving renovations can be a powerful catalyst for the successful adoption of such measures⁴⁷.

H5a(b)

There is a positive correlation between the willingness to adopt low-carbon energy-saving measures and the willingness to undertake energy-saving renovations (traditional energy use).

Situational factors

Situational factors refer to external influences that shape residents’ energy-related behaviors. These include interpersonal influences (such as persuasion and demonstration), social norms, costs, public relations efforts, and material incentives^10,48.

In China, residents’ low-carbon consumption behaviors are often influenced by a tendency to conform to others. They are more likely to imitate the actions of those around them, particularly when it comes to adopting energy-saving devices. For instance, when neighbors, friends, or members of the same community start using energy-efficient appliances—such as energy-efficient household devices or solar water heaters—individuals are typically influenced by these behaviors and inclined to follow suit⁴⁹. The impact of collective culture in China is particularly strong, as it emphasizes social norms and shared values. As a result, individuals often take into account the behaviors and opinions of others when making decisions about adopting energy-saving measures^50,51.

H6a(b)

There is a positive correlation between group mentality and the willingness to undertake energy-saving renovations (traditional energy use).

Policy plays a crucial role in influencing energy consumption behavior and encouraging residents’ active participation in residential energy-saving renovations. For example, due to the policy prohibiting the burning of crop straw, by 2015, only about 8% of rural households continued this practice⁵². Similarly, the implementation of policies such as the industrial structure optimization policy in Jiangsu Province has contributed to improvements in air quality⁵³. Economic subsidies also have a significant impact on household energy consumption and energy-saving renovations. Electricity subsidy policies, for instance, incentivize rural residents to switch to clean electricity sources⁵⁴. Policies are instrumental in shaping rural residents’ willingness and behavior regarding energy savings. In Northern China, direct energy-saving subsidy policies have successfully encouraged residents to adopt energy-efficient technologies. To effectively reduce household energy consumption, it is essential for the government to develop tailored energy-saving strategies that account for the unique conditions of different regions and populations⁵⁵.

H7a(b)

There is a positive correlation between policies and regulations and the willingness to undertake energy-saving renovations (traditional energy use).

Greek residents, like those in many other countries, demonstrate a strong focus on economic policies related to residential energy-saving renovations⁵⁶. This trend is also observed in other parts of the world. In Western Europe, for instance, a well-structured energy pricing system has proven effective in encouraging households to voluntarily adopt energy-saving measures⁴⁵. In Germany, residents are particularly motivated to participate in energy-saving renovations when the financial benefits, such as savings and a favorable payback period, are clear⁵⁷. This highlights the significant role of policy incentives—such as tax breaks and subsidies—and long-term economic analyses in encouraging residents to adopt energy-saving renovation measures.At the same time, residents often take into account the construction and maintenance costs associated with energy-efficient buildings, which can influence their decision to undertake energy-saving renovations⁵⁸. Comprehensive energy-saving renovations, including upgrades to roofs, windows, and walls, can lower overall costs and significantly reduce energy consumption, making large-scale energy-saving renovations across entire communities more feasible⁵⁹.

H8a(b)

There is a negative correlation between economic costs and the willingness to undertake energy-saving renovations (traditional energy use).

In Ethiopia, the increase in household energy use is often accompanied by a rise in energy expenditure. As the demand for energy grows, residents tend to opt for cleaner and more cost-effective energy sources, particularly when energy costs become a significant factor in their decision-making⁶⁰. The choice between electricity or clean coal for heating is frequently influenced by the use of firewood or soft coal. With improved economic conditions or policy support, households may gradually transition to electricity or clean coal for heating³. Similarly, in rural China, households with higher per capita energy consumption are generally more willing to adopt energy-efficient appliances and are more likely to embrace clean energy options⁶¹.

H9a(b)

There is a positive correlation between energy use and the willingness to undertake energy-saving renovations (traditional energy use).

Sociodemographic factors

Numerous studies have shown that sociodemographic factors such as age, income, family size, education level, and population density play a significant role in shaping energy consumption patterns and the choice of energy-efficient appliances⁶². Larger households tend to have higher heating demands and are more likely to invest in convenient household appliances for heating, thus reducing their reliance on biomass energy⁶⁰. Additionally, individuals with higher levels of education and income are more inclined to use electrical appliances rather than biomass energy⁶³. In Jiangsu, urban residents’ sociodemographic characteristics—including age, gender, income, family structure, and educational background—are key factors influencing their energy-saving behaviors⁶⁴. Furthermore, a household’s economic status, past renovation experience, house location, and energy expenditures all play a crucial role in determining residents’ willingness to undertake energy-saving renovation projects⁶⁵.

Materials and methods

Study area

The study area for this research is Hai’an City, located in Nantong, Jiangsu Province, China. Hai’an has a typical subtropical monsoon climate, characterized by an annual average precipitation of 989 mm, an average temperature of 14.7 °C, and an average relative humidity of 75%. As of 2022, Hai’an’s resident population reached 868,000, with 30.43% of the population engaged in agriculture, indicating a substantial agricultural community. According to the Global Atmospheric Research Emissions Database (EDGAR), carbon emissions from a rural area in Hai’an (at 120.5°E, 32.4°N) are estimated to be 4.3 × 10⁵ tons.

Sample selection

In 2020, Hai’an City defined its urban areas as “Zhongcheng, Nancheng, Xicheng, and Beicheng Streets,” collectively referred to as “Hai’an Street,” while all other areas were classified as suburban. For the purpose of this study, households located in Hai’an Street were categorized as near suburban, while those in the remaining areas were classified as far suburban. This research, conducted in January 2024, employed a multi-stage stratified random sampling method. In the first stage, the research team distributed online questionnaires to gather general information about rural households across Hai’an City. To investigate differences in energy structures between far suburban and near suburban areas, six villages were selected, considering factors such as township size, population, and geographical location. far suburban areas included Duntou Town and Chengdong Town, while Hai’an Street represented the near suburban area. In the second stage, two villages were randomly selected from each township for further investigation. These included Yuanzhuang and Zhoujiazhuang in Hai’an Street, Sanjiao and Shizhuang villages in Chengdong Town, and Qianbu and Maozhuang villages in Duntou Town.In the third stage, 50–80 households were randomly selected from each village, based on population size and residents’ willingness to participate. Household heads were surveyed, resulting in a total of 434 surveys. After reviewing and eliminating incomplete or inconsistent responses, 421 valid responses were used for further analysis.

Some scholars have randomly selected villages based on available resources and policies to ensure research accuracy³⁴. In this study, villages were selected according to two key factors: location and primary industry. The villages selected include P1 and P2 from the near suburban area of Hai’an Street, P3 and P4 from Chengdong Town (far suburban), and P5 and P6 from Duntou Town (far suburban). The industrial profiles of the selected villages are as follows: P1 is primarily industrial, P2, P3, and P6 focus on light industry, while P4 and P5 are mainly agricultural.By selecting villages with different primary industries, the study aimed to ensure a more random and representative sample. A total of 130 responses were collected from near suburban residents, and 291 responses from Far suburban residents. Table 1 presents the characteristics of the selected villages.

Table 1.

Village characteristics.

Village	Location	Coordinates	Main industry	Number of interviews
P1	Yuanzhuang village	Longitude:120.46 Latitude:32.51	Machinery	61
P2	Zhoujia village	Longitude:120.51 Latitude:32.57	Textile	69
P3	Sanjiao village	Longitude:120.54 Latitude:32.56	Furniture	68
P4	Shizhuang village	Longitude:120.62 Latitude:32.47	Plantation	71
P5	Qianbu village	Longitude:120.36 Latitude:32.58	Agriculture, food industry	78
P6	Maozhuang village	Longitude:120.34 Latitude:32.60	Chemical fiber industry	84

Questionnaire

To conduct this study, the research team developed a survey questionnaire based on an extensive review of relevant literature. The questionnaire was then validated by a panel of experts in rural energy use. To ensure its adequacy and accuracy, we carefully incorporated feedback from various stakeholders, making necessary revisions to enhance clarity and comprehensibility.

The questionnaire is organized into four sections: (1) Basic Household Information, (2) Awareness and Adaptation to Local Climate Change, (3) Household Energy Consumption Structure, and (4) Willingness to Use Clean Energy.

The first section collects basic household information, including details such as age, education level, household size, housing area, and number of floors.

The second section of the questionnaire investigates respondents’ awareness of climate change, with the aim of gaining deeper insights into their understanding of the connection between personal energy usage and climate change. This section also explores whether respondents perceive climate change as a factor contributing to rising living costs and its impact on indoor comfort during both summer and winter. Respondents were asked to indicate their level of agreement with various statements related to the effects of climate change on their daily lives, particularly concerning its influence on energy consumption and the rising cost of living.

The third section focuses on examining the current energy consumption patterns in rural areas of central Jiangsu. The primary energy sources used by farmers in the study area include electricity, liquefied petroleum gas, and firewood. Respondents were asked to identify the types of energy they had consumed over the past year and estimate the quantities used. To assess energy usage, respondents were provided with a scale, with options such as “never used,” “rarely used,” “occasionally used,” “frequently used,” and “used every day.”

The fourth section of the questionnaire explores respondents’ attitudes toward clean energy. It assesses their level of understanding of clean energy, awareness of clean energy technologies, and perceptions of traditional energy sources. Additionally, this section investigates the influence of social factors—such as the attitudes of people around them—and government policies on their willingness to adopt clean energy solutions. Respondents are also asked about the areas of their homes where they would be most willing to implement energy-saving renovations. All questions in this section are measured using a 5-point Likert scale, where 5 represents “strongly agree” and 1 represents “strongly disagree.”

Data analysis

Independent variables

All questions are assessed using a 5-point Likert scale to reflect the frequency of energy use behaviors among farmers, where 5 represents “strongly agree,” 4 represents “agree,” 3 represents “neutral,” 2 represents “disagree,” and 1 represents “strongly disagree.” The details of these variables are shown in Table 2.

Table 2.

Explanation of variables.

Variable type	Item assignment	Variable interpretation
Traditional energy use habits		Use of traditional energy sources in the daily life of the population
Willingness to retrofit homes for energy saving		Willingness of residents to remodel existing buildings
Individual factors	Perception of climate	Awareness of the environment
	Environmental responsibility	Willingness to protect the environment
	Indoor comfort	Perceived indoor thermal comfort
	Perception of clean energy	Willingness to use clean energy
	Willingness to save energy with low carbon	Willingness to take the initiative to implement low-carbon energy-saving behaviors
Situational factors	Group psychology	Influence of people around you
	Policies and regulations	Government’s clean energy policy
	Economic costs	Energy and equipment costs
	Energy use	Use of renewable and non-renewable energy sources
Sociodemographic factors	L.M.N.O.P.Q	Age, education, housing size, etc

To eliminate the impact of multicollinearity among the variables, this study conducted a multicollinearity test on the independent variables using SPSS. The results are shown in Table 3. It can be observed that the tolerance (TOL) of each independent variable is greater than 0.1, and the variance inflation factor (VIF) is less than 10, indicating that there is no multicollinearity issue among the explanatory variables. With the covariance problem controlled, the Hausman test showed that the model did not have significant endogeneity problem (p > 0.05).

Table 3.

Multicollinearity test.

Variant	TOL	VIF
Perception of climate	0.55	1.8
Environmental responsibility	0.55	1.85
Policies and regulations	0.14	8.8
Economic costs	0.12	8.0
Indoor comfort	0.98	1.0
Energy use	0.81	1.2
Perception of clean energy	0.14	7.2
Willingness to save energy with low carbon	0.51	1.94
Group psychology	0.13	7.18

Implicit variable

Based on section “Residential energy-saving potential and influencing factors”, this study summarizes the factors influencing the energy-saving potential of rural residences into two categories: traditional energy consumption behavior and willingness to undertake energy-saving renovations. A detailed explanation is provided in Table 4.

Table 4.

Detailed description of the dependent variable.

Categories	Items	Source
Traditional energy use habits	Windows	70–72
	Door
	Ground
	Roof
	Wall
Willingness to retrofit homes for energy saving	Habits of using traditional energy sources are difficult to change	73–75
	The phasing out of conventional energy sources has left me feeling a huge sense of loss
	Very easy to use traditional energy sources
	Conventional energy sources account for the vast majority of energy use

Model selection

The collected information was edited, coded, and analyzed using Excel® spreadsheets and SPSS 27.0 statistical software. The full name of SPSS is IBM SPSS Statistics, and its version number is 27.0(URL: https://www.ibm.com/products/spss-statistics). In this study, descriptive statistics were initially used to analyze the survey data. Subsequently, Pearson correlation analysis was applied to examine the extent to which various factors influence WTR and TEU in rural residences. Finally, multiple regression analysis was employed to establish linear models, exploring the factors influencing WTR and TEU, with the aim of identifying the factors that affect the energy-saving potential of rural residences.

1. Pearson Correlation Analysis: The Pearson correlation coefficient is used to measure the linear relationship between variables. Its mathematical definition is shown in Formula (1). Here, ‘n’ represents the sample size, and ‘x’ and ‘y’ represent two variables, with ‘r’ denoting the correlation coefficient. The correlation coefficient ranges from − 1 to 1; if its absolute value exceeds 0.6, it indicates a strong linear relationship between the two variables. In this section, the factors influencing the energy-saving potential of rural residences are divided into WTR and TEU, with a total of 11 variables.

   1
2. Multiple Regression Analysis: Equations (2), (3), and (4) indicate a linear regression model with q explanatory variables. The 17 explanatory variables include individual, situational, and Sociodemographic factors. Two of the explanatory variables represent energy consumption behavior and renovation willingness, with their values being the average scores of the three groups of residents on the two types of energy consumption behavior items. Finally, the correlation coefficients are introduced in Eq. (2).

   2

   3

   4
3. Reliability and Validity Test: According to Eqs. (5) and (6), composite reliability (CR) and average variance extracted (AVE) are used to assess the reliability of the research model. If the composite reliability (CR) is greater than 0.7 and the average variance extracted (AVE) is greater than 0.5, the latent variables are considered to be highly reliable.

   5

   6
4. Durbin-Watson Test: The Durbin-Watson test is used to examine whether the random error terms exhibit first-order autocorrelation, i.e., serial correlation. According to Eq. (7), if the value lies between 1.5 and 2.5, it indicates that the samples are independent and there is no autocorrelation.

   7

Ethics statement

This study received approval from the Ethics Committee of the School of Architecture and Design at China University of Mining and Technology. The experiment was conducted in accordance with relevant guidelines and regulations. All participants and their legal guardians provided informed consent.

Results and discussion

Reliability and validity test

This study employed exploratory factor analysis (EFA) to determine the weights of each item, examine the correlations between indicators, and select the appropriate number of items or analysis indicators for simplification. To assess the suitability of the original items for factor analysis, we conducted Bartlett’s test of sphericity and the Kaiser–Meyer–Olkin (KMO) test using SPSS 27.0³⁴. The results indicated that the KMO value was 0.791, which exceeds the critical threshold of 0.7, suggesting that the data were suitable for factor analysis. Additionally, Bartlett’s test yielded a p-value of 0, further confirming the appropriateness of the data for factor analysis.

Following the scale validity testing, SPSS 27.0 was used to extract common factors and assess the reliability of each latent variable. Items with factor loadings below 0.5 were removed, and the principal component factor loadings were re-extracted to produce the final factor analysis results for each indicator. The Cronbach’s α coefficient for the sample is presented in Table 5. All variables showed a reliability score above 0.6, indicating acceptable internal consistency. The overall reliability of the scale was calculated to be 0.816, meeting the standard for satisfactory reliability.

Table 5.

Cronbach’s Alpha coefficients for individual and situational variables.

Variable name	Number	α	AVE	CR
WTR	5	0.800	0.513	0.801
TEU	4	0.776	0.507	0.753
PC	2	0.838	0.679	0.808
ER	2	0.815	0.591	0.739
PR	3	0.709	0.560	0.718
EC	3	0.812	0.551	0.784
IC	6	0.939	0.759	0.946
EU	5	0.802	0.530	0.818
PCE	5	0.892	0.539	0.852
SE	4	0.726	0.524	0.724
GP	2	0.869	0.602	0.751

Descriptive statistical analysis

Table 6 provides a statistical summary of the socio-demographic characteristics of the sample.

Table 6.

Descriptive statistics of objective variables in the survey sample.

Demographic variables	Category	Number	Percentage (%)
Gender	Male	170	40.3
	Female	251	59.7
Age	18–30 years	61	14.5
	30–45 years	217	51.5
	45–60 years	122	29
	60 years old and above	21	5
Education	Primary school or below	16	3.8
	Junior high school	246	58.4
	College degree	151	35.9
	Graduate degree	8	1.9
Household size	1–3 people	182	43.3
	4–6 people	234	55.6
	7 people and above	5	1.1
Duration of residence	1–5 days	192	45.6
	6–15 days	48	11.4
	16–25 days	13	3.1
	26–30 days	168	39.9
Residential area	Below 100 m2	52	12.4
	100–200 m2	238	56.6
	200–500 m2	114	27
	Over 500 m2	17	4
Residential floors	1 floor	84	19.9
	2 floors	303	71.9
	3 floors and above	34	8.2

We randomly selected household heads with diverse socio-economic profiles, including variables such as education level, household size, days spent at home, housing area, and number of floors. The sample consists of approximately 59.7% females. A significant proportion of respondents are between the ages of 30 and 45, representing 51.5% of the sample. In terms of education, most respondents have completed middle school or high school (58.4%), while about 42% hold a bachelor’s degree, highlighting the importance of education in Hai’an. Regarding family size, the majority of respondents live in households with 4–6 people (55.6%), followed by 43.3% living in smaller households of 1–3 people. The length of residence varies, with 45.6% of respondents spending limited time at home, while 39.9% are home almost every day. In terms of housing, most respondents live in two-story homes, comprising 71.9% of the sample.

The average values of the individual and situational variables for the sample are shown in Table 7.

Table 7.

Mean value of the variable for residents of near and far suburbs.

Variable name	Near suburban		Far suburban		All
	Mean	SD	Mean	SD	Mean	SD
WTR	3.65	0.6	3.41	0.69	3.48	0.67
TEU	3.2	0.91	3.45	0.63	3.38	0.73
PC	3.61	0.75	3.75	0.92	3.7	0.88
ER	3.72	0.79	3.51	0.79	3.58	0.8
PR	3.69	0.68	3.65	0.7	3.66	0.69
EC	3.76	0.75	3.5	0.64	3.59	0.69
IC	2.64	0.89	2.54	0.764	2.58	0.81
EU	3.31	0.51	3.79	0.64	3.64	0.6
PCE	3.62	0.82	3.33	0.76	3.42	0.79
SE	3.68	0.75	3.58	0.62	3.61	0.66
GP	3.7	0.75	3.94	0.79	3.86	0.78

Far suburban residents generally exhibit higher values in TEU, PC, EU, and GP compared to near suburban residents. Specifically, for TEU and EU, the energy usage of far suburban residents is significantly lower than that of near suburban residents, indicating that near suburban residents have higher energy consumption levels. This also suggests that far suburban residents, compared to their near suburban counterparts, are more inclined to use traditional energy sources such as wood and coal.

In terms of PC, far suburban residences are often equipped with farmland and engage in agricultural production, making far suburban residents more attuned to climate conditions, as the success or failure of their agricultural activities is closely tied to the weather. Regarding GP, far suburban residents tend to place a higher value on social interactions with neighbors or family members. Many are accustomed to gathering and socializing outdoors, and these group activities often result in their energy usage being significantly influenced by collective behavior and social norms. In this social context, far suburban residents are likely to adjust their energy usage habits based on group behavior and communal values, making their energy consumption more strongly shaped by the influence of the community.

Pearson related analysis

Table 8 presents the results of the Pearson correlation analysis, which are explained as follows:

Table 8.

Standardized regression coefficients of subjective factors and energy saving potential of rural dwellings in central Jiangsu province.

		WTR	TEU	PC	ER	PR	EC	IC	EU	PCE	SE	GP
WTR	Near suburban	1.0	0.452**	0.672**	0.542**	0.651**	0.704**	− 0.208*	0.415**	0.603**	0.791**	0.697**
	Far suburban	1.0	0.298**	0.163**	0.374**	0.322	0.492	0.062	-0.082	0.337**	0.564**	0.325
	All	1.0	0.362**	0.295**	0.432**	0.411**	0.568**	− 0.014	0.039	0.432**	0.632**	0.417**
TEU	Near suburban	0.452**	1.0	0.588**	0.7**	0.668**	0.378	0.036	0.08	0.353**	0.601**	0.564**
	Far suburban	0.298**	1.0	0.054	0.318**	0.262**	0.26**	0.284**	-0.003	0.284**	0.264**	0.194**
	All	0.362**	1.0	0.238**	0.47**	0.409	0.328**	0.182**	0.026	0.328**	0.414**	0.328**

**Significantly related at the 0.01 level (bilateral)

*Significantly related at the 0.05 level (bilateral)

All correlations are significant at the 0.01 and 0.05 levels. PC shows a weak correlation with TEU for far suburban residents. PR shows a weak correlation with both WTR and TEU for far suburban residents. EC shows a weak correlation with WTR for far suburban residents and a weak correlation with TEU for near suburban residents. IC is significantly negatively correlated with WTR for near suburban residents, strongly correlated with TEU for all residents, and weakly correlated with TEU for near suburban residents. EU is strongly correlated with WTR for far suburban residents and negatively correlated with both WTR and TEU for far suburban residents. GP shows a weak correlation with WTR for far suburban residents.

Multiple linear regression analysis

Table 9 shows the summary information for the individual and situational factors in the multiple linear regression analysis. The Durbin-Watson values are all within the range of 1.5–2.5, indicating that the independence between the samples is satisfactory.

Table 9.

Summary of multiple linear regression for individual and situational factors.

		R2	Adjusted R2	Standard error of regression	Durbin–Watson
WTR	Near suburban	0.73	0.70	0.32	1.67
	Far suburban	0.44	0.42	0.52	2.02
	All	0.5	0.5	0.47	1.96
TEU	Near suburban	0.76	0.74	0.46	2.16
	Far suburban	0.25	0.33	0.55	1.72
	All	0.38	0.46	0.58	1.92

Table 10 presents a multiple regression analysis of the individual and contextual factors influencing the energy-saving potential of rural households in the Jiangsu central region. The non-bold numbers represent the standard error, while the bold numbers represent the standardized regression coefficients. The interpretation is as follows:

1. Willingness to retrofit homes for energy saving.

Table 10.

Standardized regression coefficients and standard errors for individual and situational factors.

		WTR	TEU	PC	ER	PR	EC	IC	EU	PCE	SE	GP
WTR	Near suburban		0.229*	0.397**	− 0.271*	− 0.603**	0.3	− 0.073	0.17**	0.058	0.558**	0.179
			0.06	0.1	0.09	0.16	0.17	0.04	0.07	0.13	0.09	0.11
	Far suburban		0.144**	− 0.03	0.122*	− 0.358*	− 0.566**	− 0.05	− 0.068	− 0.203	0.324**	0.25*
			0.05	0.04	0.05	0.16	0.12	0.043	0.058	0.1	0.06	0.11
	All		0.143**	− 0.003	0.092	− 0.419**	− 0.633**	− 0.052	− 0.11	− − 0.208*	0.347*	0.23*
			0.04	0.04	0.04	0.11	0.1	0.03	0.04	0.08	0.05	0.08
TEU	Near suburban	0.2*		-0.386**	0.653**	1.2**	− 0.623**	0.158**	0.045	− 0.057	0.077	− 0.287*
		0.13		0.15	0.12	0.19	0.22	0.05	0.1	0.18	0.15	0.16
	Far suburban	0.192*		− 0.211**	0.289**	0.45*	− 0.321*	0.206**	− 0.143*	0.202	0.025	− 0.19
		0.06		0.04	0.05	0.17	0.14	0.05	0.06	0.1	0.07	0.12
	All	0.177**		− 0.151**	0.365**	0.783**	− 0.393**	0.132**	− 0.096*	0.085	0.153**	− 0.372**
		0.06		0.04	0.05	0.13	0.13	0.03	0.05	0.1	0.06	0.1

**Significantly related at the 0.01 level (bilateral)

*Significantly related at the 0.05 level (bilateral)

For all resident groups, TEU, PR, EC, PCE, SE, and GP all show significant effects on WTR, with H1a, H2a, H3a, and H9a being rejected. Among these, PR and PCE have a negative impact on WTR, indicating that policy regulations and awareness of clean energy do not enhance residents’ interest in energy-saving renovations. Therefore, H4a and H7a are also rejected. Further analysis reveals that near suburban residents increase their WTR as PC improves, while far suburban residents are not affected by this factor. This suggests that climate concern has a positive influence on the willingness to undertake energy-saving renovations among near suburban residents, but this effect is not significant for far suburban residents. Additionally, GP has a more significant impact on far suburban residents, indicating that their decisions regarding energy-saving renovations are more influenced by group communication, which is consistent with previous research findings. Finally, H5a, H6a, and H8a are accepted.

• (2)Willingness to retrofit homes for energy saving

Except for PCE, other variables significantly affect TEU, leading to the rejection of H4b. Specifically, PC, EC, EU, and GP all have a negative impact on TEU. The negative effects of PC, EC, and EU suggest that as these factors increase, the use of traditional energy by residents decreases. This implies that residents may adopt more energy-efficient lifestyles or equipment. Based on this, H1b, H6b, and H9b are rejected. For far suburban residents, increased energy use encourages them to gradually reduce their reliance on traditional energy and shift toward more energy-efficient equipment or clean energy. This aligns with previous research findings on energy-saving behaviors in far suburban areas. In contrast, near suburban residents rarely exhibit herd mentality, and their energy consumption behaviors are less influenced by the group. Specifically, near suburban residents do not increase their use of traditional energy as GP increases, indicating that their energy choices are more dependent on personal needs and habits, and less affected by direct social group influence. Finally, H2b, H3b, H5b, H7b and H8b are accepted.

Table 11 shows the summary information for the Sociodemographic factors in the multiple linear regression analysis. Among them, the Durbin-Watson value for near suburban residents exceeds 2.8, which may be due to the more uniform housing types among near suburban residents, leading to similar responses in the data.

Table 11.

Summary of multiple linear regression for sociodemographic factors.

		R2	Adjusted R2	Standard error of regression	Durbin-Watson
WTR	Near suburban	0.24	0.2	0.54	2.36
	Far suburban	0.14	0.12	0.64	2.07
	All	0.1	0.08	0.64	2.02
TEU	Near suburban	0.26	0.22	0.8	2.8
	Far suburban	0.12	0.1	0.65	1.8
	All	0.14	0.13	0.72	2.29

Table 12 presents the multiple regression analysis of sociodemographic factors affecting the energy-saving potential of rural residences in the Jiangsu central region. The non-bold numbers represent the standardized regression coefficients, while the bold numbers represent the standard error. The interpretation is as follows:

Table 12.

Standardized regression coefficients and standard errors of sociodemographic factors.

		Age	Education	Household size	Duration of residence	Residential area	Residential floors
WTR	Near suburban	− 0.023	0.515**	− 0.226*	0.358**	− 0.2	− 0.02
		0.09	0.11	0.11	0.05	0.1	0.09
	Far suburban	0.243**	0.25**	0.28**	− 0.143*	0.063	− 0.102
		0.05	0.08	0.08	0.03	0.06	0.08
	All	0.083	0.324**	0.109**	− 0.032	− 0.005	− 0.049
		0.04	0.06	0.06	0.02	0.05	0.07
TEU	Near suburban	0.261**	0.439**	− 0.024	0.089	− 0.35**	− 0.211*
		0.13	0.17	0.17	0.07	0.15	0.14
	Far suburban	− 0.01	− 0.121	0.026	− 0.056	− 0.045	− 0.003
		0.06	0.08	0.08	0.03	0.06	0.08
	All	0.043	0.036	− 0.006	− 0.046	− 0.106*	− 0.108*
		0.05	0.07	0.08	0.03	0.06	0.08

**Significantly related at the 0.01 level (bilateral)

*Significantly related at the 0.05 level (bilateral)

1. Willingness to retrofit homes for energy saving.

Education and household size have a significant positive impact on WTR, while residential area and number of floors do not significantly affect WTR. For far suburban residents, a larger household size tends to promote energy-saving renovations. This may be related to far suburban residents’ increased focus on improving family quality of life and energy efficiency when managing larger households. In contrast, near suburban residents exhibit the opposite trend, where a larger household size is less likely to encourage energy-saving renovations. This may be due to differences in economic conditions and housing situations in near suburban areas. Additionally, the duration of residence has a varying effect on WTR. For near suburban residents, a longer duration of residence is associated with a stronger willingness to undertake energy-saving renovations, likely because they are more familiar with their homes and are more motivated to improve their living conditions. However, for far suburban residents, an increase in duration of residence negatively affects their willingness to undertake energy-saving renovations.

• (2)Willingness to retrofit homes for energy saving.

Residential area and residential floors have a significant negative impact on WTR, indicating that residents with larger homes and more floors are more likely to abandon the use of traditional energy. This phenomenon may arise from the fact that larger residences have more complex energy demands for daily activities such as heating and cooking, making it inconvenient to rely on traditional energy sources. Due to the inefficiency of traditional energy use in large residences and the operational inconveniences, residents are generally more inclined to adopt clean energy devices (such as solar energy, air-source heat pumps, etc.) to improve energy efficiency, thereby achieving energy savings and reducing energy consumption.

Robustness check

To test the robustness of the results, we conducted regression analysis by modifying some core independent variables. The specific modifications are as follows: In PCE, the statement “I think the price of clean energy equipment is acceptable” was replaced with “There are more scenarios where electricity is used.” In SE, the statement "I am more willing to cook with electricity” was replaced with “Using clean energy reduces environmental pollution.” The results are shown in Table 13.

Table 13.

Robustness check.

		WTR	TEU	PC	ER	PR	EC	IC	EU	PCE	SE	GP
WTR	Near suburban		0.238*	0.46**	− 0.397**	− 0.62**	0.145	− 0.04	0.213**	0.267*	0.428**	0.322*
	Far suburban		0.108*	− 0.114	0.119	− 0.239	0.305**	− 0.09	− 0.221**	0.113	0.36**	0.154
	All		0.098*	− 0.058	0.086	− 0.378**	0.323**	− 0.084*	− 0.115**	0.171*	0.384**	0.211*
TEU	Near suburban	0.216*		− 0.417**	0.67**	1.311**	− 0.147	0.123*	− 0.004	− 0.5**	0.044	− 0.421**
	Far suburban	0.135*		− 0.215**	0.266**	0.391*	0.015	0.215	− 0.11	− 0.019	0.053	− 0.24
	All	0.111*		− 0.191**	0.344**	0.766**	0.101	0.131**	− 0.107*	− 0.207**	0.149*	− 0.485**

**Significantly related at the 0.01 level (bilateral)

*Significantly related at the 0.05 level (bilateral)

After replacing the core independent variables, the impact of individual and situational factors on the energy-saving potential of residences remained largely consistent with the baseline model. Therefore, the results of this study’s model are robust.

Conclusion and policy implications

Conclusion

1. The willingness to undertake energy-saving renovations in rural residences in the Jiangsu central region is influenced by multiple factors.

TEU, PR, and SE significantly influence the WTR of both near suburban and far suburban residents, while IC does not have a significant impact on either group. PC and EU significantly affect the WTR of near suburban residents but have little effect on far suburban residents. This suggests that near suburban residents are more attuned to weather changes and residential energy efficiency, possibly due to their proximity to urban environments and better access to information, which enables them to actively monitor these factors and take appropriate energy-saving measures. In contrast, far suburban residents exhibit relatively lower awareness, potentially constrained by limited information flow and living conditions.EC and GP significantly negatively influence the WTR of far suburban residents, but have little impact on near suburban residents. Some studies have shown that economic costs are an important factor in energy-saving behaviors⁶⁶, suggesting that far suburban residents are more likely to consider economic costs and neighborhood group interactions when deciding to undertake energy-saving renovations. Environmental responsibility can also motivate energy-saving actions³⁰. Interestingly, environmental responsibility significantly negatively impacts the WTR of near suburban residents but has a positive significant effect on far suburban residents, indicating that far suburban residents are more concerned about environmental issues than their near suburban counterparts.

• (2)The use of traditional energy in rural areas of the Jiangsu central region is influenced by multiple factors.

Except for EU, PCE, SE, and GP, all other factors significantly influence the TEU of both near suburban and far suburban residents, with PCE having no significant impact on either group. PC and EC significantly negatively influence the TEU of all residents, indicating that far suburban residents are more concerned about economic costs and climate change, which drives them to reduce energy consumption. EU and GP also have a significant negative impact on the TEU of all residents, suggesting that as residents’ energy usage habits evolve, clean energy gradually replaces traditional energy sources. Rural residents tend to maintain a certain level of independence in their energy usage, especially when their consumption increases. They are more likely to opt for more efficient energy equipment and clean energy solutions. Photovoltaic (PV) technology, as a renewable energy source, has become a crucial means of meeting the clean energy needs in rural areas^29,67.

Regarding sociodemographic factors, residential area and residential floors have a significant negative impact on the TEU of all residents. This suggests that larger residential areas cannot meet energy needs with traditional energy sources, leading residents to prefer using clean energy devices such as air conditioners and heaters to maintain comfort. The demand for larger living spaces forces residents to choose more efficient and energy-saving equipment to reduce the use of traditional energy, which aligns with other studies: larger family sizes tend to lead to the purchase of energy-saving appliances⁶⁸.

• (3)There are regional differences in the energy-saving potential of rural residences in the Jiangsu central region.

The influence of different factors on far suburban and near suburban residents may be entirely opposite. For example, PC, ER, PR, EU, and PCE have opposing effects on residents’ WTR, while EU and PCE have opposite effects on residents’ TEU. Although both far suburban and near suburban residents live in rural areas, the narrowing of the urban–rural gap—especially in regions like Jiangsu Province, where the urban–rural disparity is significant—has led to near suburban residents’ lifestyles and work habits gradually shifting toward urbanization. As a result, their consumption habits, energy usage behaviors, and energy-saving awareness are becoming increasingly similar to those of urban residents.

Regarding WTR, near suburban residents are most influenced by PR and SE, while far suburban residents are most influenced by EC, PR, and SE. For all residents, the greatest influences are EC, PR, and SE. Therefore, in the Jiangsu central rural areas, promoting economic policies (such as subsidies for energy-saving equipment, preferential loans, etc.) and cultivating residents’ low-carbon energy-saving awareness are crucial. Research has shown that increasing environmental protection awareness can effectively enhance household energy conservation⁶⁹. By strengthening the promotion of energy-saving policies, improving residents’ environmental consciousness, and providing economic incentives, the process of rural residential energy-saving renovation can be effectively accelerated.

Regarding TEU, near suburban and far suburban residents are most influenced by PC, ER, and EC, while all residents are most influenced by PR, EC, and GP. The government needs to vigorously promote clean energy policies and regulations, as well as carry out clean energy education activities, especially at the village level, to ensure that residents better understand the long-term benefits of energy conservation and emission reduction. Through this approach, a collective action atmosphere can be created, fostering mutual learning and encouragement among villagers, and collectively promoting the use of clean energy.

Policy implications

First, the government should increase subsidies to encourage energy-saving behaviors among residents. Although most farmers have adopted energy-efficient lighting, very few have purchased energy-efficient appliances. Economic factors are a major barrier. Therefore, the government should enhance subsidies for energy-saving products, making them more affordable for a wider range of people.

Second, the government should implement flexible and diverse energy policies. Given that rural households in different regions have varying levels of concern about energy issues, and that changing energy consumption habits takes time, the government should develop long-term strategies tailored to different regions, gradually guiding rural households toward the use of cleaner and more efficient energy.

Third, the government should conduct energy-saving education activities tailored to local conditions. In remote areas with limited information, the government should launch campaigns to promote clean energy, improving residents’ understanding and acceptance of clean energy, and encouraging them to change their energy consumption habits. In suburban areas, the government should encourage rural households to adopt new energy-saving technologies and equipment, stimulating their interest in home energy-saving renovations. This will improve the overall infrastructure and facilitate the transition to more energy-efficient and environmentally friendly consumption patterns.

Abbreviations

WTR

Willingness to retrofit homes for energy saving

TEU

Traditional energy use habits

PC

Perception of climate

ER

Environmental responsibility

PR

Policies and regulations

EC

Economic costs

IC

Indoor comfort

EU

Energy use

PCE

Perception of clean energy

SE

Willingness to save energy with low carbon

GP

Group psychology

Author contributions

C.X. wrote the main manuscript text, R.Z. prepared the figures, Y.H. processed the data, and G.Y. and Z.D. reviewed the content. All authors reviewed the manuscript.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.