Group Heterogeneity of Rural Households’ Satisfaction with Good Life from the Perspective of Rural Revitalization—A Case Study from Zhejiang Province of China

Abstract

In the context of rural revitalization, this article explores the group heterogeneity and influencing factors of the satisfaction of rural households’ good life in five dimensions, including industry and economy, living environment, rural culture, rural governance, and material life, and provides micro evidence and policy suggestions for the specific strategies of deepening rural revitalization and improving the satisfaction level of rural households’ good life. Based on the field survey data in rural areas of Zhejiang Province, this article analyzes the group heterogeneity of rural households’ satisfaction with a good life through latent profile analysis and further reveals the influencing factors through multinomial logistic regression. The results indicate that most dimensions of rural households’ satisfaction with good life, except rural culture in Zhejiang, are higher than average. The rural households’ satisfaction with good life presents obvious group differences, which are aggregated into three latent classes: very-satisfied class, medium-satisfied class, and low-satisfied class. The results of multinomial logistic regression analysis showed that subjective psychological evaluation, such as communication, self-confidence, and village status evaluation, had significant positive effects on the group category of good life satisfaction. The proportion of non-agricultural income and the level of per-capita family income have a significant negative influence on the group category of good life satisfaction, and the condition of public service at the village level has a significant positive influence on the group category of good life satisfaction. The satisfaction of rural households on sanitary toilet environment, New Year celebration, cultural activities in the countryside, work of village committee, fair treatment of village affairs, and soliciting opinions of village affairs were not high. Lack of self-confidence in communication and other subjective psychological factors and poor public service conditions at the village level are the important reasons for this. Therefore, measures should be taken to deepen the construction of rural civilization, build a multi-path and long-term mechanism to increase farmers’ income, and make up for the shortage of rural infrastructure and public services.

1. Introduction

Over the past forty years of reform and opening up, China has entered a period of rapid development. However, the dual structure of urban and rural areas and the imbalance between urban and rural areas restrict the sustainable development of rural areas. During the new period, farmers have higher expectations for a higher quality of life, suitable living environments, and perfect supporting facilities, and look forward to having beautiful and livable villages [1]. In 2017, the report of the 19th National Congress of the Communist Party of China proposed the strategy of revitalizing the countryside for the first time, pointing out that the implementation of the strategy of revitalizing the countryside is not only an inevitable choice to achieve common prosperity for all people, but also an inevitable requirement to solve the current major social contradictions. The Law of the People’s Republic of China on the Promotion of Rural Revitalization, which was adopted in 2021, provides a legal guarantee for the implementation of the rural revitalization strategy. It points out that the primary task is to promote the well-being of farmers through the five major revitalizations of industry, talent, culture, ecology, and organization. The primary goal is to achieve the overall requirements of a prosperous industry, pleasant ecology, civilized countryside, effective governance, and prosperous life. The overall requirement of prosperous life is to respond to the expectation of farmers’ growing needs for a good life, to improve the farmers’ sense of access, happiness, and security.

The concept of rural revitalization is relative to the decline of rural areas. Developed countries such as Europe and the United States have taken a series of measures to narrow the gap between urban and rural areas and promote rural development in order to alleviate rural declines. For example, there is the measure of “New Town Construction” in the United States, the “New Village Movement” in South Korea, and the “Rural Revitalization Movement” in Japan [2,3]. In China, rural revitalization mainly refers to the comprehensive revitalization of agriculture, rural areas, and farmers, and ultimately realizes the wishes of a good life for a strong agriculture, beautiful countryside, and rich farmers [4]. At present, there are many articles on rural revitalization, but most of them focus on macro perspectives such as policy suggestions, rural industrial development, and institutional arrangements for revitalization [5,6,7]. Existing studies have paid little attention to the life satisfaction of micro-groups of farmers under the perspective of rural revitalization. Wang, Zhu, and Yu [1] explored the examination of residents’ satisfaction with rural livability and possible disadvantages under the perspective of rural revitalization. The ultimate goal of implementing a rural revitalization strategy is to realize the good life of farmers. Therefore, it is of great theoretical and practical significance to study the satisfaction of farmers’ good life from the perspective of rural revitalization.

In the context of rural revitalization, what are the possible variations in satisfaction with the good life among rural households? How do they differ and what are the influencing factors? To address these questions, the article examines the group heterogeneity characteristics of farmers’ satisfaction with good life based on the perspective of rural revitalization, using potential profile analysis. It considers the influencing factors of different satisfaction groups to reveal the reasons for the formation of different satisfaction levels. Considering farmers’ satisfaction as a latent variable and classifying groups by the response pattern of individually observed indicators, this method has advantages over cluster analysis in terms of class determination rules and accuracy of classification results. In addition, by comparing the characteristics and the influencing factors of different classes, it is beneficial to explain the formation of farmers’ satisfaction with their good life.

Although there are many debates on what constitutes a “good life”, scholars agree that well-being is a complex and multidimensional concept [8]. Philosophers and psychologists describe the good life as consisting of authentic expression of self, a sense of well-being, and active engagement in life and work [9]. In the literature, the concepts of the good life, life satisfaction, well-being, and quality of life tend to be used interchangeably [9,10]. Haas [11] believes that quality of life is a subjective sense of well-being encompassing physical, psychological, social, and spiritual dimensions. Cummins et al. [12] argued that satisfaction with seven domains of life adequately represents subjective well-being [13]. Chen et al. [14] consider that seven domains represent life satisfaction: career, housing situation, living environment, community life, economic situation, location, and facilities. Duboz, Macia, Diallo, Cohen, Bergouignan, and Seck [10] found through research in rural areas that all individuals agree on two dimensions that are essential to a good life: health and being able to meet material needs. Contrasting these results, the major dimensions of a good life usually include physical health, social relationships, and material living conditions. Since the 19th National Congress of the Communist Party of China pointed out that an important aspect of the transformation of the main contradiction in society is the people’s growing need for a good life, more and more studies have been conducted to understand a good life from the perspective of the need for a good life. The need for a good life is people’s basic requirements and yearning for the external living environment, which is historically rational, no longer isolated by a single person, but concrete and objective [15]. Wang et al. [16] divided the connotation of the need for a good life into three levels, including the individual material level, the family relationship level, and the national social level. Wu and Ji [17] believe that “the need for a good life” is a kind of people-centered, “life needs” as the foothold, and it is a combination of “good” quality life and “beautiful” quality life based on quantity. Yin et al. [18] believe that what people need for a better life in the new era should include material needs of abundance, diverse social needs, and rich psychological needs.

The life satisfaction of rural households is of major concern. In Đerčan et al. [19], based on the five dimensions of health, education, culture, sports, and community, the quality of life of local farmers was investigated. Li et al. [20] used the ordered Probit model to examine the differences in life satisfaction of farmers and found that the life satisfaction of poor households was generally lower than that of middle and wealthy households. Xiong et al. [21] used factor analysis to measure farmers’ happiness index from seven dimensions, including political environment, material security, social environment, physical and mental health, home environment, education and sports, and happiness and confidence. Zhao and Mao [22] used an ordered logistic regression model to explore the factors affecting the life satisfaction of farmers in the western minority areas and found that family income and income and expenditure balance, relative living standards, anti-risk ability, and social trust, among others, contributed to farmers’ life satisfaction. These factors have a significant positive impact [22]. In addition, personal characteristics [23,24], family and interpersonal relationships [25,26,27], and country and social environment [28,29] also have a significant impact on life satisfaction.

In summary, there are more existing studies on farmers’ life satisfaction, but there is a limited examination of the farmers’ satisfaction with their good life from the perspective of rural revitalization. Then, in terms of methodology, the existing literature often uses factor analysis to measure farmers’ life satisfaction. Since it is assumed that all individuals in the sample are from the same homogeneous group, factor analysis focuses on the classification of test items and ignores the group differences. Otherwise, latent profile analysis is more suitable for studying group differences. To be specific, this study focuses on the following three objectives: (1) based on the existing studies and combined with the rural revitalization strategy, this article proposes to examine the satisfaction of farmers’ good life in five dimensions, including industry and economy, living environment, rural culture, rural governance, and material life; (2) this article uses the latent profile analysis (LPA) to explore the heterogeneity of rural households’ life satisfaction under the rural revitalization policy; (3) this article discusses the differences and influencing factors on different life satisfaction classes and provides recommendations for future government efforts.

2. Materials and Methods

2.1. Study Area

The study area of this article is Zhejiang Province (27°02′–31°11′ N, 118°01′–123°10′ E) located on the southeast coast of China, which is the southern region of the Yangtze River Delta (Figure 1). The total area is 105,500 square kilometers. Zhejiang Province has a complex topography and varied landscape types, with the terrain sloping from southwest to northeast. Mountainous and hilly areas account for more than 70% of the total area while the plains and basins account for 23.2%. Zhejiang Province is famous for fish and rice because of its moderate annual temperature and abundant rainfall, which belong to a humid subtropical climate. In 2020, the gross domestic product (GDP) of Zhejiang Province was CNY 6.46 trillion, approximately USD 936.4 billion. It accounts for 6.4% of mainland China’s total GDP, and ranks fourth among all provinces. From the perspective of urban and rural structure, the total resident population of Zhejiang Province is 64,567,600 in 2020, while the rural population is 17,969,100, accounting for 27.8%. In 2020, the annual disposable income of rural residents in Zhejiang Province is CNY 31,930, exceeding the national average and ranking second in all provinces in China [30,31]. In 2021, the Central Committee of the Communist Party of China and the State Council will support Zhejiang’s high-quality development and construction of a demonstration area for common prosperity and create a high-quality demonstration province for rural revitalization [32].

2.2. Data Source

Stratified sampling was conducted to select participants. Firstly, considering the differences in economic development in different regions of Zhejiang Province, this study divided it into high-income regions, middle-income regions, and general-income regions according to the average annual income level of urban and rural residents. Two counties were randomly selected from each region, and 25–30 rural households were surveyed in each country. After random selection, rural households from 14 counties (Figure 2) in Lishui, Quzhou, Hangzhou, Jiaxing, Taizhou, Ningbo, and Jinhua in Zhejiang Province participated in surveys. These farm households have strong typicality and representativeness as the rural areas in the rural revitalization demonstration province (Zhe Jiang) [33].

The survey includes two stages. Stage 1: Design the questionnaire according to the research purpose. Stage 2: Issue the questionnaire from July to September 2020. Six trained postgraduates from China Jiliang University served as investigators. Semistructured interviews were conducted by investigators after obtaining the informed consent of the heads of households. The overwhelming majority of interviewees were composed of heads of households or spouses of the heads of households. Each interview lasted more than 20 min. A total of 396 families received the survey, and 367 families completed the survey and could be further analyzed effectively (effective response rate: 92.6%). The characteristics of the survey data are as follows: (1) Gender distribution: there were 349 men, accounting for 95.10% of the total; there were 18 women, accounting for 4.90% of the total. (2) Age distribution: there were 58 people aged 30–45, accounting for 15.80% of the total; 202 people aged 46 to 60, accounting for 55.04% of the total; and 107 people over the age of 61, accounting for 29.16% of the total. (3) Education distribution: there were 156 people with primary school education and below, accounting for 42.51% of the total; 159 people with junior middle school education, accounting for 42.32% of the total; and 52 people with high school or above, accounting for 14.17% of the total. (4) Ethnic minorities distribution: 360 of the Han nationality, accounting for 98.09% of the total; 7 ethnic minorities, accounting for 1.91% of the total. (5) Political appearance distribution: there were 314 members of the masses and other party members, accounting for 85.56% of the total; 53 members of the Communist Party of China, accounting for 14.44% of the total. (6) Annual household income distribution: 87 households earned below CNY 60,000, accounting for 23.71% of the total; 74 households with CNY 60,000 to 100,000, accounting for 20.16% of the total; 57 households with CNY 100,000 to 150,000, accounting for 15.53% of the total; 63 households with CNY 150,000 to 200,000, accounting for 17.77% of the total; and 86 households above CNY 200,000, accounting for 23.48% of the total.

2.3. Research Tools

Combined with the five guidelines of rural revitalization, the questionnaire was designed for rural households in five aspects, including industry and economy, living environment, rural culture, rural governance, and material life. The questionnaire includes three industrial economy entries, four living environment entries, three rural culture entries, three rural governance entries, and four material life entries. The questionnaire was scored using the Likert 5-point scale, with 1 indicating very dissatisfied/no, matching number = 1, 2 indicating dissatisfied, matching number = 2, 3 indicating fair, matching number = 3, 4 indicating satisfied, matching number = 4, and 5 indicating very satisfied, matching number = 5. The statistical results show that the questionnaire has good reliability and good overall fit. The specific indices are as follows: (${\chi }^2$ = 339.213, CFI = 0.938, TLI = 0.926, RMSEA = 0.073, SRMR = 0.077, internal consistency reliability Cronbach’s $\mathsf{\alpha }$ = 0.810).

2.4. Research Method

2.4.1. Latent Profile Analysis

Latent profile analysis (LPA), similar to cluster analysis, is an extension of latent class analysis (LCA) on continuous explicit variables, aiming to explore the heterogeneous classification situation within groups [34]. However, unlike clustering analysis, latent profile analysis (LPA) is a probabilistic and model-based approach [35].

Latent class analysis (LCA) is a statistical method that explains the association of indicators through category latent variables and subsequently maintains their local independence.

$$P(y_i)={\displaystyle \sum _{k=1}^{K}P(c_i=k)}P(y_i=1|c_i=k)(i=1,2,3,L,k)$$

(1)

In Equation (1), $y_i$ denotes the score of the individual $i$ for the two options $y=1$ or $y=0$ of the indicator; $P(c_i=k)$ represents the proportion of a certain class $k$ to the population, also known as latent category probability; $P(y_i=1|c_i=k)$ represents the conditional probability; $c$ represents a categorical latent variable; $k$ represents a class.

When the observed indicators are continuous variables, LCA is called latent profile analysis (LPA). The principle and procedure are the same as traditional LCA, but the difference is that the probability distribution is expanded into a density distribution. Then, the formula becomes

$$P(y_i)={\displaystyle \sum _{k=1}^{K}P(c_i=k)f(y_i=1|c_i=k)}$$

(2)

The selection of the latent profile model is based on the fitting statistics, and the selection between different classes is based on AIC, BIC, aBIC, entropy, LMR, and BLRT. When the values of AIC, BIC, and aBIC are smaller, the fitting result is better. When the entropy is closer to 1, the classification result is better. LMR and BLRT are mainly used for the fitting difference between k−1 and k classes. Significant LMR and BLRT indicate that the k-th class model is better than the k−1 class model. Retaining the model of the correct class is still an issue under discussion. When the optimal model displayed by each indicator is inconsistent, a comprehensive judgment is required and a suitable one is selected. The interpretability of the model should also be considered when selecting classes [36].

2.4.2. Multinomial Logistic Regression

The heterogeneity of farmers’ satisfaction with a good life was analyzed through latent profiles. The result was discrete data dominated by categorical data. When analyzing discrete choice problems, the use of probability models (logistic model) is an ideal estimation method [37]. This article divides farmers’ satisfaction with a good life into more than two classes with different satisfaction levels; hence, a multinomial logistic model is used.

The multinomial logistic model is a simple extension of the binary logistic model, allowing multiple types of dependent variables [38]. It can be regarded as a joint estimation of multiple binary logistic models formed by pairing each type of the explained variables.

$$logit(P(Y=j))=ln[\frac{P(Y=j)}{P(Y=i)}]={\alpha }_i+{\beta }_{1j}{X}_1+\dots +{\beta }_{pj}{X}_j+\mu (j\ne i)$$

(3)

In Equation (3), $logit(P(Y=j))$ is the dependent variable, indicating the class of good life satisfaction to which the farmer belongs, $X_j$ is the explanatory variable, indicating each influencing factor affecting the category of satisfaction; ${\alpha }_i$ indicates the model intercept; ${\beta }_{pj}$ indicates the corresponding regression coefficient; $j$ is the number of explanatory variables; $\mu$ is the random residue.

The explanatory variables include the age of the head of the household, the number of years of education of the head of the household, the self-confidence in communication, the status in the village, whether he is engaged in government occupation, and whether he is a party member, income per capita, and infrastructure in the village (including schools, hospitals, courier services, banking services, and agricultural material services). The meaning, assignment, and descriptive statistical analysis results of each variable are shown in

Table 1. Descriptive statistics of influencing factors.

	Variables	Variables Definition	Mean/Percentage (%)	Sd
Individual factors	Age	≤45 = 1, 45–60 = 2, >60 = 3	2.99	0.66
	Years of education	≤9 years = 0	85.83
		>9 years = 1	14.17
	Confidence in communication	5 = very confident, 4 = not confident, 3 = average confident, 2 = somewhat confident, 1 = very confident	3.78	0.75
	Village status	Low = 1, Medium = 2, High = 3	2.43	0.056
	Whether in a government career	No = 0	74.39
		Yes = 1	25.61
	Whether you are a party member	No = 0	85.56
		Yes = 1	14.44
Family factors	Family size	≤4 = 1, 4–6 = 2, >6 = 3	1.22	0.47
	Close family contacts	Number of close families	9.34	11.79
	Housing area per capita	House size/Total household size	74.45	51.01
	Share of non-agricultural income	Nonfarm income/Total income	0.59	0.41
	Household income level per capita	Ln(family income per capita)	9.21	1.00
Village characteristics	Public service conditions	Have 1–5 kinds of infrastructure	3.74	1.27

.

3. Results

Mplus 8.3 statistical software was used to perform the latent profile analysis, gradually increasing the number of classes in the model from the initial model, from which the best-fit model was selected based on discriminant indicators. In addition, Stata15.0 statistical software was used for preliminary data collation, descriptive statistical analysis, and multinomial logistic regression analysis.

3.1. Common Method Deviation Test

Since the variables were obtained based on the participants’ self-responses, there might be a common method bias. In this regard, process treatments such as anonymous responses, blurring of the main participants’ responses, and Harman’s one-way test were performed [39]. The results showed that five factors with eigenvalues greater than 1 were obtained after factor analysis, and the variance explained by the first factor was 34.30%, which was less than the critical value of 40%, indicating that there was no obvious common method bias.

3.2. Mean Value Description of Rural Households’ Satisfaction with the Good Life

Table 2. Rural households’ satisfaction with good life (mean ± SD, score).

Item	Score
Industrial Economy	$3.581 \pm$ 0.825
1. Family income satisfaction	$3.538 \pm$ 0.893
2. Satisfaction with household income stability	$3.610 \pm$ 0.870
3. Evaluation of the stability of homeworkers	$3.591 \pm$ 0.882
Living Environment	$3.411 \pm$ 0.736
4. Satisfaction with the village ecological environment	$3.523 \pm$ 0.829
5. Satisfaction with village roads and greening	$3.444 \pm$ 0.837
6. Field waste recycling satisfaction	$3.431 \pm$ 0.830
7. Satisfaction with toilet construction	$3.245 \pm$ 0.938
Rural Culture	$2.984 \pm$ 1.347
8. Whether there are complete fitness facilities	$3.638 \pm$ 1.898
9. Whether to hold a New Year’s celebration	$3.267 \pm$ 1.985
10. Whether to hold cultural activities in the countryside	$2.046 \pm$ 1.760
Rural Governance	$3.321 \pm$ 0.805
11. Satisfied with the work of the village committee	$3.343 \pm$ 0.844
12. The village committee handles village affairs fairly	$3.332 \pm$ 0.826
13. Village affairs solicit public opinion	$3.286 \pm$ 0.873
Material Life	$4.292 \pm$ 0.825
14. Satisfaction with basic education guarantees	$4.499 \pm$ 1.326
15. Whether home appliances are fully equipped (refrigerator, air conditioner or heating, computer, water heater, washing machine)	$4.490 \pm$ 1.042
16. Household savings (emergency savings, New Year’s red envelopes, pension money for elders, control of gift money, debt)	$4.616 \pm$ 0.811
17. Extra consumption after meeting the basic living guarantee (holiday dinners, outings, participating in serious illness medical insurance, participating in old-age insurance, and paying for serious illness at your own expense)	$3.564 \pm$ 1.395

shows the five dimensions of farmers’ satisfaction with good life and the mean for each item. The participants’ satisfaction in each dimension except for rural culture is slightly higher than the median value of 3. The result shows that most dimensions of rural households’ satisfaction with good life in Zhejiang are higher than average. In addition, the satisfaction levels were found to vary across the five dimensions of rural households’ satisfaction with good life. It was determined that the dimension with the highest satisfaction was material life (4.292 ± 0.825 SD). This was followed by the industrial economy (3.581 ± 0.825 SD) and living environment (3.411 ± 0.736 SD). Meanwhile, the dimensions with relatively low satisfaction were found to be rural governance (3.321 ± 0.805 SD) and rural culture (2.984 ± 1.347 SD). Otherwise, the standard deviation data further showed that the rural culture results showed large differences among the different participants.

The mean satisfaction of the items of the rural culture dimension was (in descending order): Whether there are complete fitness facilities (3.638); Whether to hold a New Year’s celebration (3.267); Whether to hold cultural activities in the countryside (2.046). These results indicated that Zhejiang farmers are more satisfied with public fitness facilities, but less satisfied with public cultural facilities. The highest satisfaction level among the evaluation items of the living environment dimension is the satisfaction with the village ecological environment (3.523), and the lowest is the satisfaction with toilet construction (3.245). These results indicate that Zhejiang has a good ecological environment, but the popularity of sanitary toilets is not as good as it should be.

3.3. Latent Profile Analysis of Rural Households’ Satisfaction with Good Life

LPA is a model-based procedure that allows for more flexible model specifications [40]. LPA classifies individuals into classes in which individuals within classes are similar to each other and different from individuals in other classes. An attempt was made to fit a model of 1–4 classes (classifying residents into 1 to 4 classes), and the results are shown in

Table 3. Model fit statistics for 1- to 4-class models. AIC: Akaike information criteria; BIC: Bayesian information criteria; aBIC: adjusted Bayesian information criteria.

Number of Classes	AIC	BIC	aBIC	Entropy	LMR	BLRT
1	18,376.937	18,509.719	18,401.850
2	17,122.045	17,325.124	17,160.148	0.927	0.000 ***	0.000 ***
3	16,666.711	16,940.086	16,718.003	0.944	0.016 ***	0.000 ***
4	15,787.461	16,131.133	15,851.942	1.000	0.636	0.000 ***

*** p < 0.01.

: Akaike information criterion (AIC), Bayesian information criterion (BIC), and adjusted Bayesian information criterion (aBIC) decrease with the increase of the number of classes, and the fitting of the four-class model is better; entropy indicates that the fitting of the four-class model is better. The likelihood ratio test (LMRT) and bootstrap-based likelihood ratio test (BLRT) indicate that the three-class model fits better than the four-class model. When different indicators show inconsistent optimal models, the simplicity and practical significance of the models are considered. The study results led to the selection of the model with three classes as the final model, and the matrix of the attribution probabilities is shown in

Table 4. The average attribution probability of subjects in different latent classes.

Class	Attribution Probability
	C1	C2	C3
C1	0.975	0.025	0.000
C2	0.024	0.976	0.000
C3	0.000	0.005	0.995

, where the average probability of rural households in each class belonging to each class ranged from 97.5% to 99.5%, with the final three classes accounting for 9.6%, 32.4%, and 58%.

The characteristics of the distribution on the three types of rural households’ satisfaction with the good life in the five dimensions were further analyzed to reveal the regularity of the three classes. Figure 3 shows that the mean values of satisfaction with good life in four dimensions of the industrial economy, living environment, rural culture, rural governance, and material life of the third class of households are higher than those of the second class of households. In addition, the satisfaction with material life of the second class of farmers is almost similar to that of the first class of households, but the mean values of satisfaction with good life in the other four dimensions of the second class of households are higher than those of the first class of households. Furthermore, unlike the second class, the first class has a low satisfaction rating for sanitary toilet environment, New Year celebration, cultural visits to the village, village committee work, fair handling of village affairs, and village affairs consultation, with a mean value below 3 points. Finally, combining the regular characteristics of increasing life satisfaction of households between the first class and the third class, the first class was named the low-satisfied class, the second class as the medium-satisfied class, and the third class as the very-satisfied class.

3.4. Multiple Comparisons of Farm Households’ Satisfaction with Good Life among Different Classes

To further verify the heterogeneity of latent classes of households’ satisfaction with good life, multiple comparative analysis of three latent classes on five dimensions of satisfaction with good life was conducted, and the results are shown in

Table 5. Multiple comparisons of rural households’ satisfaction with good life among different classes.

	Class			F Value	Post Hoc Test
	Low-Satisfied Class (C1) (n = 213)	Medium-Satisfied Class (C2) (n = 119)	Very-Satisfied Class (C3) (n = 35)
Industrial Economy	3.30 ± 0.75	3.84 ± 0.73	4.41 ± 0.75	44.097 ***	C1 < C2 < C3
Living Environment	3.03 ± 0.55	3.68 ± 0.44	4.80 ± 0.41	210.43 ***	C1 < C2 < C3
Rural Culture	2.63 ± 1.31	3.29 ± 1.29	4.09 ± 0.84	24.717 ***	C1 < C2 < C3
Rural Governance	2.78 ± 0.42	3.80 ± 0.33	4.99 ± 0.06	668.13 ***	C1 < C2 < C3
Material Life	4.23 ± 0.83	4.26 ± 0.87	4.79 ± 0.37	7.211 ***	C2 < C3 C1 < C3

*** p < 0.01.

. Post hoc test analysis and two-way comparison showed that the different classes of samples showed significant differences in the scores on the five dimensions. The very-satisfied class scored significantly higher than the low-satisfied class and the medium-satisfied class on the five dimensions; the medium-satisfied class scored significantly higher than the low-satisfied type on the four dimensions except for material life. Therefore, the latent classification of farmers’ life satisfaction can well distinguish the degree of households’ good life satisfaction.

3.5. Analysis of Factors Influencing Latent Classes of Rural Households’ Satisfaction with Good Life

Taking the households’ satisfaction with good life classes (low-satisfied class = 1, medium-satisfied class = 2, very-satisfied class = 3) as the dependent variable, and the very-satisfied group as the reference group to carry out multinomial logistic regression, the main explanatory variables include individual factors, family factors, and village characteristics. The parameter estimation results are shown in

Table 6. Multinomial logistic regression result of rural households’ satisfaction with good life in the classes.

Variables	Low-Satisfied Class		Medium-Satisfied Class
	RR Ratios	S.E.	RR Ratios	S.E.
Individual factors
Age	0.598	0.246	0.854	0.348
Years of education	0.519	0.302	0.363 *	0.211
Confidence in communication	2.821 ***	1.096	2.056 *	0.792
Village status	5.446 ***	3.442	1.890	1.211
Whether in a government career	0.623	0.352	0.428	0.240
Whether you are a party member	0.679	0.399	0.964	0.543
Family factors
Family size	2.574	2.013	2.205	1.724
Close family contacts	1.018	0.020	1.014	0.020
Housing area per capita	0.995	0.004	0.998	0.004
Share of non-agricultural income	0.158 ***	0.100	0.234 **	0.146
Household income level per capita	0.713	0.201	0.632 *	0.177
Village characteristics
Public service conditions	1.918 ***	0.443	1.463 *	0.337
Log likelihood	−268.371 ***
Pseudo R2	0.190

*** p < 0.01, ** p < 0.05, * p < 0.1.

.

Among the individual factors, subjective psychological evaluations such as self-confidence in communication and evaluation of status in the village have a significant positive impact on the group class of satisfaction with a good life.

In the “medium-satisfied class/very-satisfied class” model, the farmers with lower self-confidence evaluation of social interaction are more likely to be in the medium-satisfied class.

In the “low-satisfied class/very-satisfied class” model, social interaction self-confidence will further strengthen the negative effect on life satisfaction; farmers with lower self-evaluation of status in the village are more likely to be in the low-satisfied class. Finally, education has a significant negative effect on the group class of good life satisfaction. Compared with households whose householders have the education level of junior high school and below, households whose householders have education level of high school and above are more likely to be in a medium-satisfied class.

Among the family factors, the proportion of non-agricultural income and per-capita income level of the family has a significant negative impact on the group class of good life satisfaction. In the “medium-satisfied class/very-satisfied class” model, the higher the proportion of non-agricultural household income, the higher the level of per-capita household income, and the less likely it is to be a “medium-satisfied class”. In the “low-satisfied class/very-satisfied class” model, the higher the proportion of non-agricultural household income, the further the probability of “low-satisfied class” is reduced. Variables such as family size and the number of closely related families had no significant effect.

Among village characteristics, village-level public service conditions have a significant positive impact on the group class of satisfaction with a good life. Villages with poorer public service conditions such as village-level schools, hospitals, express delivery services, banking services, and agricultural materials services have a higher probability of households being in the medium-satisfied class, and the impact on the low-satisfied class is further strengthened.

4. Discussion and Conclusions

This study used 367 valid questionnaires from 14 counties of Zhejiang. The results showed that participants’ satisfaction with each dimension except rural culture was greater than the median value of three. Material life and industrial economy were the two dimensions with the highest satisfaction, which is consistent with existing studies [41]. The possible reason for these results may have been that despite the respondents originating from rural areas with a relatively less-developed level, it was located in Zhejiang Province, which is considered to be one of the most developed provinces in China. Farmers in Zhejiang Province have diversified livelihood strategies, many stable employment opportunities, and relatively high income levels. They were slightly dissatisfied with the dimensions of the rural culture. This may be due to the imperfect construction of rural public cultural facilities [42], the problem of functional failures [43], the lack of connection between cultural activities in the countryside and the needs of farmers [44], and the gradual weakening of rural traditional culture [45]. Rural Zhejiang possesses a high level of livability but a low level of satisfaction with sanitary toilets. This result is consistent with study [1].

Class heterogeneity of aggregated good life satisfaction was analyzed using latent profile analysis. Unlike previous studies which used the cluster analysis [46,47], LPA has strengths in dealing with measurement error and inefficiently identifying clusters [48]. The results show that the class heterogeneity of farmers’ satisfaction with a good life can be divided into three classes: very-satisfied class, medium-satisfied class, and low-satisfied class. At the same time, the three-class multiple comparison analysis further verified the difference rule. This classification supports studies showing that satisfaction in good life domains varies between rural households [49].

Specifically, 9.6% of households belong to the very-satisfied class, which has high satisfaction in industrial economy, living environment, rural culture, rural governance, and material life. The mean value of satisfaction conditions for all items is significantly higher than the other two classes. A total of 32.4% of households belong to the medium-satisfied class, which are more satisfied with the industrial economy, living environment, rural culture, and rural governance than those in the low-satisfied class, but there is little difference in satisfaction with basic education security, daily household appliances allocation, and household savings. A total of 58% of households belong to the low-satisfied class, and this class accounts for the largest proportion. The low-satisfied class has relatively low satisfaction with the sanitary toilet environment, Chinese New Year celebrations, cultural visits to the countryside, village committee work, fair handling of village affairs, solicitation of opinions on village affairs, and consultation on village affairs. The largest satisfaction gap between the low-satisfied class and others is the governance dimension (fair handling of village affairs, solicitation of opinions on village affairs, and consultation on village affairs). There are two possible reasons. On one hand, the rural Chinese is a society of acquaintances, which embodies a strong and enduring relational culture [50]. The clan organizations are mostly the “governance units” because it is difficult to separate rural human interaction from real authority. The dominant clan may gain more dominance of resources, while others may experience relative “deprivation” so that loss of happiness occurs [51]. On the other hand, rural governance ignores villagers’ autonomy and farmers’ participation. Studies have shown that people can obtain positive psychological feelings from political participation, thus enhancing their sense of well-being and access [52,53,54].

It is worth noting that the average satisfaction conditions of the three types of classes for cultural activities in the countryside are lower than the median value of three. This indicates that after the basic life is guaranteed and the living environment is improved, the farmers’ demand for culture is stronger. There is still a big gap in the activities of going to the countryside in meeting the needs of farmers for a better culture, which leads to the fact that households’ satisfaction with the activities of going to the countryside is not high.

In terms of influencing factors, subjective psychological factors, such as lack of self-confidence in communication and low evaluation of status in the village, are important factors that lead to the low satisfaction of farmers with a good life. This finding was consistent with previous research that stated that satisfaction is an individual’s subjective evaluation of the life quality based on their own standards, which is greatly influenced by personal cognition [55,56,57]. According to the social comparison theory, it is human nature to compare themselves with the people around them, and it is also a result of the individual pursuit of social fairness. People with higher self-confidence in social status and communication will have higher levels of subjective well-being [58]. Among the factors in the village, this study finds that poor public service conditions will lead to lower satisfaction of farmers with a better life. Yang [59] has similar views that insufficient public services for people’s livelihood have a significant negative impact on life satisfaction. Higher education level is an individual factor that can significantly improve satisfaction with a good life [60,61]. Previous research has demonstrated that households with high-quality employment or higher monthly income show higher life satisfaction [49,62]. The conclusions of this study support this view that providing farmers with more quality and efficient off-farm livelihood strategies to increase their household income can significantly improve their satisfaction with a good life. Based on the above research conclusions, this article considers that the following five aspects should be implemented according to households.

First, according to the five-dimensional scores of 367 households’ satisfaction with a good life, the most important task is to improve the rural culture. Government departments should strengthen the construction of rural culture according to the actual conditions, and in-depth search of the characteristics of local folk customs. In addition, the data in this article show that the average satisfaction of the three classes with rural cultural activities is low, which deserves the attention of the government and researchers. Cultural activities in the countryside should connect with the traditional culture and folk culture of the countryside, to ensure that such activities are closer to the farmers’ lives.

Second, increased attention should be paid to the life satisfaction of specific classes. The rural households in the low-satisfied class have relatively low satisfaction with the sanitary toilet environment, village committee work, fair handling of village affairs, and solicitation of opinions on village affairs. For areas where rural households belonging to a low-satisfied class are located, the popularization of sanitary toilets and the supervision of village cadres should be strengthened, and the level of villagers’ self-governance should be improved.

Third, attention should be paid to the psychological counselling of farmers. According to the empirical results of this article, subjective psychological factors are important factors affecting satisfaction with the good life of rural households. Therefore, it is an effective measure to improve the life satisfaction of rural households to continuously improve farmers’ self-confidence and positive attitude towards life.

Fourth, this article finds that high-quality employment contributes to life satisfaction. The government should build multiple paths and long-term mechanisms to increase farmers’ income to ensure the continuous growth of farmers’ income.

Finally, the public service conditions have a significant positive impact on the group class of satisfaction with a good life. It is recommended that the government should focus on the following factors: shortcomings of rural infrastructure and public services, quality of education in rural primary and secondary schools, medical services, and postal and communication facilities.

This study has several limitations. On one hand, the research design of this study is cross-sectional, which neither determines the causation nor offers a dynamic perspective. On the other hand, the households analyzed in this study are from the economically developed Zhejiang Province. The conclusion of this study is that it is uncertain whether it can be extrapolated to other rural areas. It is recommended that the survey area is expanded to different regions of China in future research.