A Study on the Mechanism of Female Participation in Rural Development of Yunnan on Their Capacity Building for Sustainable Development—Based on Cognitive, Emotional and Behavioural Perspectives

Abstract

Rural women’s development is a problem related to current and future rural development, as well as the development of society as a whole. This paper takes the theory of planned behaviour as the basis, researches the mechanism of women’s rural development participation with the five indicators of participation behaviour, determines the indicators of rural development participation with the theory of informed behaviour; explores the relationship between the external environment, women’s family economy, human capital, social network and family roles and the persistent poverty and determines the indicators of sustainable and responsible capacity; and constructs a hypothetical model of the influence mechanism of rural development participation on the sustainable and responsible capacity building. It also constructs a hypothetical model of the influence mechanism of rural development participation on sustainable and responsible capacity building; conducts a questionnaire survey and collects data from women in the former poverty-stricken areas of Yunnan Province; empirically analyses and verifies the hypothetical model using structural equation modelling and, finally, puts forward policy recommendations, which will serve as important references for poor rural women to improve their sustainable development capacity. The results of the empirical study indicate that the impact of participation indicators on sustainability is that cognitive participation, emotional participation and behavioural participation of female subjects all positively affect sustainability. Cognitive participation affects behavioural participation, and increasing the cognitive level of rural development participation helps increase female rural development participation. Cognitive participation and affective participation also affect sustainability through behavioural participation but not exclusively through behavioural participation. In the analysis of the role of the external environment, the external environment directly affects the effect of rural development participation on the sustainable development capacity. According to the results of the data analysis, the external environment plays a significant positive moderating effect on the economic dimension and human capital in rural development participation and sustainable development capacity. The innovations of the study on the impact mechanism of rural women’s rural development participation on sustainable development capacity building are expanding the research perspective and research methods for studying rural women’s rural development participation, constructing a scale of rural development participation and sustainable development capacity building measurement indicators and exploring the impact mechanism of rural development participation on sustainable development capacity building of rural women in Yunnan.

1. Introduction

With the further promotion of rural revitalisation, mobilising the participation of residents in backward areas to enhance their endogenous motivation has become more and more important, and “empowering farmers” has become the fundamental motivation for the development of backward areas. The participation behaviour of rural development participants in backward areas, especially poor women, is closely related to their willingness to participate, attitude and perception of behavioural control and subjective norms, among which, the more positive women’s emotion and perception of sustainable development in backward areas, the higher their participation in rural development; the higher their behavioural intention to participate in rural development projects, the more obvious their participation behaviour [1]. Therefore, in promoting women’s participation in rural development, it is necessary to promote the concept of industrial revitalisation and enhance the awareness of government services, as well as to cultivate the awareness of rural elites, strengthen their identity and improve the mechanism of interest demand [2].

1.1. Poverty and the Feminisation of Poverty

The definition of poverty is complicated, and academics have never given a uniform definition or specific judgement criteria. Because living standards and economic development vary greatly from country to country and region to region, the definition of poverty and the criteria for judging it also vary greatly. In economics, poverty refers to the material aspect, and the Chinese Bureau of Statistics clearly states that poverty refers to a family’s or individual’s standard of living being below the social minimum. Historically, poverty has two components: poverty in the broad sense and poverty in the narrow sense. Poverty in the narrow sense refers to economic poverty—that is, the inability of a family, region or individual to secure basic survival needs due to insufficient economic conditions and lack of material goods. Poverty in the broader sense, on the other hand, not only includes the content of poverty in the narrower sense but also covers environmental, social and cultural aspects of poverty, such as life expectancy of the population, culture and education, living environment and health care. According to Amartya Sen (1982), in essence, poverty is not so much about lower income as it is the lack of ability and opportunity to access benefits [3]. The social facets of poverty are very many and are created by the need to deprive opportunities or capabilities, which is the internationally recognised definition of poverty [4].

In the 1960s and 1970s, many disciplines gradually began to focus on female poverty: the first one was sociology, the second one was anthropology, the third one was feminism and so on. Government departments also began to pay attention to the problem of women’s poverty; in addition, the Women’s Federation and other social groups have also carried out research on this issue. In 1978, the famous American sociologist Pearce for the first time put forward the concept of “feminisation of poverty” and the proposition that “women are the poorest of the poor”. Since then, the problem of feminisation of poverty has gradually gained global attention and has become a global issue requiring attention [5]. Under the current concept of gender, poor rural women do not have many resources at hand, and when making decisions on social affairs, poor women are marginalised and are not sensitive enough to the gender differences in poverty. The main manifestations are income poverty, poverty under the lack of subjectivity, poverty in the ability to participate in social development and poverty in family and social status. Women’s educational attainment is generally low, with older girls dropping out of school and failing to attend school in many poor and remote areas, and class differences, the gap between rich and poor, and urban–rural differences all have a significant impact on women’s enrolment. Although women’s access to educational resources has been greatly improved, there are still serious gender inequalities in education. Ma Dongping (2011) analysed the anti-poverty policies in ethnic areas and the problems encountered in the process of implementation from a gender perspective [6].

1.2. Knowing and Acting Theory

American scholar Frijda (1993) pointed out that individual behaviour change consists of three stages: knowledge, emotion and action. “Knowing” refers to cognition, “feeling” refers to emotion and “acting” refers to behaviour. Individuals for a certain thing first produce cognition and, through cognition, will produce the emotion of the thing, and finally, from the cognitive and emotional guidance, produce the corresponding behaviour, i.e., the theory of knowledge and action [7]. Research based on the theory of informed behaviour has gradually become a hot issue in organisational behaviour research. Cognition belongs to a mental process, while emotion is an attitudinal experience, and behaviour is a response produced under a stimulus. These terms have been mentioned in some psychology, such as cognitive learning theory and behaviourist learning theory, etc., which are divided according to the cognitive learning theory, including affective engagement, cognitive engagement and behavioural engagement [8]. For the relationship between cognition, behaviour and emotion, cognition is emotionally oriented; emotionally behavioural activities are oriented and all three influence each other, interpenetrate each other and serve as a prerequisite for each other and co-development. The cognitive process is the basis of emotion and pushes emotion forward.

1.3. Capability Theory

Amartya Sen developed the concept of Capability, which refers to the combination of functional activities that are available, where “functional activities” refer to the states or things that an individual finds worth doing or has achieved [3]. He felt that “viability deprivation” and “entitlement failure” together contribute to poverty and that lower incomes are only a symptom. The viability approach has not only made a great impact internationally but has also gradually become the theoretical basis for the formulation of poverty alleviation policies and has also been of great help to the development of poverty alleviation work in China. According to Amartya Sen, viability embodies the opportunities for various optional functional activities that an individual can achieve, and the quality of life of an individual can be judged through viability. Viability does not impose any limitations on the resources at an individual’s disposal; it emphasises what an individual can really do and reflects an individual’s true degree of freedom. Human capital theory suggest that we should be people-centred, pay attention to the subjectivity of the poor group; enhance the endogenous motivation of the poor group; carry out corresponding human capital investment in the poor subject; provide standardised public services; cultivate management-related technical personnel and carry out poverty alleviation, legislation and so on.

Xie Ming et al. (2017). believe that the viable ability standard is a supplement and expansion of the income standard, which has a high degree of compatibility with accurate identification, accurate support and accurate assessment and attaches importance to the measurement of multidimensional poverty and long-term help mechanism, so as to make the poverty alleviation target have higher viable ability [9], and based on the theory of “viable ability”, it is necessary to strengthen the cultivation of “viable ability” for rural poor people with disabilities and gradually improve their awareness of participation and give attention to their role as the main body. Ma Wenfeng (2017) suggests that there is a great correlation between the theory of feasible ability and the long-term effectiveness, accuracy and initiative of carrying out accurate poverty alleviation work, and pointed out that the key to long-term poverty alleviation is to help poor groups regain their viable ability, as having viable ability means that they can obtain economic resources through labour, which provides the basic conditions for long-term poverty alleviation [10].

1.4. Theory of Capacity Building and Empowerment

“Empowerment” (empowerment) mainly emphasises the need to have a variety of abilities in order to improve the certainty of the social subject in the process of achieving the goal and to stimulate the potential abilities and strengths, so as to successfully achieve the goal [11], which also reflects the mobility of the resident subject and the initiative of the individual. Barbare Solomon (1976) was the first to use “empowerment” in the field of social work, and he believed that the actions taken by social workers in providing appropriate services “aim to reduce the sense of powerlessness of group members due to negative evaluations, and change the negative evaluations of group members by dealing with special barriers in the problem. particular barriers, changing the definition of the group that has been negatively evaluated, so that the members within the group redefine and recognise the group and regain their self-confidence and self-esteem” [12].

When researching on women’s empowerment, Yang Ziling (2014) believes that there is a link and interaction between empowerment and participation; women’s participation in affairs will allow them to improve their abilities, the improvement of their abilities will improve their participation and the participation platform will feel that farmers’ co-operative organisations are a platform for efficient development and that it can build up a pool of female talent, which will be able to participate in economic and social services in the countryside [13]. Lv Qing and Zhang Shufang (2014) believe that the use of internal and external mechanisms can achieve self-empowerment, and the most effective way is to build a complete talent training programme and make full use of external forces [14]. The most important problem encountered in the process of community governance nowadays is the lack of participation, which, in essence, is the lack of subjectivity of the residents. In order to improve the participation of residents, it is necessary to make them realise their subjectivity and, most importantly, to improve their ability to act. Huang Sijing (2015) suggests that rural left behind women lack the ability of self-control, survival and development, access to resources and democratic participation, and it is also very difficult to improve their ability in all aspects. In the process of building a new socialist countryside, the issue of capacity building for rural left behind women is particularly important [15].

The basic assumption of the empowerment theory is that the disempowered or disempowering state of individuals is caused by the exclusion and oppression of the external social environment and that indirect and direct barriers in the external social environment restrict the exercise of individual capabilities but that such barriers are not insurmountable, because it is emphasised that disadvantaged groups can be active subjects who are not incapable and who, with appropriate assistance, are fully capable of enhancing their rights, as they are not incapable, and with appropriate assistance, they can advance their rights and capabilities. The lack of role in rural women’s subjectivity and the lack of action in rural development participation constitute a circular closed loop that seriously affects and limits the effectiveness of rural community governance [16]. Rural women’s participation in rural development is a manifestation of their subjective initiative. The establishment, expansion and presentation of rural women’s subjectivity require the capacities that support women’s participatory behaviour, i.e., their economic capacity, human capital, social networks and family roles in achieving sustainable development.

How women can enhance their self-worth and contribute to rural revitalisation is through economic participation, political participation and cultural participation, so as to find, build and manifest their subjectivity and sense of existence and change from the “existence of the other” as an “outsider inside the bureau” to the “existence of the subject” as an “outsider inside the bureau”. At the same time, the study of women’s work–family balance, personal ideals and life strategies, livelihood patterns, social interactions and lifestyles, and so on, will help women to achieve a certain degree of gender order reconstruction [17]. At the same time, we study the transformation of women’s work–family balance, personal ideals and life strategies, livelihood patterns, social interactions and lifestyles and the dual possibilities of opportunities and dilemmas faced by women as individuals in the context of rural revitalisation strategy and mobile subjectivity [18]. Will there be more complicated and stronger conflicts between rural women’s professional roles and family roles, social identities and gender identities than before? Will the traditional gender order and patriarchy in the countryside change? Will rural women’s lifestyles change? Will their subjectivity be constructed? Will their ability to participate in rural community development be enhanced? Will the exogenous power be strengthened and the endogenous power be stimulated to build sustainable development? Whether the change in rural women’s lifestyle has occurred, whether their subjectivity has been constructed, whether their ability to participate in rural community development has been enhanced, whether exogenous power has been enhanced and whether endogenous power has been stimulated to build a stable and sustainable development mechanism are the issues to be explored in this study [19]. The study should give full attention to the advantages of regional enterprises and industries to attract more rural women to work and increase their income and take multiple measures to promote employment and activate the “blood-making” function of rural villages to build a new normal of rural revitalisation with the concerted efforts of government, market and society and the active participation of various social forces and to cultivate the ability of women to get rich sustainably. The ultimate effectiveness of poverty eradication will also depend on whether the poor areas and the targets of poverty eradication have endogenous development power.

Based on this, with the enhancement of sustainability as the basis and entry point; the enhancement of women’s cognitive participation, emotional participation and behavioural participation as the driving force and the new progress of poverty eradication at present as the background, this paper provides feasible ideas and paths, which also constitute the main idea of this paper’s discussion.

2. Hypothesis on the Mechanism of the Impact of Rural Development Participation on Building Women’s Capacity for Sustainable Development

2.1. Interaction Hypothesis of Rural Development Participation Indicators

Cognition, affect and behaviour are not independent or dispersed in different perspectives; they are highly correlated, and changes in any one of them may cause changes in the others. It has been shown that, in the causal chain in which behaviour influences intention, it is the individual’s self-beliefs that are the key variable leading to changes in intention and that individual cognition plays an important mediating role in the behaviour–intention relationship: behaviour quite naturally triggers thoughts that support intentions consistent with behaviour [20]. During the development of participation behaviour, cognition, emotion and behaviour influence each other, while cognition forms the basis of emotion; cognition indirectly influences behaviour, emotion directly influences behaviour and cognition, emotion and behaviour form a cascading relationship [21]. In the rural development participation index, both positive emotion and cognition will positively influence behaviour, and based on previous scholarly research combined with the actual situation of this study, the following hypotheses are proposed:

H1. 

The higher the level of cognitive involvement, the more active the behavioural involvement of women in rural development participation.

H2. 

The higher the level of emotional involvement, the more active the behavioural involvement of women in rural development participation.

2.2. Hypothesis of the Impact of Rural Development Participation on Capacity Building for Sustainable Development

The results of Sachs (2017) suggest that cognition, participation identity and responsibility are key variables that translate into future willingness to participate in the process of rural development participation realisation and play an important linking role in past rural female behaviour and future rural female participation [22]. Cognition is a prerequisite for certain emotions and behaviours based on inherent perceptions of objective conditions and one’s own needs [23]. However, for such objective perceptions to eventually be transformed into deep and stable individual values, women need to be guided in the rural development participation process to internalise perceptions and strengthen positive self-beliefs through careful consideration of the relationship between their existing experiences and their self-value beliefs; on this basis, once rural women are aware of their roles and responsibilities, they will be able to enhance their family’s economic level, human capital, social network and contribution of the family as a booster [24]. In the impact of rural development participation on sustainable development capacity building, the level of cognitive participation will directly affect the enhancement of sustainable development capacity or not; based on this, the following hypothesis is proposed:

H3. 

The higher the level of cognitive participation of women in rural development participation, the more significant the increase in sustainable development capacity.

H3a. 

The higher the cognitive participation of women in rural development participation, the more significant the economic uplift of the household.

H3b. 

The higher the cognitive participation of women in rural development participation, the more significant the human capital enhancement.

H3c. 

The higher the cognitive involvement of women in rural development participation, the more pronounced the social network enhancement.

H3d. 

The higher the cognitive involvement of women in rural development participation, the more pronounced the family role enhancement.

Rural women’s rural development participation in poor areas has slowly become a new and complex social issue in the new era since it was mentioned, and a new kind of philosophical thinking is needed to coordinate various participation factors while viewing women’s rural development participation rationally [25]. It includes women’s emotion, cognition, subject position and participation style. The influence of emotions on human actions also contains both positive and negative effects in a situation where there are more and more different influencing factors. Positive emotions significantly increase motivation and drive people to action, creating a “force multiplier” effect [26]. Based on the hypothesis that the level of emotional involvement has a direct impact on the growth of sustainable development capacity, the following hypothesis is proposed:

H4. 

The higher the level of emotional involvement of women in rural development participation, the more significant the increase in sustainable development capacity.

H4a. 

The higher the emotional involvement of women in rural development participation, the more significant the economic uplift of the household.

H4b. 

The higher the level of emotional involvement of women in rural development participation, the more significant the human capital enhancement.

H4c. 

The higher the level of emotional involvement of women in rural development participation, the more pronounced the social network enhancement.

H4d. 

The higher the level of emotional involvement of women in rural development participation, the more pronounced the family role enhancement.

From the perspective of viability, rural development participation is not only about ensuring the basic needs of women in poor areas and improving their income levels but also about developing their viability to “get out of poverty and get rich” through concrete actions [27]. In the process of rural development participation, rural women improve their agricultural and other related skills, and their rural development participation also promotes their participation in rural production and construction, which increases their economic level and enables them to gradually accumulate human capital and social networks; it also promotes the transformation of family relationships and family attitudes and improves the intergenerational transmission of poverty [28]. Based on the hypothesis that the level of participation in rural development has a direct impact on the growth of sustainable development capacity, the following hypothesis is proposed:

H5. 

The higher the level of behavioural participation of women in rural development participation, the more significant the increase in sustainable development capacity.

H5a. 

The higher the behavioural involvement of women in rural development participation, the more pronounced the economic uplift of the household.

H5b. 

The higher the level of behavioural participation of women in rural development participation, the more significant the human capital enhancement.

H5c. 

The higher the level of behavioural participation of women in rural development participation, the more pronounced the social network enhancement.

H5d. 

The higher the behavioural involvement of women in rural development participation, the more pronounced the family role enhancement.

H6. 

Behavioural participation mediates the impact of women’s participation in rural development in terms of emotional participation and cognitive participation in the building capacity for sustainable development.

2.3. Influence of External Environment Assumptions

Organisational support and the external policy environment have an important influence on rural women’s behavioural choices, and therefore, this external environment factor is taken into account in the study of the specific impact of rural development participation on sustainability. The moderating role of the external environment is also worth exploring in the study of the relationship between rural development participation, the external environment and sustainability [29]. In different empirical studies, it was found that the degree of influence of behavioural participation on sustainable development capacity building was often different from case to case, indicating that sustainable development capacity building can be influenced by external environmental factors [30]. Based on this, the following hypothesis is proposed in conjunction with the design of this study:

H7. 

The external environment positively moderates the impact of behavioural participation on capacity building for sustainable development in women’s participation in rural development.

H7a. 

The external environment positively moderates the impact of behavioural participation on household economy in women’s rural development participation.

H7b. 

External environment positively moderates the impact of behavioural participation on human capital in women’s participation in rural development.

H7c. 

The external environment positively moderates the impact of behavioural participation on social networks in women’s rural development participation.

H7d. 

The external environment positively moderates the impact of behavioural participation on household roles in women’s rural development participation.

3. An Empirical Study on the Impact Mechanism of Rural Development Participation on Building the Sustainable Development Capacity of Rural Women

3.1. Introduction to the Research Site and Questionnaire Distribution

According to the way scholars obtain and process data samples, it is known that, in order to achieve the purpose of the study, the same paper can have different case sites and sample sizes; of course, the settings of different data samples are somewhat logical and scientific, so according to the initial setting of this study, the study of women’s participation in rural development, in order to increase the universality of this study, the selection of the research scope is first based on the counties, townships and villages women were from. The relevant statistics of industrial participation, government policy documents on poverty alleviation and statistics of the Women’s Federation are used; the questionnaire survey is based on the basic interviews of female grassroots Women’s Federation cadres, entrepreneurs, wage earners and cooperative leaders in the previous section; looking for farm households radiating from surrounding villages and towns and screening the required samples from them for the model. Involving 10 prefectures (cities) and 18 counties in Yunnan Province, according to the coverage of the four contiguous areas of special hardship in Yunnan Province, one to two counties were selected in each prefecture (city) included in the area according to the principle of universality and representativeness, and one to two natural villages were randomly selected in each county for the household survey; the survey involved 30 natural villages, with a sample size of 843 households, including five prefectures (cities) and nine counties in the border area of Western Yunnan Province; 468 households, three prefectures (cities) and three counties in the rocky desertification area; 102 households, three prefectures (cities) and three counties in the Diqing Tibetan area; 165 households, three prefectures (cities) and three counties in the Wumeng mountainous area; and 108 households. The above sample data are the number of invalid questionnaires that were sifted, and the specific survey areas and questionnaire groupings are shown in Figure 1. After that, the 843 questionnaires were screened, and the main questionnaires were selected as the main requirement of this study: either the head of the household was female or the female labour force accounted for more than 50% of the overall labour force of the household. According to the screening conditions, 702 questionnaires were selected to meet the requirements of this study. At the same time, in order to ensure the overall validity and data integrity of the questionnaire, the questionnaire results were sorted again to eliminate the results data with vacant answers and poor questionnaire completeness. Finally, 679 valid questionnaires were screened, with an effective rate of 80.54%.

3.2. Questionnaire Content and Question Codes

3.2.1. Rural Development Participation Indicator Question Items

Combined with the reality of this study, women’s participation in helping organisations is not only the learning process of participation but also the preliminary cognition, and the enthusiasm of participation will affect women’s rural development participation behaviour. This study combined with the actual questionnaire questions designed a scale containing three major indicators of cognitive participation, emotional participation and behavioural participation (question interpretation) and, finally, compiled the question design about the rural development participation scale, and the specific topic indicators are shown in

Table 1. List of question codes for rural development participation indicators.

Dimensionality	Secondary Dimension	Title Number	Title	Assignment
Cognitive Engagement	Benefit Perception	RZCY1	Whether the family’s standard of living has improved	Yes = 1, No = 0
		RZCY2	Medical Assistance Help Degree	Greatly negative impact = 1, Negatively impacted = 2, Not helpful = 3, Fairly helpful = 4, Very helpful = 5
		RZCY3	Income Enhancement Ease	Difficult = 1, Less difficult = 2, Achievable = 3, Easier to achieve = 4, Easy to achieve = 5
		RZCY4	Household income and expenditure impact	No effect = 1, slight = 2, moderate = 3, large = 4, great = 5
	Self-confidence perception	RZCY5	Do you understand the help program	Yes = 1, No = 0
		RZCY6	Whether or not to fight for help projects	Yes = 1, No = 0
		RZCY7	Fairness in helping projects	Very poor = 1, poor = 2, fair = 3, good = 4, very good = 5
	External Perception	RZCY8	Poverty alleviation project initiator	Government = 1, Business = 2, Individual = 3, NGO = 4
		RZCY9	Poverty alleviation work for family help	Likert 5 scale with multiple choice quantities indicating skewed object variability
		RZCY10	Ease of access to external funding	Difficult = 1, difficult = 2, with some difficulty = 3, no difficulty = 4
Behavioural Engagement	Willingness to participate	XWCY1	Willingness to participate in poverty alleviation projects	Yes = 1, No = 0
		XWCY2	Tilt of the request for help	Likert 5 scale with multiple choice quantities indicating skewed object variability
	Participation Behaviour	XWCY3	Poverty Assistance Program Participation	Likert 5-point scale, willingness distribution
		XWCY4	Collective activity participation	Likert 5 scale, with higher scores for more participation items;
Emotional Engagement	QGCY1		Whether to receive cash assistance	Yes = 1, No = 0
	QGCY2		Ease of access to cash assistance	Difficult = 1, Less difficult = 2, Achievable = 3, Easier to achieve = 4, Easy to achieve = 5
	QGCY3		Degree of implementation of support measures	Rarely = 1, Less = 2, Average = 3, More = 4, Many = 5
	QGCY4		Help project familiarity	Likert 5 scale, where the more unfamiliar reasons selected, the lower the score

.

3.2.2. External Environment Indicators Question Items

According to the general definition of environment, the environment is mainly divided into the external and internal environment. The external environment usually exists outside the individual and is the sum of various objective factors and forces that affect the individual’s participation in activities and their development and does not change in the short term by the individual’s influence. According to the need of this study and the specific questionnaire design items of the Questionnaire on the Situation of Farmers in Poor Areas and the Status of Poverty Alleviation, the items on the external environment were compiled and obtained as shown in

Table 2. External environment scale question code list.

Dimensionality	Title Number	Title	Assignment
External Environment	WBHJ1	Whether to participate in government support or mutual aid organisations	Yes = 1, No = 0
	WBHJ1	Whether there is a long-term relationship with the company	Yes = 1, No = 0
	WBHJ1	The degree of help from mutual aid organisations	Greatly negative impact = 1, Negatively impacted = 2, Not helpful = 3, Fairly helpful = 4, Very helpful = 5
	WBHJ1	Initiator of production skills training	Likert 5 scale, where the more initiators, the higher the score;

.

3.2.3. Sustainable Development Capacity Building Indicators Question Item

The Sustainable Livelihoods analytical framework focuses on access to livelihood capital, and its policy recommendations emphasise the external granting of capital and rights, while the Viable Capability analytical framework focuses on addressing the long-term nature of poverty, and its policy recommendations emphasise the empowerment of rights, opportunities and capabilities. In terms of sustainable poverty eradication, Sen’s viable capacity is more compatible with the other three analytical frameworks; thus, in terms of selecting the corresponding analytical framework, this paper, based on Sen’s viable capacity, the sustainable livelihood capacity framework and the research literature on farmers’ capacity, as well as the study of intergenerational transmission influencing factors and sustainable development requirements in Section 4, identifies women’s sustainable development capacity in rural poor areas as economic dimension, human capital, social network and family role as the four indicators for the measurement analysis. The results are shown in

Table 3. Code list of sustainability questions.

Latitude	Secondary Dimension	Title Number	Title	Assignment
Economic Dimension	Arable land situation	JJWD1	Number of acres of cultivated land	1–5 acres = 1, 2–10 acres = 2, 11–15 acres = 3, 16–20 acres = 4, 21 acres or more = 5
		JJWD2	Whether to own agricultural appliances	Yes = 1, No = 0
		JJWD3	Main topography of cultivated land	Likert 5 scale, with higher scores for higher terrain complexity
	Residence situation	JJWD4	Residential area	1–100 m2 = 1, 101–200 m2 = 2, 201–300 m2 = 3, 301–500 m2 = 4, 500 m2 or more = 5
		JJWD5	Residential support situation	Likert 5-point scale, where the more matches, the higher the score
	Life & Entertainment	JJWD6	Number of household appliances	Likert 5-point scale, where the more matches, the higher the score
		JJWD7	TV viewing time	0.5–1 h = 1, 1.1–2 h = 2, 2.1–3 h = 3, 3.1–4 h = 4, more than 4 h = 5
		JJWD8	Traffic Convenience	Very convenient = 5, more convenient = 4, average = 3, less convenient = 2, very inconvenient = 1
	Income and Expenditure	JJWD9	Growth in revenue indicators	With growth = 1, no growth/decrease = 0
		JJWD10	Life Improvement Status	Improved a lot = 5, Improved = 4, No improvement or no opinion = 3, Decline in living condition = 2, Decline a lot = 1
		JJWD11	Family side business	Likert 5 scale, the more side hustles, the higher the score (no side hustle = 1, 4 or more side hustles = 5)
		JJWD12	Major Payment Items	Likert 5 scale, where the higher the expenditure, the higher the score
Human Capital	Labour Capital	RLZB1	Whether to go out to work	Yes = 1, No = 0
		RLZB2	Number of personnel with junior high school education	0 people = 1, 1–2 people = 2, 3–4 people = 3, 4–5 people = 4, more than 5 people = 5
		RLZB3	Presence of major diseases	Yes = 1, No = 0
		RLZB4	Amount of medical expenses	Less than 3000 CNY = 1, 3000 to 10,000 CNY = 2, 10,000 to 20,000 CNY= 3, 20,000 CNY or more = 4
	Skill Capital	RLZB5	Have participated in production skills training	Yes = 1, No = 0
		RLZB6	Degree of production skills training assistance	Greatly negative impact = 1, Negatively impacted = 2, Not helpful = 3, Fairly helpful = 4, Very helpful = 5
Social Networks	SHWN1		Number of Minorities	0 people = 1, 1–2 people = 2, 3–4 people = 3, 4–5 people = 4, more than 5 people = 5
	SHWN2		Whether to participate in religious activities	Yes = 1, No = 0
	SHWN3		Whether there are village officials in the family	Yes = 1, No = 0
	SHWN4		Neighbourhood familiarity	Very well understood = 5, better understood = 4, average = 3, not well understood = 2, not at all understood = 1
Family Role	Family Member Relationships	JTJS1	Tightness of family relationships	Less than one year = 1, more than one year and less than three years = 2, more than three years and less than five years = 3, more than five years = 4
		JTJS2	Number of outworkers	Likert 5-point scale, with higher scores for higher numbers (0 = 1, 1 = 2, 2 = 3, 3 = 4, more than 3 = 5)
		JTJS3	Length of service (member exchange)	Likert 5-point scale, with higher scores for higher total duration (0 = 1, 6 months = 2, 12 months = 3, 18 months = 4, 18+ months = 5)
	Members’ attitude towards life	JTJS4	Is there any financial help for working?	Yes = 1, No = 0
		JTJS5	Going out to work sites (taking on family responsibilities)	Neighbouring village = 1, county = 2, provincial capital = 3, out of province = 4
		JTJS6	Number of domestic staff	Likert 5-point scale, with higher scores for higher numbers (1–2 = 1, 3–4 = 2, 5–6 = 3, 7–8 = 4, 8+ = 5)

.

3.3. Descriptive Statistical Analysis of Household Demographic Characteristics

Before conducting data analysis, this study compiled a survey of female participants’ family situations. Previous scholars have argued that differences in background factors such as age, gender and education affect the variability of model results. To further argue and ensure the completeness of this study, the indicators of the background factors involved were also collated and summarised. The final results of the background factors were obtained as shown in

Table 4. Descriptive statistical analysis of the background factors.

Category	Indicators	Sample Size	Proportion (%)
Number of women in the workforce	1 person	276	40.648
	2 people	271	39.912
	3 people	121	17.82
	4 or more	11	1.62
Number of household labourers	Less than 2 people	331	48.748
	2–5 people	311	45.803
	More than 5 people	37	5.449
Number of family members	Less than 4 people	141	20.766
	4–6 people	412	60.677
	More than 6 people	126	18.557
Female labour force age	18 years old and below	162	23.859
	19–40 years old	117	17.231
	41–50 years old	140	20.619
	51–60 years old	117	17.231
	Over 60 years old	143	21.06
Annual household income	40,000 CNY and below	644	94.845
	40,000–60,000 CNY	25	3.682
	60,000–90,000 CNY	4	0.589
	100,000–120,000 CNY	5	0.736
	120,000 CNY or more	1	0.147
Education level	No cultural tutorials accepted	95	13.991
	Primary School	304	44.772
	Junior High School	229	33.726
	High school or junior college	46	6.775
	College or high school	4	0.589
	Bachelor’s degree or above	1	0.147
Whether to participate in poverty alleviation projects	Yes	549	80.854
	No	130	19.146

. According to the results of the background factor survey, it can be seen that the number of persons in the households with females as the main labour force is low, with the average number of persons per household around 1.46, and the education level of these households is mostly below junior high school, indicating that the educational level of the household is not high. Also, in rural areas, the strength and number of labour force will directly affect the total income of the family, and according to the research, it is known that the income of the families studied in this study is generally below 40,000 CNY.

3.4. Scale Reliability Test

In order to ensure the rationality of the scale and measurement items selected for the study, the study will test the scale through an overall reliability analysis and an exploratory factor analysis.

3.4.1. Analysis of Revision Scale Reliability Test

(1)

Reliability test

Reliability is the consistency and stability of test results obtained when a scale is repeatedly tested on the same subject. Commonly used reliability analyses include Cronbach’s alpha coefficient, half-measure reliability, etc. In this study, Cronbach’s alpha coefficient was used as one of the reliability indicators of each scale. For the acceptable threshold of the Cronbach’s alpha coefficient, it is generally considered to be greater than 0.7. Combining with this study and professional judgment, the threshold standard of each scale in this study was also set at 0.7, and the secondary dimensions under each scale were also measured together to ensure the reliability of the final data.

(2)

Exploratory factor analysis

In order to ensure the structural validity of the scale, exploratory factor analysis (EFA) was used for this study. Exploratory factor analysis is a technique to reduce the dimensionality of data by extracting a few core variables from variables with complex relationships to determine the essential structure of multiple factors. The exploratory factor analysis can be used to determine the structural validity of the variables by the distribution of factor loadings obtained from the exploratory factor analysis. At the beginning of the study, the KMO (Kaisex–Meyer–Olkin) measure and Bartlett’s sphere test were the main indicators considered in conjunction with the arguments of experts and scholars. The KMO threshold was chosen to be above 0.7, and the p-value of Bartlett’s sphere test should be less than 0.05. The factors were fitted and extracted according to the principle of eigenvalues greater than 1 and cumulative explained variance ratios greater than 50% at the minimum.

(3)

Validation factor analysis

Validation factor analysis is different from exploratory factor analysis, and this study was conducted after extensive research and review based on previous scholarly research. After the exploratory factor analysis, the architectural model was determined, and AMOS software was used to measure the stability of the model. In the validation of the results, it is necessary to ensure that the AVE value of each factor is greater than 0.5 and the CR value is greater than 0.7 to effectively indicate that the studied scale has good convergent validity, and it is also generally required that the factor loading coefficient (factor loading) value corresponding to each measurement item is greater than 0.7.

3.4.2. Revision Scale Reliability Test

Based on the results of the initial scale reliability test, it was found that certain items were revised and organised, and the final overall scale retained one item. In this study, we further conducted a secondary reliability test on the finalised scale. The previous study had already conducted exploratory factor analysis, and this reliability test still used Cronbach’s alpha test to determine the reliability, while the validity was verified by the AMOS structural equation model to obtain better convergent validity and discriminant validity test results.

The results of the reliability tests on the formal scale, as shown in

Table 5. Scale reliability test results.

Scale	Secondary Dimension	Cronbach’s Alpha Coefficient	Number of Items	Total Scale Reliability
Participation Indicators	Engagement Awareness	0.778	10	0.724
	Participation Behaviour	0.742	4
	Emotional Engagement	0.739	4
Sustainable Development Capacity Building Indicators	Economic Dimension	0.787	12	0.812
	Human Capital	0.741	6
	Social Networks	0.804	3
	Family Role	0.74	6
External Environment		0.758	4	0.758

, show that the reliability result value of the economic dimension increased from 0.748 to 0.787, the social network reliability value increased from 0.727 to 0.804 and the overall reliability value of the sustainability capacity scale increased from 0.722 to 0.812, while the reliability value of the participation indicator scale also increased from the initial 0.715 to 0.724. The study again demonstrated that the deleted items were worthwhile and valid.

Validation factor analysis was performed on the identified external environment scale, and model validation was performed using AMOS 24.0 software. To ensure the validity results of the scale results, the recommendations of experts and scholars were combined. The model fitness indicators were collated and obtained as shown in

Table 6. Judgment of the fitted indicators for the validation factor of the participating indicators.

Reference Indicators	X2/DF	GFI	NFI	CFI	IFI	RMSEA
Judgment criteria	<3.00	>0.90	>0.90	>0.90	>0.90	<0.08

.

(1)

Validation factor analysis of the participation index scale

A validation factor analysis was conducted on the identified participation index scale, and the participation questions were the identified questions totalling 18 items. The results according to the model fitting indicators are shown in

Table 7. Result of the fitted indicators for the validation factor of the participating indicators.

Scheme	X2/DF	GFI	NFI	CFI	IFI	RMSEA	Judgment Results
Cognitive Engagement	1.738	0.919	0.933	0.943	0.943	0.043	By
Behavioural Engagement	2.025	0.902	0.917	0.914	0.935	0.056	By
Emotional Engagement	2.735	0.909	0.905	0.949	0.958	0.061	By

, indicating that the model fitting indicators for the three dimensions of cognitive, behavioural and affective participation passed the model requirements.

According to the results of the validation factor analysis, as shown in

Table 8. Results of validation factor analysis of the participation indices.

Paths	Standardised Path Coefficient	S.E.	C.R.	p	CR	AVE
RZCY1<---Benefit perception	0.444	□	□	□	0.895	0.526
RZCY2<---Benefit perception	0.603	0.359	8.321	***
RZCY3<---Benefit perception	0.763	0.48	8.744	***
RZCY4<---Benefit perception	0.921	0.555	8.756	***
RZCY5<---Confidence perception	0.675
RZCY6<---Confidence perception	0.916	0.029	15.674	***
RZCY7<---Confidence perception	0.703	0.061	15.762	***
RZCY8<--External Perception	0.872
RZCY9<--External Perception	0.842	0.058	21.359	***
RZCY10<--External Perception	0.716	0.054	19.231	***
XWCY1<---Willingness to participate	0.671	□	□	□	0.914	0.514
XWCY2<---Willingness to participate	0.724	0.134	23.453	***
XWCY3<---Participation behaviour	0.717	0.135	23.47	***
XWCY4<---Participation behaviour	0.606	0.023	18.217	***
QGCY1<---Emotional engagement	0.622	□	□	□	0.878	0.503
QGCY2<---Emotional engagement	0.858	0.109	26.793	***
QGCY3<---Emotional engagement	0.526	0.044	6.116	***
QGCY4<---Emotional engagement	0.558	0.083	15.836	***

Note: *** indicates significance at p < 0.001.

, the AVE values with the three dimensions of cognitive, behavioural and emotional involvement were 0.526, 0.514 and 0.503, respectively, which were all greater than the requirement of 0.5. Also, the corresponding CRs were all greater than 0.8; most of the standardised regression coefficients for each question item on its corresponding latent variable were above 0.4, and all of these standardised estimates were statistically significant at p < 0.01, indicating that this study had good convergent validity.

(2)

Validation factor analysis of the sustainability capacity scale

The validation factor analysis was conducted on the identified sustainable development capacity scale, and the participation questions were the identified questions totalling 31 items. According to the model fitting index results as shown in

Table 9. Judgment of the fitting indicators for the sustainability validation factors.

Name	X2/DF	GFI	NFI	CFI	IFI	RMSEA	Judgment Results
Economic Dimension	1.947	0.989	0.986	0.987	0.987	0.074	By
Human Capital	1.9424	0.908	0.947	0.948	0.948	0.068	By
Social Networks	2.776	0.903	0.904	0.905	0.905	0.076	By
Family Role	1.439	0.994	0.986	0.998	0.998	0.025	By

, the four latitude model fitting indices of participation economic dimension, human capital, social network and family role all passed the model requirements.

According to the results of the validation factor analysis as shown in

Table 10. Results of the validation factor analysis of the capacity building indicators.

Paths	Standardised Path Coefficient	S.E.	C.R.	p	CR	AVE
JJWD9<---Income and Expenditure	0.55	□	□	□	0.906	0.543
JJWD10<---Income and expenditure status	0.566	0.084	11.854	***
JJWD11<---Income and expenditure status	0.821	0.495	14.679	***
JJWD12<---Income and expenditure status	0.866	0.592	14.73	***
JJWD6<---Living Entertainment	0.857
JJWD7<---Living Entertainment	0.89	0.02	29.889	***
JJWD8<---Living Entertainment	0.859	0.051	28.822	***
JJWD1<---Arable land situation	0.794
JJWD2<---Arable land situation	0.524	0.027	7.34	***
JJWD3<---Arable land situation	0.473	0.07	7.126	***
JJWD4<---Residential situation	0.545
JJWD5<---Residential situation	0.63	0.826	7.189	***
RLZB1<---labour capital	0.844	□	□	□	0.916	0.518
RLZB2<---labour capital	0.732	0.696	6.588	***
RLZB3<---labour capital	0.865	0.411	6.487	***
RLZB4<---labour capital	0.701	0.241	7.465	***
RLZB5<---Skill capital	0.76
RLZB6<---Skill capital	0.644	0.236	9.015	***
SHWN1<---Social Network	0.745	□	□	□	0.907	0.521
SHWN2<---Social Network	0.428	0.075	3.681	***
SHWN4<---Social Network	0.503	0.019	3.499	***
JTJS1<---Family member relationship	0.525				0.888	0.509
JTJS2<---Family member relationship	0.678
JTJS3<---Family member relationship	0.637	0.142	23.133	***
JTJS4<---Members’ attitude towards life	0.545	0.018	14.131	***
JTJS5<---Members’ attitude towards life	0.703	0.117	5.081	***
JTJS6<---Members’ attitude towards life	0.682	0.571	5.1	***

Note: *** indicates significance at p < 0.001.

, the AVE values of economic dimension, human capital, social network and family role were 0.543, 0.518, 0.521 and 0.509, respectively, all of which were greater than 0.5, while the corresponding CRs were all greater than 0.8; the standardised regression coefficients of each question item on the corresponding latent variables were greater than 0.4, and these standardised estimates were all at p < 0.01, indicating that this study has good convergent validity.

(3)

Validation factor analysis of the external environment scale

The validation factor analysis was conducted on the identified external environment scale, and the main participating questions were the finalised external environment scale questions. The scale fits are shown in

Table 11. Judgment of the fitting indicators for the validation factor of the external environment scale.

Name	X2/DF	GFI	NFI	CFI	IFI	RMSEA	Judgment Results
Criteria or thresholds for adaptation	<3.00	>0.90	>0.90	>0.90	>0.90	<0.08
External Environment	2.736	0.957	0.936	0.938	0.938	0.062	By

Data source: Compiled by this study.

, and it can be seen that the values of the indicators used to evaluate the model can meet the threshold value requirements, indicating that the fit indicators of the external environment scale model of this study pass.

According to the results of the validation factor analysis as shown in

Table 12. Results of the external environment validation factor analysis.

Paths	Standardised Path Coefficient	S.E.	C.R.	p	CR	AVE
WBHJ1<---External environment	0.516				0.817	0.532
WBHJ2<---External environment	0.552	0.061	12.94	***
WBHJ3<---External environment	0.418	0.171	18.055	***
WBHJ4<---External environment	0.501	0.144	17.91	***

Note: *** indicates significance at p < 0.001. Data source: Compiled by this study.

, it can be seen that the AVE values of the external environment of the questionnaire exceeded 0.5 and the combined reliability CR of the scale was greater than 0.8. The standardised regression coefficients of each question item on its corresponding latent variable were above 0.4, and all of these standardised estimates were statistically significant at p < 0.01, indicating that this study had good convergent validity.

3.5. Empirical Analysis and Hypothesis Testing

3.5.1. Correlation Analysis and Differential Validity

Based on the expansion of the size of the case sites of the research questionnaire in this section and the expansion of the sample size, correlation analysis was performed before the analysis of the variables according to the statistical principles. Based on p ≤ 0.05, it was determined whether the correlation was significant or not, while the positive and negative strong correlations were determined based on the Pearson correlation result values, and the correlation coefficient matrix between the study variables is shown in

Table 13. Correlation analysis: matrix of the correlation coefficients between study variables.

	Indicators	Number of Women in the Workforce	Number of Household Labourers	Number of Family Members	Annual Household Income	Education Level	Engagement Awareness	Participation Behaviour	Emotional Engagement	Economic Dimension	Human Capital	Social Networks	Family Role	External Environment
		1	2	3	5	6	7	8	9	10	11	12	13	14
Background factors	1	0.082
	2	0.080	1
	3	0.401 **	0.216 **	1
	4	0.103 **	0.036 **	0.612
	5	0.216	0.051 **	0.241 **	1
	6	0.193 **	0.042 **	0.026 **	0.147 **	1
Participation Indicators	7	0.183 **	0.031 **	0.213	0.324	0.188 **	0.725
	8	0.268 **	0.213 *	0.652	0.623	0.012 **	0.287 **	0.717
	9	0.042	0.075	0.329 **	0.216	0.230 **	0.384 **	0.357 **	0.709
Sustainable Development Capability	10	0.024	0.412	0.257 *	0.398	0.331 **	0.642 **	0.155 **	0.488 *	0.737
	11	0.561	0.190 **	0.339	0.451	0.489	0.311 **	0.515 **	0.458 *	0.199 **	0.720
	12	0.045	0.005 **	0.417	0.181 **	0.432	0.163 **	0.208 **	0.136 **	0.271 **	0.430 **	0.722
	13	0.241	0.322	0.325 **	0.47	0.643	0.322 **	0.331 **	0.156 **	0.106 **	0.606 **	0.234 **	0.713
External Environment	14	0.035	0.519	0.002 **	0.397	0.279	0.311 **	0.156	0.325	0.015 *	0.209 **	0.017 **	0.462	0.729

Note: ** significant correlation at the 0.01 level (two-tailed); * significant correlation at the 0.05 level (two-tailed); square root of AVE on the diagonal.

. From the results of the data in this table, it can be seen that only some of the six indicators of background factors have some correlation with the variables of this study, but the overall correlation shows a weak correlation, and also, according to the comparison of the square root of the average variance extracted (AVE) and the correlation coefficient result value between the variables, it is significantly greater than the correlation coefficient between their corresponding variables, which indicates that the results of the determined scale have a good discriminant validity.

Meanwhile, according to Table 13, it can be seen that the results of each dimension of the participation indicators and sustainable development capacity building show a positive correlation. The moderating variable (external environment) has a certain correlation with some of the variables studied, which initially verifies the hypothetical relationship proposed in this paper, but the correlation only indicates that there is a certain correlation between the two variables and does not reflect the causal relationship between them, which needs to be further tested by means of regression analysis.

3.5.2. Analysis of Variability of Indicators before and after Participation

In the pre-hypothesis test, we determined whether or not participation in rural development projects causes individual differences across households—in particular, for the subject of this study, households with more female contributions. Differences were analysed by examining whether or not households participated in rural development projects. Statistical principles were used to make comparisons using independent sample t-tests. The final scores for each dimension were also calculated using the regression method in conjunction with the pre-factor analysis process, and the results are shown in

Table 14. Variance analysis of whether to participate in rural development projects.

Dimensionality	Participation	Did Not Participate	T	p
Economic Dimension	0.071 ± 1.994	0.353 ± 1.999	−4.061	0.000
Human Capital	0.305 ± 1.325	0.236 ± 1.452	−3.109	0.000
Social Networks	0.511 ± 0.901	0.554 ± 1.258	1.959	0.104
Family Role	0.098 ± 1.323	0.493 ± 1.726	−6.104	0.000
Cognitive Engagement	0.128 ± 1.742	0.042 ± 1.681	4.956	0.000
Behavioural Engagement	0.108 ± 0.829	0.043 ± 1.494	6.508	0.000
Emotional Engagement	0.209 ± 0.992	0.044 ± 1.041	3.510	0.011
External Environment	0.251 ± 0.981	0.254 ± 1.056	0.980	0.323

Note: If p ≤ 0.05, this indicates a significant difference between participation and non-participation.

. The data study showed that, among the participating and non-participating families, the overall data performance differences were in the economic dimension, human capital, social network, family role, cognitive involvement, behavioural involvement, emotional involvement and external environment. According to the results, the non-participating families were significantly more economically well off than the participating families. This situation was found for both human capital and family role. Then, according to the cognitive participation, behavioural participation and emotional participation, the participating families showed significantly higher results, indicating that the participating families had a higher recognition of rural development, while the support given by the government/organisations was biased. In conjunction with this study, the families involved in the rural development project were selected as the main subjects in the later stage of the study to consider the changes in the role and impact of the indicators.

3.5.3. Direct Action Mechanism Study

Mechanism of the Role of Cognitive Involvement and Affective Involvement on Behavioural Involvement

In this study, a multiple linear regression equation model was used to verify the hypothesised effects of cognitive involvement and affective involvement on behavioural involvement based on statistical principles. First, six major background factors, namely, the number of female labour force, the number of household labour force, the number of household size, age, annual household income and education level, were put into the model to verify whether there was an effect on behavioural involvement. Second, on the basis of model 1, cognitive involvement and emotional involvement variables were added to construct model 2. The results of the specific analysis model are shown in

Table 15. Results of regression analysis of the cognitive involvement, emotional involvement and behavioural involvement.

Variables	Model 1	Model 2
Family Labour	0.022 (3.567) **	0.017 (2.449) *
Female labour force age	−0.026 (−0.674)	−0.028 (−0.739)
Annual household income	−0.146 (−3.813)	−0.209 (−3.761) **
Number of family members	0.001 (0.005)	−0.003 (−0.071)
Education level	0.035 (2.405) *	0.034 (2.702) *
Cognitive Engagement		0.205 (5.135) ***
Emotional Engagement		0.312 (6.291) ***
R2	0.222	0.508
Adj R2	0.175	0.491
F	9.121 *	46.467 ***

Note: *** indicates significance at p < 0.001; ** indicates significance at p < 0.01; * indicates significance at p < 0.05.

.

From Table 15, model 1, it can be seen that household labour (β = 0.022, p < 0.01) has a significant relationship with behavioural participation, indicating that sufficient household labour makes it easier for more labour to participate in rural development organisations. Educational attainment (β = 0.035, p < 0.05) has a significant relationship with behavioural participation, implying that higher educational attainment leads to stronger behaviours of participation. From model 2, it is clear that when the two variables of cognitive involvement and affective involvement are included in the model, the effect of cognitive involvement on behavioural involvement is significant (β = 0.205, p < 0.001), and similarly, the effect of affective involvement on behavioural involvement is significant (β = 0.312, p < 0.001), and the adjusted R² of model 1 is 0.491, indicating that the model has strong explanatory power for behavioural involvement; combined with the model F-test results values, it is clear that the model has a p-value significantly lower than 0.001 and is statistically significant. Ultimately, based on the model results, we conclude that cognitive involvement and affective involvement both positively affect behavioural involvement, and hypotheses H1, H1a and H1b hold.

Mechanism of the Role of Behavioural Participation on Sustainability

Based on the results of the preliminary analysis, it is known that sustainability is divided into four main dimensions, namely, economic dimension, human capital, social network and family role. Based on this, this study will build four model results to investigate the mechanism of behavioural involvement on economic dimension, human capital, social network and family role, respectively. The model results are shown in

Table 16. Results of the regression analysis of behavioural engagement on sustainability.

Variables	Model 3	Model 4	Model 5	Model 6
	Economic Dimension	Human Capital	Social Networks	Family Role
Family Labour	0.01 (3.268) **	0.013 (2.749) *	0.071 (2.972) *	0.049 (3.356) **
Female labour force age	0.046 (1.181)	−0.037 (−0.952)	0.036 (1.026)	−0.007 (−0.178)
Annual household income	−0.001 (−0.985)	−0.008 (−0.201)	0.022 (0.627)	0 (−0.007)
Number of family members	0.035 (0.904)	−0.012 (−0.317)	−0.006 (−0.177)	−0.012 (−0.332)
Education level	0.044 (2.148)	0.036 (2.917) *	0.035 (3.756) **	0.076 (3.073) **
Behavioural Engagement	0.137 (6.332) ***	0.251 (8.311) ***	0.194 (11.052) ***	0.217 (8.612) ***
R2	0.472	0.551	0.319	0.547
Adj R2	0.415	0.523	0.296	0.527
F	35.645 ***	36.486 ***	22.050 ***	41.873 ***

Note: *** indicates significance at p < 0.001; ** indicates significance at p < 0.01; * indicates significance at p < 0.05.

.

As seen in Table 15, model 1, behavioural involvement (β = 0.137, p < 0.001) is significantly related to behavioural economic dimensions, indicating that higher behavioural involvement is more likely to lead to an increase in household economy. Behavioural involvement (β = 0.251, p < 0.001) is significantly related to human capital, indicating that higher levels of behavioural involvement are more likely to lead to higher human capital. Behavioural engagement (β = 0.194, p < 0.001) has a significant relationship with social networks, indicating that higher levels of behavioural engagement are more likely to enhance people’s social relationships. Behavioural involvement (β = 0.194, p < 0.001) was significantly related to family roles, indicating that higher levels of behavioural involvement are more likely to enhance positive changes in people’s family roles. Ultimately, based on the model results, we conclude that higher behavioural involvement enhances sustainability, and hypotheses H5, H5a, H5b, H5c and H5d hold.

3.5.4. Impact Study of the Moderating Role of the External Environment

Previously, researchers divided sustainability into four main dimensions: economic dimension, human capital, social network and family role. In this test of the moderating effect, multiple linear regression equations were used to test the interaction terms for verifying the existence and direction of the moderating effect. The regression equation was used to test the moderating effect, and the regression equation included two as follows:

y = a + bx + cm + e

(1)

y = a + bx + cm + c′mx + e

(2)

In the above equation, m is the moderating variable, mx is the moderating effect and whether the moderating effect is significant is the analysis of whether c′ is significant up to the level of the critical ratio of 0.05 in statistical significance.

Test of the Moderating Role of the External Environment between Behavioural Involvement and Economic Dimensions

To test the moderating effect of the external environment between behavioural involvement and economic dimensions, firstly, contextual factors, independent variables (behavioural involvement) and moderating variables (external environment) were added to the model to analyse their effects on the economic dimensions; secondly, external environment*behavioural involvement was put into the model as an interaction term to test whether there was a moderating effect, and the results of the analysis are shown in

Table 17. Regression test of the external environment as a moderator between behavioural involvement and economic dimensions.

Variables	Model 7	Model 8
Family Labour	0.01 (2.561) *	0.001 (3.287) **
Female labour force age	0.045 (1.156)	0.045 (1.171)
Annual household income	−0.001 (−0.789)	−0.002 (−0.058)
Number of family members	0.035 (0.906)	−0.039 (−1.016)
Education level	0.045 (1.157)	0.048 (1.226)
Behavioural Engagement	0.016 (7.408) ***	0.102 (3.405) **
External Environment	0.029 (2.748) *	0.032 (2.783) *
External Environment × Behavioural Engagement		0.046 (3.361) **
R2	0.369	0.451
Adj R2	0.337	0.438
F	37.681 ***	35.443 ***

Note: *** indicates significance at p < 0.001; ** indicates significance at p < 0.01; * indicates significance at p < 0.05.

. From the data in this table, it can be seen that the regression coefficient of the external environment × behavioural involvement is significant (β = 0.046, p < 0.01), and the R² significance of model 8 is enhanced compared to model 7, while the F-values of both models are significant, indicating that the moderating effect of the external environment as a moderating variable is significant. Therefore, according to the coefficient of the interaction term, the direction of regulation is positive regulation. In other words, a good or bad external environment positively affects the relationship between behavioural participation and economic dimensions. Therefore, hypothesis H7a is valid.

Testing the Moderating Role of External Environment between Behavioural Engagement and Human Capital

To test the moderating effect between behavioural involvement in the external environment and human capital, firstly, contextual factors, independent variable (behavioural involvement) and moderating variable (external environment) were added to the model to analyse their effects on human capital; secondly, external environment*behavioural involvement was put into the model as an interaction term to test whether there was a moderating effect, and the results of the analysis are shown in

Table 18. Regression test of the external environment as a moderator between behavioural engagement and human capital.

Variables	Model 9	Model 10
Family Labour	0.014 (3.365) **	0.014 (4.357) ***
Female labour force age	−0.035 (−0.907)	−0.035 (−0.91)
Annual household income	−0.008 (−0.2)	−0.007 (−0.183)
Number of family members	−0.012 (−0.322)	−0.014 (−0.352)
Education level	−0.036 (−0.936)	−0.037 (−0.953)
Behavioural Engagement	0.057 (3.458) **	0.053 (7.272) ***
External Environment	0.056 (1.458)	0.057 (3.466) **
External Environment × Behavioural Engagement		0.011 (2.774) *
R2	0.511	0.565
Adj R2	0.489	0.54
F	44.520 ***	41.751 ***

Note: *** indicates significance at p < 0.001; ** indicates significance at p < 0.01; * indicates significance at p < 0.05.

. From the data in this table, it can be seen that the regression coefficient of external environment*behavioural involvement is significant (β = 0.011, p < 0.05), and the R² significance of model 10 is enhanced compared to model 9, while the F-values of both models are significant, indicating that the moderating effect of the external environment as a moderating variable is significant. Therefore, according to the coefficient of the interaction term, the direction of regulation is positive regulation. In other words, a good or bad external environment positively affects the role between behavioural engagement and human capital. Therefore, hypothesis H7b is valid.

Test of the Moderating Role of the External Environment between Behavioural Involvement and Social Networks

In order to test the moderating effect of the external environment between behavioural involvement and social network, firstly, contextual factors, independent variable (behavioural involvement) and the moderating variable (external environment) were added to the model to analyse their effects on social network; secondly, external environment*behavioural involvement was put into the model as an interaction term to test whether there was a moderating effect, and the results of the analysis are shown in

Table 19. Regression test of the external environment in moderating between behavioural involvement and social network.

Variables	Model 11	Model 12
Family Labour	0.07 (2.988) *	0.068 (2.930) *
Female labour force age	0.035 (0.981)	0.036 (1.021)
Annual household income	−0.022 (0.626)	−0.018 (−0.491)
Number of family members	−0.006 (−0.173)	−0.004 (−0.102)
Education level	0.036 (1.016)	0.042 (1.187)
Behavioural Engagement	0.389 (10.851) ***	0.119 (10.995) ***
External Environment	0.05 (3.41) **	0.053 (7.497) **
External Environment × Behavioural Engagement		0.086 (2.263)
R2	0.382	0.389
Adj R2	0.369	0.372
F	37.681 ***	37.443 ***

Note: *** indicates significance at p < 0.001; ** indicates significance at p < 0.01; * indicates significance at p < 0.05.

. From the data in this table, it can be seen that the regression coefficient of external environment × behavioural involvement is not significant (β = 0.086, p > 0.05), indicating that the moderating effect of the external environment as a moderating variable is not significant. In other words, the effect of good or bad external environment on the relationship between behavioural involvement and social network is not significant. Therefore, hypothesis H7c is not valid.

Examination of the Moderating Role of the External Environment between Behavioural Involvement and Family Roles

To test the moderating effect of the external environment between behavioural involvement and family role, firstly, contextual factors, independent variable (behavioural involvement) and moderating variable (external environment) were added to the model to analyse their effects on family role; secondly, external environment × behavioural involvement was put into the model as an interaction term to test whether there was a moderating effect, and the results of the analysis are shown in

Table 20. Regression test of the external environment as a moderator between behavioural involvement and family role.

Variables	Model 13	Model 14
Family Labour	0.049 (4.354) **	0.05 (3.364) **
Female labour force age	−0.007 (−0.182)	−0.006 (−0.175)
Annual household income	0 (−0.007)	−0.001 (−0.03)
Number of family members	−0.012 (−0.331)	−0.01 (−0.281)
Education level	−0.076 (−2.07)	−0.075 (−2.033)
Behavioural Engagement	0.316 (8.545) ***	0.122 (8.147) ***
External Environment	0.004 (3.108) **	0.005 (3.123) *
External Environment × Behavioural Engagement		−0.015 (1.391)
R2	0.365	0.371
Adj R2	0.344	0.359
F	31.681 ***	33.443 ***

Note: *** indicates significance at p < 0.001; ** indicates significance at p < 0.01; * indicates significance at p < 0.05.

. From the data in this table, it can be seen that the regression coefficient of external environment × behavioural involvement is not significant (β = −0.015, p > 0.05), indicating that the moderating effect of the external environment as a moderating variable is not significant. That is, the effect of good or bad external environment on the relationship between behavioural involvement and family roles is not significant. Therefore, hypothesis H7d is not valid.

4. Structural Equation Model Testing

To further investigate the mechanism of the influence of participation indicators on sustainability, the study further uses structural equation modelling to do path–test analysis on them. Since the effect of control and moderating variables are explored in detail in this paper in the hierarchical regression analysis, they are not introduced into the structural equation here to maintain the parsimony of the model. In addition, considering the large number of variable measurement question items in this paper, principal component analysis is used in this paper to obtain the factors under each dimension as the source of question items, and the combined item values are taken as the three-level factors. According to the theoretical model designed in this paper, AMOS statistical software was used to construct the initial structural equation model of cognitive/emotional involvement → behavioural involvement → sustainable development (poverty eradication) ability. The whole study on the impact of rural participation on women’s ability is based on the goal of sustainable long-term poverty eradication and prosperity, so women who are still in rural areas with a low development level are in need of sustainable poverty eradication as the core element of sustainable development. The term “ability to escape poverty” appears below for this reason and is hereby explained.

4.1. Schematic Diagram of the Structural Equation Model

According to the classification of sustainable development capacity indicators, the model construction analysis of rural development participation indicators and economic dimensions was carried out first, and the sample data were imported for the model operation, and the results of the model run are shown in Figure 2. The X²/DF = 2.595, GFI = 0.879, RMSEA = 0.073, NFI = 0.902, IFI = 0.908 and CFI = 0.908 of the model show a good fit.

According to the values in

Table 21. Participation indicators and economic dimension model path fitting results.

Paths			Standardisation Factor	S.E.	C.R.	p
Behavioural Engagement	<--	Cognitive Engagement	0.276	0.046	7.013	***
Behavioural Engagement	<--	Emotional Engagement	0.215	0.051	6.595	***
Economic Dimension	<--	Behavioural Engagement	0.257	0.058	4.325	***
Economic Dimension	<--	Emotional Engagement	0.022	0.163	1.012	0.215
Economic Dimension	<--	Cognitive Engagement	0.158	0.049	3.765	***

Note: *** indicates significance at p < 0.001; X² = 397.48, DF = 153, GFI = 0.879, RMSEA = 0.073, NFI = 0.902, IFI = 0.908 and CFI = 0.908.

of the model path results, the value of the coefficient of the path emotional involvement → economic dimension is not significant, while the rest of the paths are significant. Combining the statistical principles and the previous regression results, the final study concluded that there is no significant direct effect of emotional involvement on the economic dimension (β = 0.022, p = 0.215 > 0.05). Also, based on the comparison of other path coefficients, it is clear that there is a significant path effect of both cognitive involvement and emotional involvement on behavioural involvement with coefficient results of β = 0.276, p ≤ 0.001 and β = 0.215, p ≤0.001, respectively. Also, based on the path results of behavioural involvement and cognitive involvement on the economic dimension, the coefficients were β = 0.257, p≤ 0.001 and β = 0.158, p ≤0.001, respectively, indicating that both had a significant effect relationship. Combined with the theoretical and pre-post regression results, behavioural engagement partially mediated the relationship between cognitive engagement and economic latitude.

4.2. Model of the Impact of Rural Development Participation on Human Capital

According to the classification of the sustainable development capacity building indicators and then rural development participation indicators and human capital model construction analysis, import the sample data for model operations, and the model results are shown in Figure 3; according to the “Model Fit Summary” fitting results, the model fit indicators X²/DF = 1.770, GFI = 0.901, RMSEA = 0.053, NFI = 0.908, IFI = 0.921 and CFI = 0.921 showed good fit.

According to the values in

Table 22. Participation indicators and human capital model path fitting results.

Paths			Standardisation Factor	S.E.	C.R.	p
Behavioural Engagement	<--	Cognitive Engagement	0.212	0.062	3.751	***
Behavioural Engagement	<--	Emotional Engagement	0.184	0.081	6.335	***
Human Capital	<--	Behavioural Engagement	0.173	0.076	5.678	***
Human Capital	<--	Emotional Engagement	0.133	0.056	2.056	0.006
Human Capital	<--	Cognitive Engagement	0.279	0.068	3.688	***

Note: *** indicates significance at p < 0.001; X² = 132.814, DF = 75, GFI = 0.901, RMSEA = 0.053, NFI = 0.908, IFI = 0.921 and CFI = 0.921.

of the model path results, the coefficient value of the path emotional involvement → human capital is significant, and the rest of the paths are significant. Based on the comparison of the path coefficients, it can be seen that there is a significant path effect of cognitive involvement and emotional involvement on behavioural involvement with coefficient results of β = 0.212, p ≤ 0.001 and β = 0.184, p ≤ 0.001, respectively. Also, according to the path results of behavioural involvement, emotional involvement and cognitive involvement on human capital, the coefficients were β = 0.173, p ≤ 0.001, β = 0.133, p ≤ 0.01 and β = 0.279, p ≤ 0.001, respectively, indicating that all had significant influence relationships. Combining the theoretical and pre-post regression results, behavioural engagement partially mediated the relationship between cognitive engagement and human capital while partially mediating the relationship between affective engagement and human capital.

4.3. Model of the Impact of Rural Development Participation on Social Networks

According to the classification of sustainable development capacity building indicators, the rural development participation indicators and social network model construction analysis, import the sample data for model operations, and the model results are shown in Figure 4; according to the “Model Fit Summary” fitting results, the model fit indicators X²/DF = 3.180, GFI = 0.901, GFI = 0.887, RMSEA = 0.074, NFI = 0.896, IFI = 0.901 and CFI = 0.901 showed a good fit.

According to the values in

Table 23. Participation indicators and social network model path fitting results.

Paths			Standardisation Factor	S.E.	C.R.	p
Behavioural Engagement	<--	Cognitive Engagement	0.202	0.091	4.521	***
Behavioural Engagement	<--	Emotional Engagement	0.189	0.089	3.554	***
Social Networks	<--	Behavioural Engagement	0.418	0.077	5.227	***
Social Networks	<--	Emotional Engagement	0.053	0.322	1.977	0.076
Social Networks	<--	Cognitive Engagement	0.184	0.077	2.936	0.001

Note: *** indicates significance at p < 0.001; X² = 305.280, DF = 96, GFI = 0.887, RMSEA = 0.074, NFI = 0.896, IFI = 0.901 and CFI = 0.901.

of the model path results, the value of the coefficient of the path emotional involvement → social network is not significant, while the rest of the paths are significant. Combining the statistical principles and the results of the prior regression, the final study concluded that there was no significant direct effect of emotional involvement on social networks (β = 0.053, p = 0.076 > 0.05). Based on the comparison of other path coefficients, it is clear that there is a significant path effect of both cognitive involvement and emotional involvement on behavioural involvement, with coefficient results of β = 0.202, p ≤ 0.001 and β = 0.189, p ≤ 0.001, respectively. Based on the path results for behavioural and cognitive involvement on social networks, the coefficients were β = 0.418, p ≤ 0.001 and β = 0.184, p ≤ 0.01, respectively, indicating that both had significant influence relationships. Combined with the theoretical and pre-post regression results, behavioural engagement partially mediated the relationship between cognitive engagement and social networks.

4.4. Model of the Impact of Rural Development Participation on Household Roles

According to the classification of sustainable development capacity building indicators, the rural development participation indicators and household role model construction analysis, import the sample data for model operations, and the model results are shown in Figure 5; according to the “Model Fit Summary” fitting results, the model fit indicators X²/DF = 2.754, X² = 261.643, DF = 95, GFI = 0.894, RMSEA = 0.069, NFI = 0.899, IFI = 0.907 and CFI = 0.907 showed a good fit.

According to the values of the model path results in

Table 24. Participation indicators and family role model path fitting results.

Paths			Standardisation Factor	S.E.	C.R.	p
Behavioural Engagement	<--	Cognitive Engagement	0.162	0.052	2.013	0.007
Behavioural Engagement	<--	Emotional Engagement	0.221	0.076	5.665	***
Family Role	<--	Behavioural Engagement	0.23	0.083	4.428	***
Family Role	<--	Emotional Engagement	0.034	0.241	1.235	0.197
Family Role	<--	Cognitive Engagement	0.187	0.066	2.984	0.001

Note: *** indicates significance at p < 0.001; X² = 261.643, DF = 95, GFI = 0.894, RMSEA = 0.069, NFI = 0.899, IFI = 0.907 and CFI = 0.907.

, the value of the coefficient of the path emotional involvement → family role is not significant, while the rest of the paths are significant. Combining the statistical principles and the previous regression results, the final study concluded that there was no significant direct effect of emotional involvement on family role (β = 0.134, p = 0.197 > 0.05). Based on the comparison of the other path coefficients, it is evident that there is a significant path effect of both cognitive involvement and emotional involvement on behavioural involvement, with coefficient results of β = 0.162, p ≤ 0.001 and β = 0.221, p ≤ 0.001, respectively. Based on the path results for behavioural involvement and cognitive involvement on family roles, the coefficients were β = 0.230, p ≤ 0.001 and β = 0.066, p ≤ 0.01, respectively, indicating that both had significant influence relationships. Combined with the theoretical and pre-post regression results, behavioural involvement partially mediated the relationship between cognitive involvement and family role.

5. Analysis of the Results of the Empirical Study

Based on the hypothetical basis of the influence mechanism of rural development participation on rural women’s sustainable development capacity building, this section uses exploratory factor analysis and validation factor analysis methods to test the validity and reliability of the initial measurement questions of each variable in the constructed model and makes adjustments and revisions according to the requirements of the indicators to form a measurement model with a high degree of fit.

The results of the scale on women as rural development subjects were compiled through the indicator regression of the Questionnaire on the Situation of Farming Households and Poverty Alleviation in Poor Areas of Yunnan Province, and the indicator scales of the three dimensions of rural development participation, sustainable development capacity and external environment were explored. The analysis of the rural development participation indicators yielded three dimensions of rural development participation indicators, including cognitive participation, emotional participation and behavioural participation. The SEM structural equation modelling study also showed that the impact mechanism of rural development participation on sustainability showed that cognitive participation, affective participation and behavioural participation all affect sustainability in female-dominated rural development. Cognitive participation influences behavioural participation with an impact coefficient of 0.312, indicating that, in female-dominated rural development, increasing the cognitive level of rural development participation and deepening cognition helps to assist in rural development efforts. Cognitive participation and affective participation also influence the sustainable development capacity through behavioural participation. In the analysis of the role of the external environment, the goodness of the external environment is directly related to the effect of rural development participation on sustainable development capacity building. According to the results of the data analysis, the external environment plays a moderating role in the impact of rural development participation on sustainable development capacity building, and it is a positive moderating effect. It indicates that, the better the external environment is, the more significant is the effect of female rural development participation on sustainable development capacity building. It further indicates that the change of external environment in the process of female subjects’ participation is elevated, which will help to improve their sustainable development capacity rapidly.

Innovative points of this study. In response to the requirements of development in the post-poverty alleviation era, this paper combines the research foundations of previous research on rural women’s issues, rural women’s participation and women’s ability to study the impact mechanism of women’s participation in rural development in rural areas of Yunnan on the building of sustainable and responsible capacity: (1) Expand the research perspectives and methods of researching women’s participation in rural development. It enriches the research theory on women’s issues by studying the special role of Yunnan’s rural female labour force in the fight against poverty in the countryside and its capacity for sustainable development. Currently, research based on the theory of planned behaviour is applied to the study of farmers’ participation in agricultural land improvement and environmental management, tourism and entrepreneurial behaviour, etc., while the research on rural women’s participation in rural development as the object of study is relatively rare. Therefore, this paper measures the mechanism of women’s participation in rural development from the perspective of the theory of planned behaviour and proposes the concept of women’s capacity building for sustainable development to make up for the shortcomings of this research. (2) Construct a scale of indicators for measuring rural development participation and sustainable and responsible capacity building. This study breaks with the traditional concepts and measurement standards of rural development participation; combines the theory of planned behaviour and the theory of informed behaviour and establishes measurement scales for behavioural attitudes, subjective norms, perceptions of behavioural control, behavioural willingness and behavioural responses and investigates their relationships through structural equation modelling. (3) Explore the influence mechanism of rural development participation on the capacity building of Yunnan rural women for sustainable development. Currently, there are research ideas that incorporate female subjectivity into community participation and participation capacity, but there are very few studies that take anti-poverty as the research background and take rural women in rural development participation as the research object with the goal of enhancing women’s capacity for sustainable development. This paper is the first to establish a framework for the influence mechanism of rural development participation on women’s sustainable and responsible capacity building based on the questionnaire of the former poverty-stricken areas in Yunnan Province and empirically analyses and verifies the influence of cognitive participation, affective participation and behavioural participation on the sustainable and responsible capacity building, as well as the moderating role of the external environment.

This study further collates and illustrates the study hypotheses that hold true, as shown in

Table 25. Summary of the hypothesis testing results.

Assumption No.	Hypothetical Content	Validation Results
H1	The higher the cognitive participation of women in rural development participation, the more active their behavioural participation;	Establishment
H2	The higher the level of emotional involvement and the more active the behavioural involvement of women in rural development participation;	Establishment
H3	The higher the level of cognitive participation of women in rural development participation, the more significant the increase in sustainable development capacity;	Established
H3a	The higher the perceived participation of women in rural development participation, the more pronounced the economic uplift of the household;	Establishment
H3b	The higher the level of cognitive participation of women in rural development participation, the more significant the human capital enhancement;	Establishment
H3c	The higher the level of cognitive participation of women in rural development participation, the more pronounced the social network enhancement;	Establishment
H3d	The higher the cognitive involvement of women in rural development participation, the more pronounced the enhancement of family roles;	Establishment
H4	The higher the level of emotional involvement of women in rural development participation, the more significant the increase in sustainable development capacity;	Not Established
H4a	The higher the emotional involvement of women in rural development participation, the more pronounced the economic uplift of the household;	Not Established
H4b	The higher the level of emotional involvement of women in rural development participation, the more significant the human capital enhancement;	Established
H4c	The higher the level of emotional involvement of women in rural development participation, the more pronounced the social network enhancement;	Not Established
H4d	The higher the level of emotional involvement of women in rural development participation, the more pronounced the enhancement of family roles;	Not Established
H5	The higher the level of behavioural participation of women in rural development participation, the more significant the increase in sustainable development capacity;	Establishment
H5a	The higher the behavioural involvement of women in rural development participation, the more significant the economic uplift of the household;	Establishment
H5b	The higher the level of behavioural participation of women in rural development participation, the more significant the human capital enhancement;	Establishment
H5c	The higher the level of behavioural participation of women in rural development participation, the more pronounced the social network enhancement;	Establishment
H5d	The higher the level of behavioural participation of women in rural development participation, the more pronounced the enhancement of family roles;	Establishment
H6	Women’s behavioural participation in rural development participation mediates between emotional participation and cognitive participation in building capacity for sustainable development;	Established
H7	Women’s participation in rural development in which the external environment positively moderates the impact of behavioural participation on capacity building for sustainable development;	Partially established
H7a	Women’s participation in rural development where the external environment positively moderates the impact of behavioural participation on the household economy;	Established
H7b	Women’s participation in rural development where the external environment positively moderates the impact of behavioural participation on human capital;	Establishment
H7c	Women’s participation in rural development in which the external environment positively moderates the impact of behavioural participation on social networks;	Not Established
H7d	Women’s participation in rural development where the external environment positively moderates the impact of behavioural participation on family roles.	Not Established

.

6. Results Model Policy Recommendations

6.1. Policy Recommendations for Enhancing Rural Women’s Ability to Poverty Eradicate on a Sustainable Basis

6.1.1. Improving the External Environment for Income Generation and Wealth Creation

Rural residents, especially women, need good policy support. For those in deep poverty who have been impoverished or returned to poverty due to illness, the “establishment of a database and card” should be used as the basis for categorisation, targeting and precise management. A health system based on the New Rural Cooperative System should be established, supplemented by medical insurance for major illnesses and emergency relief for illnesses. At the same time, commercial health insurance and charitable assistance are being actively developed. In terms of the organisational system, a rural hierarchical diagnosis and treatment model has been established, giving full attention to the role of primary medical and health care institutions and easing the pressure on the medical assistance system. Refine the funding arrangements for education and training according to the age structure, industrial background, ethnic composition and other realities of the deeply impoverished and carry out skills training adapted to the characteristics of local industries so as to block the intergenerational transmission of poverty with multilevel education and training. Constantly improve infrastructure construction, such as electricity and transportation, so that backwards rural areas are no longer closed off and can interact and communicate with the outside world. A fair and equitable policy guarantee system should be established so that all disadvantaged people can enjoy their basic rights, including political and economic rights, and all privileges should be rejected. Rural women should be guaranteed fair and equal participation in community activities. In order to change the persistence of poverty, rural women should be guided to enhance their awareness of self-improvement, provide reasonable conditions for rural women, provide a broader participation platform and improve their social status and self-identity, which will reduce the duration of persistent poverty. We should also ensure special treatment for special problems, differentiate between different categories of poverty, provide internal and external guarantees that are conducive to stabilising poverty eradication and adopt a differentiated approach to providing assistance to families with different levels of poverty.

6.1.2. Enhancing the Capacity for Sustainable Development

Provide high-quality education and training for rural women to avoid the intergenerational transmission of poverty. Based on the above research, which shows that mothers have a great influence on their children, their productive capacity and awareness of poverty alleviation and enrichment have an important impact on their children’s development; therefore, through the vocational skills training of parents, especially mothers, to improve their vocational skills, this can help them to improve from their ability to use their ideology and thus positively influence their children. Strengthen the guidance, especially for those disadvantaged farmers with a certain degree of labour ability, more to do a good job of education, to help them break free from the “wait and rely on” thought shackles; poverty alleviation practice experience is also enough to show that the most effective way to carry out poverty alleviation work is vocational education, according to the ethnicity of the poor group. Therefore, education and training funds can be divided according to the poor group’s ethnicity, age, industrial background, etc., so as to provide the poor group with appropriate skills training.

6.2. Improving the Participation Mechanism for Rural Women’s Income Enhancement and Wealth Creation

Enhancing female cognitive participation. The results of female cognitive participation found that many poor people have the ability to work but do not think they are poor, do not have the desire to rise out of poverty or have the desire to rise out of poverty but lack the courage and action. Women’s cognitive participation can be effectively enhanced by guiding them to know how to use different channels to obtain opportunities to increase their income and give full attention to their talents, to protect their rights and interests and to participate in various organisations and trainings. Publicity, technology promotion and service work have been achieved to enhance women’s awareness and ability to participate. At the same time, with the support and guidance of the government, women’s self-knowledge is enhanced through training and other means to better understand the programme, thus increasing women’s motivation to participate. Enhance women’s knowledge of various projects and industrial cultures through cultural training and other means.

Enhance women’s emotional participation. Cultivate women’s sense of responsibility and participation. Provide support and assistance to rural women’s organisations. Women’s federation organisations at all levels should bring the role of female leaders and elites into full attention to promote the better development of economic organisations. They should constantly improve the rules and regulations of rural women’s organisations, coordinate with the relevant units, provide financial and policy support for women’s organisations and build up a sense of joint participation and cooperation. Effective participation requires that women have the interest and ability to actively participate in poverty alleviation and enrichment industries, as well as the ability to reward correct cognition and emotion. Female cognition and female responsibility constitute the psychological foundation of female participation, which is the prerequisite for realising effective female participation. To enhance women’s cognitive and emotional input and stimulate women’s sense of responsibility, women should be given the necessary knowledge and skills training and be guided to closely link their active participation with their own interests and feel the importance of participating in the poverty alleviation and wealth-raising industries. At the same time, a sound mechanism for multi-party participation in poverty alleviation has been established to activate the spirit of initiative and endogenous motivation.

Enhancing women’s behavioural participation. Continue to maintain effective participation in rural industries. Increase the number of rural women’s internal employment. Make efforts to develop industries in which women are the dominant force, such as farming, handicraft manufacturing and processing, and give women technical guidance. The mode of industrial integration has been gradually improved, with the primary industry as the basis, gradually developing into secondary and tertiary industries, focusing on the development of new forms of industry so that rural industries are gradually developing in the direction of diversification and a focus point for industrial revitalisation is found. Encourage and guide rural areas to bring in industries with distinctive advantages, give full attention to the region’s strengths and explore high-quality resources, so that these resources can help the development of industries and gradually form distinctive leading industries. Increase training for women, especially for cadres of women’s federations and members of organisations, and improve training for rural women in management, skills, marketing and sales, so as to improve their overall quality and ability to adapt to the market.