Retain in the City, Return Flow, or Blind Direction: A Study on the Differentiation Mechanism of Migrant Workers’ Migration Willingness under the Background of China’s Strategy for Integrated Urban–Rural Development

Abstract

Against the backdrop of urban–rural integration development in China, the government has further strengthened its support for rural migrant workers’ urban employment and entrepreneurship, as well as their urbanization. Nevertheless, influenced by the current urban–rural development environment and public policies, a portion of rural migrant workers have evolved into two distinct groups: those with clear intentions to settle in cities or return to their hometowns and those trapped in a “neither-nor” situation, who are unable to fully integrate into urban life or seamlessly return to the countryside. This study, based on field investigations conducted in 21 cities across seven major geographical regions in China, reveals a ternary differentiation in migration intentions among rural migrant workers: “Retain in the cities”, “Return flow”, and “Blind direction”. In advancing the urban–rural integration strategy, the “clearly-oriented” groups, choosing to stay or return, can serve as dynamic forces in reconciling resources between urban and rural areas, thereby fostering closer urban–rural ties. Conversely, the “aimless wandering” group, characterized by stagnation, confusion, frequent mobility, and recurrent unemployment, poses numerous negative impacts on urban–rural interaction and coordination, hindering the in-depth integration of urban and rural areas to some extent. Drawing upon the internationally recognized Push-Pull Theory and Social Integration Theory within the field of population migration, this study conducts an empirical analysis of large-scale sample data to explore the characteristic factors and formation mechanisms underlying the “staying”, “returning”, and “blind direction” intentions, with a particular focus on the internal dynamics shaping and evolving the “aimless wandering” group. Finally, grounded in the empirical findings, this study advocates a collaborative approach between the Chinese government and various sectors of society to address the issue by promoting employment guidance, enhancing security measures, and other initiatives that encourage rural migrant workers to either stay in cities or return to their hometowns. It aims to provide policy recommendations for a definitive resolution of the rural migrant worker issue during the crucial period of urban–rural integration development.

1. Introduction

Labor migration has always been a topic of widespread concern among scholars both domestically and internationally. In international research, Canada and the United States primarily focus on the seasonal migration and rights protections of agricultural workers. In recent years, Europe has gradually shifted its research perspective to issues related to the labor rights of female migrants among international immigrants. In Southeast Asia, a considerable portion of international migrants work as domestic helpers in their host countries, making the protection of their rights a key issue in migration research in the region. Since the outbreak of the COVID-19 pandemic, research on the physical and mental health and medical insurance of migrants has gradually increased. Zhou Daming (2022) conducted fieldwork to investigate the migration characteristics of Vietnamese laborers who came to China in autonomous counties and explored how individual factors such as hometown and ethnicity differentiate Vietnamese laborers’ willingness to work in China under cross-border labor cooperation policies [1]; Alina Maciejewska et al. (2023) analyzed how the willingness to migrate among Ukrainian immigrants is formed in the context of war and how to incorporate the basic living needs of war refugees into local policy planning to achieve economic and social sustainability [2]; and distinct from international migration and internal migration in other countries, urban migrant workers in China emerged from China’s unique national conditions. Due to the long-standing planned economic management system of “each performing its own duties” in industry and agriculture, the Chinese government adopted a strict household registration management system to restrict the floating population between urban and rural areas, resulting in a dualistic development pattern between urban and rural areas. Since the reform and opening-up, improved agricultural productivity has led to a large rural population surplus. The gradual relaxation of household registration restrictions has allowed some rural residents to migrate to cities. Compared to engaging in agricultural production in their rural hometowns, working in cities offers higher incomes and more convenient living conditions. Therefore, a large number of rural residents have begun to migrate to cities for work on a large scale. Limited by their household registration status and their own cultural and skill levels, most of this group engages in low- and medium-end manufacturing and service industries after entering the cities. As a result, this part of the working population has gradually become an independent “third element” group separate from urban residents and rural villagers—urban migrant workers (Gan Mantang, 2003) [3].

Related domestic and international research mainly focuses on four aspects, including migration patterns, basic characteristics, influencing factors, and the effects of migrant populations. In terms of the classification of migration patterns, Zelinsky W (1971) divided population migration into five forms: rural-to-urban migration, domestic-to-remote area migration, international migration, urban-to-urban and intra-urban migration, and circular migration [4]. In China, Li Qiang (2004) divided the family patterns of migrant workers in cities into five types from the perspective of family structure: single-child migrant type, siblings’ migrant type, couple-separated type, couple–child-separated type, and whole-family migrant type [5]. In addition, Ren Yuan (2006) found that the migrant population in Shanghai exhibited a gradual and cumulative residency pattern [6], while Yang Juhua et al. (2013) explicitly pointed out that the trend in migrant population migrating with their families will become increasingly common, and the process of family based migration and reunification will follow a gradual hierarchy [7]. Under the background of population aging and urban–rural integration, Zhang Yaojun et al. (2024) classified inter-provincial population mobility in China into four major modes: rural–rural, rural–urban, urban–rural, and urban–urban. The floating population is shifting from a single direction to a diversified flow trend [8].

In terms of the basic characteristics of the migrant population, both T.W. Schultz (1961) and Narasimhan S (1995) emphasized the special role of human capital, believing that human capital is a combination of various stocks, such as cultural knowledge, labor management skills, and health quality embedded in the labor force. Laborers with stronger human capital tend to have stronger migration capabilities [9,10]. In China, Tian Ming (2013) believes that the number of years of working away from home largely represents the accumulation of human capital, such as work skills and experience [11]. As the number of years of migrant workers’ employment increases, their job proficiency and market competitiveness will also improve accordingly. The household registration system is a special national condition that has existed in China for a long time. Yu Pei et al. (2024) focused on the impact of public services, such as health care, children’s education, and social security, which are linked to the household registration system, on the willingness of migrant workers in cities to permanently migrate. They believe that the separation of urban public services from the household registration system should be achieved as soon as possible to realize equal access to public services, thereby enhancing the willingness of migrant workers to settle down in cities [12].

In terms of research on the influencing factors of population migration, Martin P (2003) summarized the main factors into three types: demand pull factors at the destination, supply push factors from the origin, and network factors connecting the origin and the destination [13]. Michael P. Todaro (1969) established the Behavioral Model of Rural–Urban Labor Migration, which identified two variables in rural labor migration to cities: the actual income difference between urban and rural areas and the possibility of obtaining urban jobs. It proposed that the expected income gap is the internal reason and fundamental driving force for rural–urban labor migration [14]. In China, studies have found that social security conditions, such as pension insurance and children’s education, have a significant impact on migrant workers’ willingness to become urban residents (Wang Guixin et al., 2015) [15], and higher socio-economic status helps to enhance migrant workers’ willingness to urbanize (Xu Yanhui et al., 2022) [16]. Furthermore, Yue Zhongshan et al. (2011) believed that the social network of non-relatives between migrant workers and urban residents significantly positively affects the cultural and psychological integration of migrant workers [17]. Ye Jingzhong et al. (2009) argued that families with only left-behind children, women, and the elderly have a stronger desire for migrant workers to return [18]. Zhou Chuang et al. (2022) have used fixed-term contracts as a measure of the job stability of migrant workers. Their research found that having a stable job can enhance the persistent income expectations and social participation of migrant workers in cities, especially among the new generation and those with higher education levels, thereby strengthening their willingness to permanently migrate [19]. Zhou Yihu et al. (2023) argue that migrant workers’ decisions to migrate are determined by the coupling effect of social pressure and their own behavioral preferences [20].

Regarding the social effects of population mobility, S Goldstein (1993) argued that migration can alleviate population pressure in rural areas, achieving a better balance between skill availability and demand in the labor market, and is an important factor in urbanization and modernization [21]. In China, Huang Ping (1997) used Anthony Giddens’ structuration theory to explain the reasons for rural population migration for employment, arguing that migrant workers entering cities for employment is a process of dualization between migrant workers and China’s urban–rural social structure [22], resulting in the evolution of contemporary China’s urban–rural dual social structure into a ternary social structure of peasants-migrant workers-citizens (Li Qiang, 2004) [5]. Urbanization and return migration are two directions to end the migrant worker system (Liu Chengbin, 2017) [23]. Therefore, it is urgent to promote the rapid transformation of migrant workers’ social policies from basic protection to developmental support, gradually enhancing migrant workers’ human capital and development capabilities, gradually increasing migrant workers’ political capital and social participation, and strengthening their role transformation from laborers to citizens (Zhang Ruli et al., 2018) [24]. Beyond that, urban connectivity offers a macro-level perspective on the motivations and characteristics of population migration. Lin Jinyao et al. (2022) utilized multi-source data to analyze the pivotal role of urban connectivity in regional planning and delved into the interplay between urban connectivity networks and population migration patterns [25]. Fu Miao et al. (2024) focused on the influence mechanism of urban agglomeration’s spatial structure on the spatio-temporal evolution of aging, discovering that population mobility and age structure turnover significantly impact the differentiation of urban agglomeration’s spatial structure [26]. Li Zhenlong et al. (2021) introduced a Global Multi-scale Place Connectivity Index (PCI), representing spatial interactions across regions based on geotagged tweets. Their study revealed a strong positive correlation between multi-scale PCIs at the county level in the United States and SafeGraph’s population mobility records (with a 10% penetration rate in the US population), as well as Facebook’s Social Connectedness Index (SCI), a popular connectivity index based on social networks. They also proposed the supportive role of PCI in future research on urban interactions and population migration worldwide [27].

Currently, migrant rural workers remain the core human resource for urban employment in China. As the most proactive and dynamic production factor, they continue to play an irreplaceable pillar role in accelerating urban construction, driving urban development, and promoting urban transformation and upgrading. At the same time, they have triggered profound changes and integrated development in urban and rural societies. However, despite being industrial workers who contribute to urban construction, they are seen as a special group separate from urban residents. Living in urban society, they struggle to access equal urban public services, leading to severe social issues within the migrant rural worker population. Under the influence of various urban–rural integration policies, migrant rural workers have not significantly increased their willingness to return to their hometowns for entrepreneurship and employment, but instead, some unexpected consequences have emerged. These are mainly manifested in the trifurcation of migration intentions among migrant rural workers: strong migration intentions, vague migration intentions, and weak migration intentions. Correspondingly, the migrant rural worker population has fragmented into three groups: those who stay in cities, those who are uncertain about their direction, and those who return to their hometowns. Previous academic research has accurately described the universal characteristics and general patterns of population mobility or migrant rural worker migration, playing a significant theoretical guiding role in effectively addressing issues such as inter-regional population mobility, international migration, and migrant rural worker migration. However, overall, relevant research still lacks the necessary attention and deep exploration into the multifaceted differentiation of migrant rural workers’ migration intentions before their mobility behaviors occur. In particular, there is a notable lack of focus on a special group objectively present among migrant rural workers—those who are uncertain about their direction. Therefore, based on an analysis of the integration factors and restrictive factors influencing the differentiation of migrant rural workers’ migration intentions, this study separately explores the characteristics and formation mechanisms of migrant rural workers who stay in cities, return to their hometowns, and are uncertain about their direction. At the same time, this study emphasizes investigating how migrant rural workers who are uncertain about their direction differentiate from the traditional “stay-return” migration framework. By comparing their connections and differences with migrant rural workers who have a clear direction, this study analyzes the transformation mechanisms and driving factors that convert migrant rural workers who are uncertain about their direction into those with a clear direction.

2. Theoretical Basis and Hypothesis

2.1. Theory of Push-Pull

The “push-pull” theory is internationally recognized as the earliest theory of population migration. This theory can be traced back to Ravenstein’s “The Laws of Migration”, published in 1885, which summarized the mechanisms, structures, and spatial characteristics of population migration and proposed that “population migration is primarily motivated by the desire to improve one’s economic situation”, a viewpoint considered the origin of the Push-Pull theory. Subsequently, Herber introduced the fundamental ideas and core concepts of the “push-pull” theory in 1938, arguing that population migration arises from a series of “pushes” and “pulls” that interact to shape migration patterns. The primary goal of migration is to improve living conditions, with favorable factors in the destination region acting as pulls and unfavorable conditions in the region of origin serving as pushes. Thus, migration is determined by these two forces acting in tandem. In the late 1950s, Bogue further refined the “push-pull” theory from a kinematic perspective, comprehensively and concisely outlining 12 push factors and 6 pull factors, thereby enhancing and supplementing the theoretical framework. Building upon this foundation, Lee made the most significant contribution to the development of the “push-pull” theory. In 1966, he systematically summarized the theory and established a comprehensive analytical framework, aiming to explain the attractions and resistances encountered by individuals during migration. Lee also introduced intermediate obstacle factors, categorizing the determinants of migration behavior into four aspects: destination-related factors, origin-related factors, various intermediate obstacles, and individual factors. His most notable contribution lies in the explicit description of intermediate obstacles, including distance, physical barriers, linguistic and cultural differences, as well as the migrant’s personal value judgments regarding these factors. Ultimately, Lee concluded that population migration is the result of a multifaceted interplay of factors. By this point, the “push-pull” theory had matured significantly and was widely applied in various studies of population migration.

The “push-pull” theory establishes a basic framework for the factors influencing population migration. For instance, “push” factors encompass population growth, low living standards, lack of employment opportunities, and political exclusion, among others, while “pull” factors include labor demand, favorable economic opportunities, ample land carrying capacity, political freedom, and so forth. In the context of China’s migrant worker migration, the “push-pull” theory holds fundamental advantages. Since the 1980s, with the rise of the migrant worker tide in China, Chinese scholars have integrated the “push-pull” theory with the practical migration patterns of Chinese migrant workers, identifying migration laws with distinct Chinese characteristics. This study organizes the “push-pull” factors discussed above in

Table 1. “Push-Pull” factors in the rural-to-urban migration of Chinese migrant workers.

Rural Push Factors	Rural Pull Factors	Urban Push Factors	Urban Pull Factors
Low rural income	Stay-at-home families need care	High cost of living	More earning opportunities in cities
Agricultural labor is too strenuous	Unwillingness to give up rural land	Difficulty in finding and adapting to work	Abundant job opportunities
Lack of development prospects	Low cost of living	Separation from loved ones and emotional distress	Many development opportunities
Limited and poor career choices	Good interpersonal relationships	Poor living environment	Ability to learn skills
Inability to acquire skills	Sense of satisfaction in rural areas	Lack of security and rights protection	Desire to experience city life and broaden horizons
Backward basic public services	Relatively fast rural development	Feeling of exclusion from urban society	Access to city’s public services

.

Currently, the new generation of migrant workers has become the main force of migrant workers in cities. Compared with their parents, the factors influencing their migration intentions are more complex. While more people choose to return home to start their own businesses, there emerges a group of “blind” individuals who are uncertain whether to stay in the city or return home in the future, resulting in a ternary differentiation structure of migrant workers’ migration intentions, namely, staying in the city, returning home, and being uncertain. This paper argues that the emergence of this structure is the result of the interplay between the integration and restrictive factors in the cities where they work. Based on this, this paper analyzes the “pushes” and “pulls” that influence the migration choices of migrant workers in China’s cities and villages, respectively, and explores the internal mechanism of the ternary differentiation of migrant workers’ migration intentions. In this study, the reference significance of push-pull theory to the study of migrant workers’ willingness to migrate differentiation is shown in Figure 1.

2.2. Theory of Social Capital

On the basis of relational methodology, Pierre Bourdieu first proposed the concepts of “field” and “capital”, defining “field” as a social relation network interconnected by various social relations and dividing capital into economic capital, cultural capital, and social capital. Bourdieu believed that different social elements serve as nodes in the field network, which interact and change by occupying distinct positions, with social capital being the driving force behind these changes. James S. Coleman extended the concept of social capital as an extension of social resources, defining social resources as the persistent social relations formed by actors to achieve their own interests, while social capital is these social resources possessed as individual capital assets. Building on Coleman’s work, Robert D. Putnam elevated social capital from the individual level to the collective level, arguing that social capital stems from dense networks of civic participation, which are strengthened through various means of punishing those who undermine trust among people or engage in inappropriate behaviors. This civic spirit and participation embody social capital. Nan Lin, through his research on social networks, developed the theory of social capital, distinguishing resources into personal resources and social resources. Personal resources refer to wealth, knowledge, status, and other factors attached to individuals, while social resources are the resources embedded in an individual’s social network, representing potential accessible resources connected through social networks. Social capital, on the other hand, refers to the resources that an individual actually acquires and utilizes through social relation networks.

The theory of social capital delves into the formation mechanisms of migrant workers’ migration preferences from a structural perspective. It emphasizes that social capital, as the total resources an individual can mobilize within their social relation networks, has core elements, including the construction of social relation networks, the maintenance of trust and norms of reciprocity, and the ability to engage in social participation and resource integration, all of which profoundly influence the entire process of migrant workers’ migration decisions. Specifically, the social relations and resources accumulated by migrant workers in rural social networks and in the cities where they work serve as a bridge to reduce information asymmetry in migration decisions, while emotional support enhances their psychological motivation to make clear choices about returning to their hometowns or staying in the cities. Simultaneously, the trust mechanisms and norms of reciprocity formed through long-term interactions during migration facilitate the flow of critical resources, such as funds and information, constructing a safety net during the migration process. After entering cities, some migrant worker groups actively participate in community activities to expand their social relations, thereby realizing the reproduction of social capital. This process not only aids their social integration in cities but also enhances their social status and life satisfaction through resource integration, thereby reinforcing their willingness to reside permanently. In summary, the complex interactions and profound impacts of social relation networks, trust and reciprocity, social participation, and other factors revealed by the theory of social capital on migrant workers’ migration preferences and subsequent urban integration provide a solid theoretical foundation for understanding the phenomenon of rural–urban migration. In this study, the reference significance of social capital theory for the study of migrant workers’ willingness to migrate differentiation is shown in Figure 1.

2.3. Theory of Social Integration

The theory of social integration is a comprehensive assimilation theory with cross-disciplinary, compatibility, and openness. Based on a multi-dimensional and multifaceted research perspective, it objectively presents the influencing factors, interactive characteristics, and dynamic processes of special social groups, such as immigrants, in terms of social integration. For instance, Xu Jing et al. (2024) conducted a thorough analysis of the intermediary role of social support in enhancing the social integration of mobile elderly populations, exploring the supportive effects of social integration on alleviating loneliness among immigrant seniors [28]. Meanwhile, Zhang Ping et al. (2024) studied the impact of cultural diversity on the social integration of immigrants, finding that a low level of social integration can increase obstacles for immigrants seeking to start businesses in their host localities [29]. It is a systematic summary and scientific generalization of the basic connotations, behavioral characteristics, main patterns, integration standards, and general laws of social integration. In this context, promoting high-quality urban social integration is an important topic in the study of migrant workers’ willingness to migrate to cities. Although migrant workers have their unique essential attributes and diverse integration methods, the construction of their integration willingness and the implementation of integration behaviors still cannot escape the requirements of general laws and the role of basic characteristics of social integration. Therefore, it is of great significance to use the analytical framework and research methods of social integration theory to examine, analyze, and explain the differentiation of migrant workers’ migration willingness, which can effectively promote the development and improvement of research on the differentiation of migrant workers’ migration willingness. The theory of social integration has the following values in explaining the differentiation of migrant workers’ migration willingness:

First, it provides theoretical support for defining the two core concepts of integration and limitation. As special immigrants in urban society, migrant workers’ migration willingness will undergo drastic changes under the background of drastic economic and social changes, manifesting as various restrictive forces gradually emerging from the dimension of social exclusion, while integration forces are also gradually growing and developing into the core force to promote the urban integration of migrant workers, leading to more complex and diverse migration willingness of migrant workers. Migrant workers are not only facing urban exclusion or urban acceptance in their first transfer from rural to urban areas, but more importantly, the inherent power of urban–rural restrictions, urban–rural interactions, and urban–rural integration that accompanies their work experience. This requires that the research on the differentiation of migrant workers’ migration willingness needs to further broaden its horizons, shifting from focusing on localized examinations of micro-factors, such as migrant workers’ individuals and families to comprehensive exploration of macro-factors such as social systems, social environments, social structures, and social integration.

Second, it can encourage the academic community to pay more attention to applying the core connotations and dialectical thinking of social integration theory to analyze and study the differentiation of migrant workers’ migration willingness. In the theory of social integration, exclusion and integration are both distinct and separated from each other, yet also interconnected and interactive, displaying the clear characteristics of mutual complementarity and transformational integration. This has important enlightenment functions for the theoretical research on migrant workers’ migration. By referring to the logical thinking and dialectical thinking of social integration theory, this article deeply explores and scientifically integrates the major contradictions and prominent issues in the migration process of migrant workers, studies and summarizes the specific reasons and temporal-spatial characteristics behind these contradictions, especially focusing on the various limitations and integration effects generated by migrant workers’ interactions with urban populations, society, environment, culture, and other factors, and carefully sorts out the specific relationships of mutual interaction, mutual penetration, and mutual promotion among various temporal–spatial dimensions in the work history of migrant workers.

Third, it is conducive to referring to and drawing on the indicator system of social integration theory to guide and improve research work. The indicator systems related to social integration at home and abroad are complex and diverse, providing rich and sufficient reference materials for the in-depth development of this research. Based on referring to various indicator dimensions of social integration theory at home and abroad, this article specifically divides various indicators in the work history of migrant workers into four dimensions: individual, family, work, and interaction. Scientific collation and analysis of these factor indicators reveals the compound effects of basic factors on the ternary differentiation of migrant workers’ migration willingness. In this study, the reference significance of social integration theory for the study of migrant workers’ willingness to migrate differentiation is illustrated in Figure 1.

Based on the corresponding dimensions of various indicators, this study mainly categorizes various characteristics in migrant workers’ work experience into individual, family, work, and interaction levels, comprehensively examines the combined effects of basic factors on the differentiation of urban migrant workers’ willingness to migrate and establishes the following research framework.

3. Sample Situation and Model Establishment

3.1. Sample Situation

The primary data for this study comes from the National Natural Science Foundation project “Research on the Impact of Localization Factors on the Urban Integration of Migrants during Urbanization”, hosted by the School of Social Development and Public Policy at Beijing Normal University. The research team conducted a sample survey of migrant workers and their families in 21 cities across seven major regions in China, including east China, south China, north China, central China, southwest China, northwest China, and northeast China. Based on the results of the China Migrants Dynamic Survey 2018 (CMDS2018), the research team selected cities with significant proportions of migrant workers flowing in or out, including Shanghai, Wuxi, Jiangyin, Shenzhen, Jiangmen, Heshan, Beijing, Langfang, Bazhou, Wuhan, Xiangyang, Zaoyang, Chengdu, Mianyang, Chongzhou, Xi’an, Xianyang, Xingping, Changchun, Jilin, and Jiaohe. From the perspective of survey regions, the sample distribution across the seven surveyed areas is relatively even. Among them, there are 585 valid samples from county-level cities, accounting for 18.6% of the total; 1144 valid samples from prefecture-level cities, accounting for 36.5% of the total; and 1408 valid samples from sub-provincial cities and above, accounting for 44.9% of the total. This study adopts differentiated sampling in cities of different levels to present the agglomeration effects of migrant workers in various urban tiers.

Firstly, stratified random sampling is adopted for the survey. Step 1: Randomly select a province from the municipalities directly under the central government and provinces for the next level of survey city selection. Step 2: Design a sampling frame based on the urban administrative division level, namely, design two types of sampling frames for sub-provincial and above cities and prefecture-level cities in each province, and then randomly select one sub-provincial and above city and one prefecture-level city from the two sampling frames. Step 3: Randomly select an investigated county-level city from the selected prefecture-level city.

Secondly, regarding the scale of questionnaire distribution:

Step 1: Set the scale of questionnaire distribution based on city levels. Specifically, distribute 250 questionnaires in provincial or sub-provincial cities, 200 questionnaires in prefecture-level cities, and 150 questionnaires in county-level cities.

Step 2: Allocate the specific number of questionnaires in each surveyed city according to its administrative areas. Using the population census data provided by the “Sixth Population Census” of China, calculate the long-term floating population by subtracting the permanent resident population from the total population of each administrative area. Then, distribute the sampling quota for each area based on the ratio of floating population between administrative areas.

Finally, during the field survey, whole district sampling and accidental survey methods are employed:

Step 1: Based on urban planning and functional structures, investigators are dispersed to densely populated areas such as industrial parks, residential areas, leisure areas, and shopping malls to ensure the diversity of sample occupational sources and the internal differences within the survey areas.

Step 2: When conducting surveys in specific locations, investigators use non-probability accidental sampling to interview migrant workers who are currently working or seeking work in the area.

Simultaneously, rigorous quality control procedures are implemented throughout the field survey, data entry, and cleaning processes. Survey instructors are responsible for collecting and reviewing questionnaires on the same day, making modifications to problematic questionnaires or arranging follow-up visits for supplementation the next day. The data entry process strictly adheres to the “double entry” principle, ensuring the quality of the survey questionnaires. Furthermore, two rounds of data cleaning procedures are conducted by professional members of the research team, with careful examination of the internal logic of the data. Any logical issues are recorded and verified against the original questionnaires for modification. For a few error questionnaires that cannot be modified, respondents are contacted for a second visit to make corrections, thereby reducing mechanical errors in the survey data.

A total of 4250 questionnaires were distributed to migrant workers, and 3137 valid questionnaires were ultimately recovered, with a recovery rate of 73.81%. Furthermore, this study strictly adheres to the requirements related to personal data protection and privacy protection outlined in the Law of the People’s Republic of China on Scientific and Technological Progress, the Law of the People’s Republic of China on the Protection of Personal Information, and the Opinions on Strengthening the Governance of Science and Technology Ethics. The writing of this paper has been approved by the Ethics Committee of Shenyang Normal University. This study follows the principle of voluntary participation, where each participant voluntarily participates in the survey and signs an informed consent form. Participants can withdraw from the survey at any time during the questionnaire completion process. When collecting questionnaire data, researchers do not involve respondents’ personal names and ages, only using codes to store respondent information. The basic statistics of the total sample for the migrant worker survey are shown in

Table 2. Basic statistics of the total sample of the migrant worker survey (N = 3137).

Descriptive Variable (Unit)	Proportion or Mean
Gender	Male	60.5%
	Female	39.5%
Age (years)	33.86
Education Level	Primary School	15.8%
	Junior High School	34.2%
	High School, Vocational High School, Technical Secondary School	26.4%
	College	12.1%
	Bachelor’s Degree and Above	11.5%
Marital Status	Unmarried	38.1%
	Married	58.8%
	Others, including divorced and widowed	3.1%
Location Origin	Urban	4.6%
	County Town	16.2%
	Small Town	17.6%
	Village	61.6%
Number of Household Members (persons)	4.29
Family Members Left Behind	None Left Behind	44.3%
	Family Members Left Behind	55.7%
Family Members Migrating Together	Alone	49.0%
	With Family Members	51.0%
Family Contracted Land	No Contracted Land	53.8%
	Contracted Land	46.2%
Family Home Ownership	Only Home Village	73.3%
	Only Local Area	2.2%
	Both Places	7.6%
	Neither Place	16.9%
Annual Household Income (yuan/year)	73,253.73
Annual Household Surplus (Yuan/year)	26,873.53
Occupational Distribution	National and Social Managers	0.7%
	Managers	1.4%
	Private Business Owners	3.1%
	Professionals	8.4%
	Clerical Staff	5.3%
	Self-employed Households	21.9%
	Business and Service Staff	36.7%
	Industrial Workers	12.4%
	Agricultural Workers	2.3%
	Not Sure	7.7%
Current Employment Status	Self-employed Worker	25.6%
	Domestic Helper	2.0%
	Others	5.6%
Current Monthly Income (Yuan/month)	4081.76
Working Days per Month (days/month)	26.54
Duration of Working Away from Home (years)	9.98
Current Work Distance	Within the Same City	14.1%
	Other Cities within the Province	38.9%
	Other Provinces	47.0%
Urban Life Adaptation	Very Accustomed	13.5%
	Accustomed	52.7%
	Average	27.7%
	Not Accustomed	5.6%
	Very Not Accustomed	0.5%
Location Identity	Locals	11.0%
	People from Hometown	58.6%
	Both Locals and People from Hometown	24.4%
	Neither locals nor people from hometown	6.0%
Main Communication	Locals	46.1%
	Fellow Townsmen	28.9%
	Other Outsiders	25.0%
Regular Contact with Friends	7.43
Days Spent at Home Per Year (days/year)	23.37
Migration Intention	Stay in the Current Working City	30.1%
	Stay in the City of One’s Hometown	16.1%
	Go to Another Place to Explore	6.9%
	Not Sure	17.8%
	Return to the Rural reas of One’s Hometown	18.1%
	Return to Small Towns, including County-Level Central Towns, of One’s Hometown	11.0%

.

3.2. Variable Selection

3.2.1. Dependent Variable

Based on the core issue and research design, this study takes the migration intention of migrant workers in cities as the dependent variable for specific research. The operationalization is divided into clear-cut intentions and vague intentions.

Clear-cut intentions: This refers to migrant workers who clearly express their intention to stay in the city or return to the countryside in the future. Among them, staying in the city includes two variables: “Staying in the current working city” and “Staying in the city of one’s hometown”. Returning to the countryside includes two variables: “Returning to the rural areas of one’s hometown” and “Returning to small towns, including county-level central towns, of one’s hometown”.

Vague intentions: This refers to migrant workers who are not sure where they will settle down in the long run, including two variables: “Not sure” and “Going to another place to explore”.

3.2.2. Independent Variables

This study takes the characteristic factors that affect the migration intention of migrant workers in cities as independent variables, which are divided into individual characteristics, family characteristics, work characteristics, and interaction characteristics. The assignment of each variable and its subdivisions is shown in

Table 3. Selection, definition, and assignment of independent variables in the quantitative analysis of this study.

Variable Level	Variable Name	Definition and Assignment of Variables
Individual Characteristics	Gender	Male = 1
		Female = 2
	Age	18–29 years old = 1
		30–39 years old = 2
		40–49 years old = 3
		50–59 years old = 4
		60 years old and above = 5
	Education Level	Primary school and below = 1
		Junior high school = 2
		High school (vocational high school, technical secondary school, technical school) = 3
		College = 4
		Bachelor’s degree and above = 5
	Marital Status	Unmarried = 1
		Married = 2
		Others (including divorced, widowed) = 3
	Geographical Origin	Rural area = 1
		Small town = 2
		County town = 3
		City = 4
Family Characteristics	Family Members Left Behind	No family members left behind = 1
		With family members left behind = 2
	Family Members Accompanying	Alone = 1
		Together with family members = 2
	Contracted Farmland	No contracted farmland = 1
		With contracted farmland = 2
	Home Ownership	Only have a house in the hometown = 1
		Only have a house in the local area = 2
		Have houses in both places = 3
		Have no houses in both places = 4
	Family Annual Income in the Past Year	Below CNY 30,000 = 1
		CNY 30,000–59,999 = 2
		CNY 60,000–89,999 = 3
		CNY 90,000–119,999 = 4
		CNY 120,000 and above = 5
Work Characteristics	Duration of Migration	1–3 years = 1
		4–6 years = 2
		7–9 years = 3
		10–12 years = 4
		13 years and above = 5
	Current Work Distance	Within the same city = 1
		Other cities within the province = 2
		Other provinces = 3
	Current Employment Status	Employee = 1
		Employer = 2
		Self-employed worker = 3
		Domestic helper = 4
		Others = 5
	Current Monthly Income	Below CNY 2000 = 1
		CNY 2000–3999 = 2
		CNY 4000–5999 = 3
Interactive Characteristics	Location Identity	Native of the town = 1
		Local resident = 2
		Both native and local resident = 3
		Neither native nor local resident = 4
	Main Communication Partners	Fellow townsman = 1
		Local people = 2
		Other outsiders = 3
	Number of Frequent Friends	0–3 = 1
		4–7 = 2
		8–11 = 3
		12 and above = 4
	Number of Times Returning Home Every Year	0–1 time = 1
		2–3 times = 2
		4–5 times = 3
		5 times and above = 4
	Number of Days Staying at Home Every Year	0–9 days = 1
		10–19 days = 2
		20–29 days = 3
		30 days and above = 4

.

Individual Characteristics of Migrant Workers: It is subdivided into five variables: gender, age, marital status, education level, and geographical origin. The flow of labor force towards urban areas with relatively higher incomes is a primary manifestation of interest-driven migration. Due to differences in individual capabilities, those with relatively stronger abilities tend to have higher mobility in pursuit of greater benefits, while those with weaker abilities exhibit lower mobility, either working locally or staying in rural areas to engage in simple, traditional agricultural labor and low-income jobs. Among these factors, migrant workers’ gender, age, and education level are commonly used indicators to measure labor capabilities, while marital status relates to the form of labor mobility and may impact individual migration outcomes. Furthermore, given the complexity of the current migrant worker source locations, there are variations in how migrant workers perceive their own capabilities and attributes, as well as in their understanding of urban and rural locations. Therefore, the attribute of location source is selected as an identity characteristic of migrant workers at the individual level.

Family Characteristics of Migrant Workers: It is subdivided into five variables: family members left behind, family members accompanying the migrant worker, contracted farmland, home ownership, and family annual income in the past year. For migrant workers entering cities, the distribution structure of family members objectively reflects the labor allocation strategy of the entire migrant worker family. Specifically, the situation of family members left behind can reveal the restraint effect of population in the hometown, while the situation of family members migrating together can demonstrate the concentration state of family members during the work process. These family structure elements play a regulatory role in the individual migration strategies of migrant workers. At the level of family economics, land has long served as a traditional safeguard for farmers’ survival, and whether a family owns contracted farmland is an important factor influencing migrant workers’ decision to work outside their hometown, often affecting their long-term migration decisions. The ownership of housing is also a reflection of a family’s economic strength, and housing has a significant effect on enabling migrant workers to stabilize in a certain location. In addition, last year’s annual income, as a direct indicator of family living standards, can demonstrate the direct role of family economic conditions on migrant workers’ migration. The last year’s family income indicator is represented by the total income from farming, working, and welfare benefits of all members in the migrant worker family in 2016.

Work Characteristics of Migrant Workers: It is subdivided into five variables: duration of migration, current work distance, current employment status, current monthly income, and the level of the city where they are currently working. From the perspective of work time and space, migrant workers’ duration of migration and distance traveled measure the effectiveness of their work in terms of time and distance, respectively. The cumulative effects of these factors influence migrant workers’ future migration decisions, especially when combined with other factors, resulting in more pronounced compound effects. Most migrant workers in cities are employed in the service industry or as industrial workers, and their most tangible labor reward is financial income. Work remuneration plays a crucial role in supporting other economic activities and sustaining urban life in the long run. Additionally, employment status at the social level is significant for migrant workers to stabilize in a job position. Furthermore, as mentioned in the research background, there are still various differences among cities in China, and migrant workers perceive different work effects in cities of different tiers.

Interaction Characteristics of Migrant Workers: Five variables are selected: geographical identity, main communication partners, regular contacts with friends, number of home visits per year, and days spent at home per year. As migrant workers continue to experience the influence of cities, it is not merely a process of active behavioral integration but also accompanied by an inner selection process. The location identity constructed during the migration process is potentially linked to future migration directions. Meanwhile, the objects of interaction and the number of friends in urban interactions clearly reflect the effects of migrant workers’ social participation and social relations. Additionally, the “migratory bird”-like behavior of migrant workers returning home for visits is the most common form of urban–rural interaction. The regular return of migrant workers to interact with family members in their hometown has a cumulative effect on their future migration choices.

3.3. Model Establishment

This study explores the migrant workers’ migration intentions, which is a multi-category variable. Therefore, the method of factor mining adopted is the multinomial logistic regression model. Specifically, through the comprehensive matching of the multinomial regression model, a quantitative analysis of the differentiation of migrant workers’ migration intentions is achieved. In the process of model building, migrant workers’ migration intention is first set as a dummy variable. If the interviewee chooses the option of “staying in the working city” or “returning to the city of their hometown”, it indicates that the migrant worker has the intention to stay in the city, and this variable is assigned a value of “1”; if the interviewee chooses the option of “not sure” or “looking for another place”, it indicates that the migrant worker has a vague intention, and this variable is assigned a value of “2”; if the interviewee chooses the option of “returning to the rural area of their hometown” or “a small town (including the central town of the county seat) of their hometown”, it indicates that the migrant worker has a weak migration intention, and this variable is assigned a value of “3”. In the model of the differentiation of migrant workers’ migration intention, the degree of their migration intention is jointly affected by multiple factors, including individual factors, family factors, work conditions, and interactive effects. However, the institutional factors that are commonly faced at the macro level are not introduced into the model. Therefore, the final overall model established in this study is as follows:

$$\mathrm{L}=\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{i}\mathrm{t}[\mathrm{P}(\mathrm{y}=\mathrm{j})]=\mathrm{l}\mathrm{n}[{\displaystyle \frac{\mathrm{p}(\mathrm{y}=\mathrm{j})}{\mathrm{p}(\mathrm{y}=\mathrm{i})}}]=\mathsf{\alpha }+{\mathsf{\beta }}_{1\mathrm{j}}{\mathrm{x}}_1+{\mathsf{\beta }}_{2\mathrm{j}}{\mathrm{x}}_2\cdots {\mathsf{\beta }}_{\mathrm{n}\mathrm{j}}{\mathrm{x}}_{\mathrm{n}}+\mathsf{\mu },\mathrm{j}\ne \mathrm{i}$$

(1)

In Formula (1), y represents the migrant workers’ migration intention choice, P(y = j) is the probability of occurrence when the migration intention choice is j, P(y = i) is the probability of occurrence when the migration intention choice is i, and P(y = j)/P(y = i) is the ratio of the probability of occurrence of one situation to the probability of occurrence of another situation, which is referred to as the odds ratio (odds); x (i = 1, 2, 3... n) is the independent variable, which is the influencing factor of the differentiation of migrant workers’ migration intention; α is the constant term; βnj is the coefficient to be estimated; and u is the random error term. Based on this, βnj represents the change in Logit [P(y = j)] brought by a unit change in the corresponding independent variable, and its interpretation is not very intuitive. Therefore, this study takes the exponential of e on both sides of the equation to obtain a new model function:

$$\mathrm{o}\mathrm{d}\mathrm{d}\mathrm{s}={\displaystyle \frac{\mathrm{p}(\mathrm{y}=\mathrm{j})}{\mathrm{p}(\mathrm{y}=\mathrm{i})}}={\mathrm{e}}^{\mathsf{\alpha }}\times {\mathrm{e}}^{{\mathsf{\beta }}_{1\mathrm{j}}{\mathrm{x}}_1}\times {\mathrm{e}}^{{\mathsf{\beta }}_{2\mathrm{j}}{\mathrm{x}}_2}\cdots {\mathrm{e}}^{{\mathsf{\beta }}_{\mathrm{n}\mathrm{j}}{\mathrm{x}}_{\mathrm{n}}}\times {\mathrm{e}}^{\mathsf{\mu }},\mathrm{j}\ne \mathrm{i}$$

(2)

In Formula (2), the value of eβnj is the odds ratio (OR value), representing the multiple of the change in odds caused by a unit change in the independent variable. It can significantly explain the meaning of the regression coefficient and more intuitively reflect the degree of influence of the independent variable on the dependent variable. In the operation of the model, when the OR value is greater than 1, the dependent variable also increases with the increase in the independent variable, indicating that the influence of the independent variable is more significant; conversely, when the OR value is less than 1, the influence is weaker. Therefore, based on the above model construction and transformation analysis, this study takes the migration intention = 2, i.e., “unclear intention”, as the benchmark category for comparison, and establishes two generalized logistic regression models for analysis, specifically as follows:

$$\mathrm{o}\mathrm{d}\mathrm{d}\mathrm{s}={\displaystyle \frac{\mathrm{p}(\mathrm{y}=1)}{\mathrm{p}(\mathrm{y}=2)}}={\mathrm{e}}^{\mathsf{\alpha }}\times {\mathrm{e}}^{{\mathsf{\beta }}_{11}{\mathrm{x}}_1}\times {\mathrm{e}}^{{\mathsf{\beta }}_{21}{\mathrm{x}}_2}\cdots {\mathrm{e}}^{{\mathsf{\beta }}_{\mathrm{n}1}{\mathrm{x}}_{\mathrm{n}}}\times {\mathrm{e}}^{\mathsf{\mu }}$$

(3)

$$\mathrm{o}\mathrm{d}\mathrm{d}\mathrm{s}={\displaystyle \frac{\mathrm{p}(\mathrm{y}=3)}{\mathrm{p}(\mathrm{y}=2)}}={\mathrm{e}}^{\mathsf{\alpha }}\times {\mathrm{e}}^{{\mathsf{\beta }}_{13}{\mathrm{x}}_1}\times {\mathrm{e}}^{{\mathsf{\beta }}_{23}{\mathrm{x}}_2}\cdots {\mathrm{e}}^{{\mathsf{\beta }}_{\mathrm{n}3}{\mathrm{x}}_{\mathrm{n}}}\times {\mathrm{e}}^{\mathsf{\mu }}$$

(4)

4. Result Analysis and Related Discussions

4.1. Result Analysis

This study used the statistical software SPSS 20.0 to conduct a multinomial logistic regression analysis on the cross-sectional data from 3137 surveyed migrant workers entering cities. During the data processing, all of the independent variables were introduced into the regression equation, and a significance test was performed on the regression coefficients. Finally, an explanatory model for the differentiation of migrant workers’ migration intentions was obtained, which integrated independent variables from four dimensions. Due to limited space, this chapter focuses on presenting and analyzing significantly influencing characteristic variables. The model estimation results are shown in

Table 4. Results of the multinomial logistic regression on the differentiated migration intentions of rural-to-urban migrant workers (N = 3137).

Description of Items		Migration Intention = 1 Stay in the City			Migration Intention = 3 Return to the Hometown
		Coefficient	Standard Error	Odds Ratio (OR)	Coefficient	Standard Error	Odds Ratio (OR)
Age (30–39 years old = reference group)	Under 20	−0.752 **	0.437	0.472	−0.510	0.505	0.600
	20–29	−0.041	0.234	0.960	−0.433 *	0.270	0.649
	40–49	0.594 **	0.271	1.812	1.033 ***	0.285	2.809
	50–59	0.493 *	0.356	1.637	1.303 ***	0.362	3.681
	60 and above	0.169	0.597	1.184	1.366 **	0.558	3.919
Education Level (Junior High School = reference group)	Primary School and Below	−0.547 **	0.268	0.579	−0.137	0.267	0.872
	High School	−0.036	0.200	0.965	−0.077	0.222	0.926
	Junior College	−0.019	0.257	0.981	−0.276	0.317	0.759
	Bachelor’s Degree and Above	0.266	0.275	1.305	−0.783 **	0.391	0.457
Marital Status (Married = reference group)	Unmarried	−0.334	0.232	0.716	−0.508 **	0.272	0.602
	Others (divorced, widowed)	−0.033	0.464	0.967	−0.156	0.480	0.856
Family Members Left Behind (No one left behind = reference group)	Family Members Left Behind	0.161	0.166	1.174	0.359 **	0.186	1.432
Family Members Migrating Together (Alone = reference group)	With Family Members	0.474 ***	0.192	1.607	−0.160	0.212	0.852
Family Ownership of Properties (Only hometown properties = reference group)	Only local properties	1.141 ***	0.480	3.129	−1.836 ***	0.688	0.159
	Properties in both places	0.282	0.311	1.326	−0.557	0.398	0.573
	No properties in both places	−0.214	0.196	0.808	−0.958 ***	0.255	0.384
Annual Household Income Last Year (CNY 60,000–89,999 = reference group)	Below CNY 30,000	0.182	0.264	1.200	0.068	0.290	1.070
	CNY 30,000–59,999	0.033	0.224	1.033	0.027	0.250	1.028
	CNY 90,000–119,999	0.440 **	0.263	1.553	−0.192	0.308	0.825
	CNY 120,000 and above	0.303 *	0.264	1.354	−0.284	0.320	0.753
Duration of Migrant Work (1–3 years = reference group)	4–6 years	0.046	0.223	1.047	0.143	0.278	1.154
	7–9 years	0.115	0.248	1.121	0.518 *	0.287	1.679
	10–12 years	−0.138	0.293	0.871	0.626 **	0.322	1.870
	13 years and above	−0.033	0.262	0.968	0.176	0.288	1.193
Current Work Distance (Local city = reference group)	Other cities within the province	−0.304	0.254	0.738	0.021	0.318	1.021
	Other provinces	−0.409 *	0.267	0.664	0.175	0.327	1.191
Current Employment Status (Employee = reference group)	Employer	0.604 *	0.424	1.830	0.360	0.495	1.433
	Self-employed Worker	0.512 ***	0.195	1.669	0.503 **	0.214	1.654
	Family Helper	−0.247	0.430	0.781	−0.579	0.525	0.560
	Others	−0.271	0.344	0.762	−0.586	0.437	0.556
Monthly Income of Current Job (CNY 4000–5999 = reference group)	Below CNY 2000	−1.044 ***	0.301	0.352	−0.337	0.334	0.714
	CNY 2000–3999	−0.235	0.204	0.790	0.285	0.235	1.330
	CNY 6000–7999	−0.087	0.297	0.917	−0.360	0.368	0.698
	CNY 8000 and above	−0.591 **	0.300	0.554	−0.251	0.357	0.778
Migrant Work City Level (County-level city = reference group)	Prefecture-level city	0.361 *	0.217	1.435	0.009	0.248	1.009
	Sub-provincial city and above	0.564 ***	0.215	1.758	0.766 ***	0.239	2.150
Identity Recognition (From hometown = reference group)	Local resident	0.235	0.274	1.265	−0.440	0.335	0.644
	Both local resident and from hometown	0.369 **	0.178	1.446	−0.608 ***	0.220	0.544
	Neither local resident nor from hometown	−0.516 *	0.278	0.597	−1.670 ***	0.405	0.188
Main Contacts (Fellow townsmen = reference group)	Local people	0.402 **	0.182	1.494	−0.006	0.204	0.994
	Other out-of-towners	0.016	0.197	1.016	−0.154	0.215	0.858
Number of Regular Friends (0–3 = reference group)	4–7	0.232	0.185	1.261	−0.311 *	0.204	0.733
	8–11	0.461 **	0.214	1.586	0.162	0.237	1.176
	12 and above	0.564 ***	0.234	1.758	−0.155	0.273	0.857
Number of Days Returning Home Each Year (0–9 days = reference group)	10–19 days	0.200	0.195	1.221	0.365	0.227	1.440
	20–29 days	0.124	0.251	1.132	0.437	0.290	1.548
	30 days and above	0.202	0.212	1.224	0.781 ***	0.239	2.184
	Constant Term	−0.129	0.554		−0.040	0.629

Note: The benchmark category for comparison is migration intention = 2: blind orientation; *, **, *** indicate that the parameters are significant at the 0.1, 0.05, and 0.01 levels, respectively.

. From the perspective of model fitness, a likelihood ratio test was performed on the model, and the overall chi-square of the multinomial logistic regression model was 538.81, with a significant level p-value of 0.000, passing the test at the 1% level, indicating that the model is significantly valid overall, and the model fit is good. The Nagelkerke R² value of the model is 0.346, indicating that the regression model is very effective.

4.2. Conclusions

4.2.1. Clear Direction

(1) 

Intention to Retain in the Cities

The availability of more comprehensive public services in cities is the main reason why most migrant workers choose to stay in cities. As the years of working in cities accumulate, bringing the whole family to cities has gradually become the primary choice for migrant workers who intend to stay. From the perspective of destination cities, with the single migrant workers as the reference group, the result shows the opposite effect of the analysis of the source place. In cities, there is a significant positive correlation, with an OR value of 1.607, indicating that migrant workers who work in cities with their family members are 1.6 times more likely to have the willingness to stay than the reference group. This objectively demonstrates the strong effect of migrating with family members on the willingness of migrant workers to stay in cities. In interviews, many migrant workers worked with their families. One case is a husband and father in the family who thought more about the future development direction from a family perspective:

[Interview Case 1]

“There are no relatives left in my hometown in the countryside. Now my wife and my children are all living with me in the city. Although my life pressure has increased, I am willing to bear it all for the sake of our family being together. No matter how hard and tiring it is, I will do it gladly. I have to be responsible for them. I need to earn more money and let them enjoy a better life in the city.”

(Qiao XX, male, 38 years old, taxi driver, Shijiazhuang City, 29 September 2017)

From the perspective of the supporting role of annual household income, choosing to stay in the city as a migrant worker means having a higher income level to support the basic livelihood of oneself and accompanying family members, compared to returning to the countryside. Using the annual income range of CNY 60,000–89,999, which is common among migrant worker families, as a reference group, the results show that the impact of each category is not significant in terms of returning to their hometowns. However, the two relatively higher income categories have significant effects on staying in the city. Among them, the relatively higher category with an annual household income of CNY 90,000–119,999 has an OR value of 1.553, indicating that the likelihood of having the willingness to stay in the city is 1.5 times that of the reference group. The even higher category with a household income of CNY 120,000 and above has an OR value of 1.354, which is relatively lower but still shows a positive effect on staying in the city compared to the reference group. These results suggest that relatively affluent family conditions enable migrant workers to better afford city expenses, and families are more likely to support the migration of migrant workers to cities. In many interview cases, the author found that family concerns far outweigh personal concerns, and future development plans are often based on family economic situations. Some cases expressed amazement at the family’s ability to support work in the city:

[Interview Case 2]

“My family’s conditions are quite good, mainly because we have a lot of land, including a 20-mu large field and a 40-mu large hazelnut orchard. Our annual income is also over 100,000 yuan. I could have chosen not to come out, but my parents think that staying in the countryside has no future, so they asked me to come to the city to gain experience and broaden my horizons. Besides, most young people in the countryside are now working in the city, and there are few young people in the village. I would only see some old people there, so there’s no fun staying at home. To let me work in Dalian without worries, my parents even bought a 90-square-meter apartment for me in Ganjingzi District. My dad said that when they can’t work anymore, they will come here to live with me.”

(Mao XX, male, 26 years old, real estate salesperson, Dalian City, 13 August 2017)

Housing security has the strongest positive impact on the migration of migrant workers (Zhou You, 2023) [30]. Under the influence of traditional Chinese farming culture, owning a house is considered fundamental to settling down and starting a career. Therefore, this article analyzes the differentiating effect of owning a house on the migration intentions of migrant workers in cities. Using the category of only having a house in the hometown as a reference group, two categories significantly affect the migration intention of migrant workers. Among them, the category of only having a local house has a significant impact on both types of intentions, but the effects are positive and negative. The OR values are 3.129 and 0.159, respectively, indicating that compared to the blind intention, the possibility of this category generating a willingness to stay in the city is 3.1 times that of the reference group, and the possibility of generating a willingness to return is only 15.9% of the reference group. Combining the analysis of the two results, it is evident that owning a house in the city of employment strongly supports the effect of migrant workers staying in the city. In interviews, the author deeply felt the migrant workers’ desire for housing. Housing is not just a form of wealth for them, but more of a spiritual sustenance. A case that has purchased a house in Xiangyang said:

[Interview Case 3]

“I have been working in Xiangyang for over 20 years since I left my rural hometown in 1998. I have always rented other people’s houses, and not only is the rent expensive, but I also feel like an outsider and inferior. I feel like the rented house is not my own home. Now that I finally have my own house, I truly feel like my home is really settled in Xiangyang. Although my household registration is still in my rural hometown, I feel like a Xiangyang person, a city person. Having my own house feels really different, and I feel very at ease, comfortable, and happy.”

(Liang XX, male, 46 years old, property manager, Xiangyang City, 11 October 2018)

From the perspective of employment, the social status of migrant workers has a significant impact. In 2022, the Chinese government issued the “Opinions on Supporting the Employment and Entrepreneurship of Migrant Workers”, which clearly supports the employment and entrepreneurship of migrant workers. Compared with the employed group, self-employed migrant workers show a stronger willingness to migrate, and their willingness to migrate with their families is also significantly higher than that of the employed group. Using the most common employees as a reference group, the results show a relatively scattered effect. There is a significant positive correlation between the employer category and staying in the city, with an OR value of 1.830, indicating that the likelihood of this category having a willingness to stay in the city is 1.8 times that of the reference group, showing a strong desire for migrant workers who achieve employer status to settle in the city. Another significant relationship exists for self-employed workers, who show a significant positive correlation in both staying in the city and returning home. This suggests that compared to the blind willingness benchmark, self-employed workers may work harder to operate their businesses in the city to achieve urban migration, with a likelihood 1.6 times that of the reference group, or they may return home to realize their own value creation, with the same likelihood of 1.6 times that of the reference group. Overall, this demonstrates the relatively clear migration intention of self-employed migrant workers. In one interview case, a fruit retail store owner had a clear plan for her future career and life, as she stated the following:

[Interview Case 4]

“I am from a rural area in Zhangjiakou, Hebei Province. My sister was originally engaged in fruit wholesale in Taiyuan. When I was 19 years old and had just graduated from high school, I went straight to my sister’s place to help her sell fruit. After working for more than 6 years, I saved over 100,000 yuan. Then I rented a house near the Financial University in Xiaodian District and opened a fruit retail store, getting my goods directly from my sister at her cost price without earning any profit from me. Now my husband and I work for ourselves, and we feel good about it. We earn a good income from this business. College students have money, so the sales volume of fruits is quite large, and we feel comfortable doing it ourselves. This business can not only support my family but also allow us to save 70,000 to 80,000 yuan every year. After saving for another two years, I plan to buy an apartment in Taiyuan. I want to buy a bigger one, around 90 square meters.”

(Ye XX, female, 35 years old, fruit retailer, Taiyuan City, 27 September 2017)

The level of psychological integration is a measure of the interaction between migrant workers and the city they live in at the emotional level, and it is a key indicator to measure the degree of urbanization of migrant workers. Whether migrant workers are truly integrated into the city psychologically depends on whether they are integrated into the local social circle. In terms of interaction objects, usual interaction objects have a certain influence. Using fellow villagers as a reference group, migrant workers whose main interaction objects are locals have a particularly significant positive relationship in their willingness to stay in the city, with an OR value of 1.494. This indicates that the likelihood of this category having a willingness to stay in the city is 1.4 times that of the reference group, demonstrating the significant influence of local main interaction objects on staying in the city. In fact, the interaction between migrant workers and city residents can reflect the need for a social circle among migrant workers who settle in the city. Many interview cases mentioned that their social circle is their emotional habitat during their work. One case stated the following:

[Interview Case 5]

“I came to Zhuhai to join my children. Both my daughter and son came to Zhuhai for work after graduating from university. Because her uncle’s family runs an electronic communication equipment manufacturing company here, which is quite large, they came here. Not only them, but almost all of my relatives on my father-in-law’s side are here. Those who are not here have gone to Siping or Changchun. Only old people are left in the countryside. Northeast China is so far from Zhuhai, and it’s difficult for the children to see us, so we just moved to Zhuhai. Because all our relatives and friends basically came here, we don’t feel lonely at all. In fact, we feel lonely and homesick when we go back to our rural hometown.”

(Miao X, male, 53 years old, property security guard, Zhuhai City, 16 June 2018)

In summary, based on considerations of their own and family factors, some migrant workers choose to stay in the city under the comprehensive influence of city level, public services, and employment policies. The choice of migrant workers to stay in the city indicates that the integrable factors of the city where they work can somehow cover the obstacles to staying in the city, enabling them to overcome the obstacles and continue to stay in the working city.

(2) 

“Return Flow” intention

Firstly, the migration intention of migrant workers returning to their hometown has changed from “passive return” to “active return”. With the deepening of reform, China’s urbanization rate has been continuously improving, and the urban–rural relationship has gradually shifted from differentiation to integration. As migrant workers transition from the “older generation” to the “new generation”, their migration concepts for the future are also changing. In 2021, the state issued the “Notice on Promoting the Pilot Experience of Supporting Migrant Workers and Other Personnel to Return Home to Start Businesses”, actively promoting the employment of migrant workers returning home; in 2022, the “Opinions on the Current Employment and Entrepreneurship of Migrant Workers” was issued, clearly supporting the employment and entrepreneurship of migrant workers. In recent years, the implementation of the rural revitalization strategy has changed the overall appearance of rural areas and increased development opportunities in rural areas. Therefore, some migrant workers are willing to return from cities to their hometown villages or small towns for further development. According to the “2023 Migrant Workers Monitoring Survey Report” issued by the National Bureau of Statistics of China (

Table 5. Distribution of migrant workers’ export and import areas.

Region	2022	2023	Increment	Growth Rate
By origin
Eastern region	10,403	10,484	81	0.8
Central region	9852	9904	52	0.5
Western region	8351	8367	16	0.2
Northeast region	956	998	42	4.4
By destination:
In the eastern region	15,447	15,277	−170	−1.1
In the central region	6771	6982	211	3.1
In the western region	6436	6552	116	1.8
In the northeast region	843	872	29	3.4
In other regions	65	70	5	7.7

), compared with 2022, migrant workers continued to return to the central and western regions.

However, some migrant workers choose to return to their hometowns based on passive backflow under the multi-dimensional restrictions of cities. They generally do not have the ability and capital to permanently settle in cities, and usually rely on staged work income to ensure the temporary survival of individuals and families, building a weak migration willingness in the urban–rural circular flow. As this group’s working years in cities increase, due to their relatively high age, these migrant workers are gradually excluded to the edge of urban life, which makes it difficult for them to fully enjoy the convenience of the city and lack a sense of belonging in urban life. Finally, this group will choose to return to their hometown after reaching a certain age. The core connotation is mainly reflected in the working duration. Taking going out for 1 to 3 years as a reference group, the category of migrant workers who go out for 4–6 years does not have a significant impact on the results, while the 7–9 year category shows a significant positive correlation in the backflow, and its OR value is 1.679, indicating that the possibility of this category having a backflow willingness is 1.6 times that of the reference group, which means that migrant workers who have been away for 7 years relatively start to have the idea of returning. This turning point is also presented in the 10–12-year category, which shows a positive correlation in the willingness to return and is more significant, and its OR value has also increased to 1.870, indicating that the possibility of this category generating a willingness to return is 1.8 times that of the reference group. It can be seen that with the increase in time away, if they do not settle down well, the idea of returning of migrant workers will become stronger. In the author’s in-depth interviews with migrant workers, many of them have been away for about 10 years, and their ideas of returning are very clear. One interview case said the following:

[Interview Case 6]

“I came to Wuxi when I was 18 years old. I have been wandering around doing this and that, always doing tiring and unprofitable work. I feel that the pressure of living in the city is too great, and I don’t feel like this is my home. If we talk about the future, I have thought about it. If I have a little savings and earn some money, I want to go back to my hometown. For example, I can do something I like, such as opening a flower shop, or a milk tea shop, or a coffee shop. I don’t need to earn a lot of money every day. It’s enough to make a living and live a plain life.”

(Liu X, female, 29 years old, hotel waiter, Chengdu City, 21 November 2017)

In addition, the restrictive elements faced by returning migrant workers mainly come from the “secondary restrictions” encountered in the city, which mainly focus on various thresholds in career choices, social interaction, and public cultural participation. They generally do not have the permanent ability and capital to settle in the city, and usually rely on phased work income to ensure the temporary survival of individuals and families, building a weak migration willingness in the urban–rural circular migration.

4.2.2. Blind Direction

The total sample size of blind migrant workers is 755, and their migration intentions include two options: “going to see other places” (216 people, accounting for 27.9%) and “not sure” (559 people, accounting for 72.1%). Blind migrant workers exhibit prominent characteristics in various basic aspects, especially some concentrated trends that can objectively reveal the group attributes of blind migrant workers and the phased contradictions in their migration intentions during the process of working, mainly manifested as follows:

(1) 

Low cultural and skill levels

In terms of the differentiation effects of education level, taking the most concentrated junior high school education level as the reference group, there are two significant results on the whole: the category of primary school and below shows a significant negative correlation in staying in cities, with an OR value of 0.579, indicating that this category of migrant workers has a 57.9% probability of having the intention to stay in cities compared with the reference group, i.e., they significantly demonstrate a kind of blindness in the process of working. Among them, 11.8% have received primary school education, and these migrant workers have a narrow range of career choices and are engaged in relatively low-end jobs; 32.3% have received junior high school education, and their career choices are also adversely affected by their lower education level and limited by the thresholds of higher education qualifications; 30.5% have received high school education, and they can cross the entry thresholds of many employers with their high school education qualifications, but they still struggle to expand their range of career choices beyond low-end jobs; 25.3% have received college education or above, and they have the most advantageous cultural background and educational foundation among the migrant worker population. Compared with migrant workers with a high school education or below, they have more job-seeking channels, fewer barriers to entry, and a higher probability of success, making them the most skilled and advanced group among blindly migrating workers. In general, the cultural education structure of blindly migrating workers is still dominated by junior and senior high school education levels, with 62.8% of migrant workers having a junior or senior high school education and 74.6% having an education level below senior high school, indicating that the overall level of cultural education among blindly migrating workers is still relatively low. This is an important factor contributing to numerous obstacles in job-seeking, narrow job-seeking spaces, low-end occupations, and low incomes among migrant workers. In terms of the receipt of professional skills and technical training among blindly migrating workers, 151 people, accounting for 19.5%, have received professional and vocational skills training, while the remaining 80.5% of blindly migrating workers have not participated in any professional or vocational skills training. This indicates that only one-quarter of blindly migrating workers have received vocational skills training and possess certain professional and technical skills. In fact, professional and vocational skills, as important components of human capital, relentlessly divide the labor market and have a direct impact on the career choices and work experiences of migrant workers. Employers in cities generally have minimum requirements for the education level of migrant workers, and most enterprises with technical content explicitly stipulate that migrant workers must have undergone professional training and possess technical qualifications when recruiting them. Low cultural education levels and a lack of vocational skills training have become two weaknesses in the job-seeking and development of blindly migrating workers, which are critical adverse factors constraining their career advancement and urban integration.

(2) 

Weak professional accumulation

Employment status with social attributes has a significant impact. Taking the most common employees as the reference group, the results show a relatively dispersed effect. The employer category has a significant positive correlation in staying in cities, with an OR value of 1.830, indicating that this category has a 1.8 times higher likelihood of having the intention to stay in cities compared with the reference group, demonstrating a strong desire for migrant workers who have achieved the status of employer to stabilize their lives in cities. Another significant relationship exists for self-employed workers, who show significant positive correlations in both staying in cities and returning to their hometowns. This indicates that based on the blind intention as a comparative benchmark, self-employed workers may work harder to operate in cities to achieve urban migration, with a likelihood 1.6 times that of the reference group, or return to their hometowns to realize their own value creation, also with a likelihood 1.6 times that of the reference group. Overall, this demonstrates a relatively clear migration intention among self-employed migrant workers. Additionally, although the results for family helpers and the other two categories are not significant, they are all negatively correlated. This may be because these two less defined occupational identities are often in more passive and marginalized situations, making it difficult to achieve stable career development and obtain the initiative to settle in cities, thereby easily leading to a sense of confusion about the future.

The monthly income from work, an economic attribute, has two significant levels of influence. Taking the moderately high CNY 4000–5999 monthly income category as the reference group, the results show that the two categories of extremely low (below CNY 2000) and extremely high (CNY 8000 and above) monthly incomes have significant negative effects on staying in cities. This indicates that migrant workers in these two income categories have relatively weak intentions to stay in cities and are more prone to a sense of blindness, with the likelihood of having the intention to stay being 35% and 55% of the reference group, respectively. The extremely low-income category may be due to a lack of viable options, while the extremely high-income category may be seeking an even more ideal development space. These results reflect that while migrant workers in cities have income, there are significant internal differences within this income, and there is also a gap between them and urban residents. Overall, the pure economic aspect of monthly income from work for migrant workers in cities is difficult to pinpoint as a definitive factor in future migration choices but plays a significant role in the current state of confusion and stagnation. Migrant workers not only compare themselves with their own small circles but also more frequently with the general income levels in the cities they reside in.

[Interview Case 7]

“I came out right after graduating from high school in 2011. At that time, I just wanted to come out to the big city and see the world. I have delivered parcels in Beijing, sold baozi in a baozi shop in Tianjin, and worked in a barbecue restaurant in Shenyang. I am not afraid of hardship or tiredness. Although my income is always a bit low and not enough to spend, I can finally support myself. But the problem is, I can’t see a bright future with this kind of work. My current job pays only about 3000 yuan per month, and the rent is 900 yuan, not to mention the cost of food and drink. There’s not much left every month. I am very confused now. There is no place to earn money in the countryside, and all the money earned in the city is spent. I don’t know where I should go in the future?”

(Wang XX, male, 26 years old, hotel security guard, Zhuhai City, 16 June 2018)

In addition, Figure 2 indicates that families with lower incomes are more likely to have a “blind” willingness to migrate.

(3) 

Dispersed social relations

Blind migrant workers often exhibit staged and dispersed social interaction effects during their relatively short work history, which places a certain degree of restraint on the emergence of their blind willingness. From the perspective of the daily interaction objects of blind migrant workers (

Table 6. Social relationships of blind migrant workers.

Description Variable	Category	Blind Direction	Clear Direction	Chi-Square Test
Main Communication Partners	Fellow townsman	27.2%	29.1%	p = 0.000
	Local people	41.0%	47.7%
	Other outsiders	31.8%	23.2%
Number of Frequent Friends	0–3	33.5%	29.3%	p = 0.031
	4–7	32.6%	32.2%
	8–11	20.3%	22.2%
	12 and above	13.6%	16.3%

), locals account for 41.0%, which is 6.7 percentage points lower than non-blind migrant workers, indicating that the interaction between blind migrant workers and locals is relatively weak; their countrymen account for 27.2%, which is 1.9 percentage points lower than non-blind migrant workers, indicating that the rural ties of blind migrant workers are not close; other foreigners account for 31.8%, which is 8.6 percentage points higher than the non-blind group, indicating that the more important interaction objects of blind migrant workers are other foreigners. These people play an important role in restraining the formation of blind willingness among migrant workers. From the perspective of the number of frequently contacted friends, blind migrant workers are concentrated in the range of 0–3, accounting for 33.5%, which is 4.2 percentage points higher than non-blind migrant workers, while the categories with eight or more friends are lower than non-blind migrant workers, indicating that blind migrant workers have difficulty establishing a wide range of friendships during their work history. Overall, dispersion, staging, and “fragmentation” constitute the basic characteristics of the social relationships of blind migrant workers, revealing that the urban social relationships of blind migrant workers are in a relatively weak state.

In terms of identity, migrant workers who hold the cognition of “neither a local nor a native” are also significant in both categories, but the results are opposite in staying in the city. The willingness to stay in the city of “double non-identification” is 59% of the reference group, while the willingness to return is 18% of the reference group, clearly showing the strong blindness of “double non-identification”. They cannot make a clear choice between staying or leaving based on blind identification. The author has a typical case of “double non-identification” in an interview, who describes this state as follows:

[Interview Case 8]

“Although I have lived in Shanghai for 10 years, I still feel that we are not Shanghainese. In my mind, Shanghainese should have a household registration, a house, and a happy life, but I have none of them. I don’t have that sense of belonging to Shanghai because I feel like I’m in a rented state and always feel like a passerby in Shanghai. My hometown, Peizhou, is even more unfamiliar to me, and I’m not used to rural life, so it’s impossible for me to return to Peizhou. I’m not happy in Shanghai, and I’m even less happy when I go back home. I don’t know where I can be happier?”

(Zhao XX, female, 31 years old, company clerk, Shanghai, 25 June 2018)

As shown in Figure 3, migrant workers have a relatively low degree of recognition of the life and behavior of residents in the cities where they work.

In general, the endowment characteristics and work accumulation of migrant workers with unclear directions can be roughly summarized into three types of elements, namely, integration elements, restrictive elements, and blinding elements. The integration elements mainly include relatively enduring work capabilities, relatively satisfactory work income, slight improvements in career transitions, a certain degree of cognition of urban society, and potential plasticity for future development. The integration elements are the most basic and positive work guarantees that migrant workers with unclear directions acquire based on group endowment and work accumulation. They are the source of motivation for migrant workers with unclear directions to stably and continuously work and live in cities, and thus have become an important guiding force for migrant workers with unclear directions to integrate into cities. Restrictive elements mainly include low levels of both culture and skills, a weak family economic foundation, fluctuations in income history, constraints on migration distance, and marginal social status. The reason why migrant workers with unclear directions fail to form clear migration goals lies in the direct effects of restrictive elements encountered in their work experience, which restrict and hinder their inner desire to take root in cities. The mixed effects derived from the collision, interweaving, coupling, and game of the “integration and restriction” elements lead to the stagnation of migrant workers who “cannot stay in the city or return to their hometown”, coupled with the infiltration, intervention, and catalysis of blinding elements, making migrant workers develop a willingness to migrate with unclear directions of “not knowing where to go”.

5. Improvement Strategies

5.1. Increase the Efforts to Address the Issues of Migrant Workers with Unclear Directions and Enhance Their Employment Capabilities and Living Standards

Currently, the scale of migrant workers with unclear directions entering cities in China is large, and there is a clear trend in continued expansion. It is of great practical significance to do a good job in the employment services and living assistance for this group: First, it is conducive to the harmony and stability of migrant workers and their families. The income of migrant workers with unclear directions is an important source of income for their families, and a stable and continuous work status is conducive to the harmonious stability and steady development of migrant workers and their families. Second, it can provide a certain scale of labor resources for urban construction. Although migrant workers with unclear directions staying in cities for a long time will derive a series of social issues like a “double-edged sword”, they also provide a certain amount of human resources for urban construction and development, which can solve the problem of insufficient labor in low-end industries in a certain range Third, it is conducive to promoting the urban integration of migrant workers. Ensuring the basic rights and interests of migrant workers with unclear directions as permanent urban residents, promoting their urban integration and social integration, is an important responsibility that urban governments and society cannot shirk. Therefore, it is necessary to establish and improve the governance system and mechanism for migrant workers with unclear directions entering cities, strengthen and improve policies to support stable employment and jobs for migrant workers with unclear directions, improve and provide basic social security and public services for them, strengthen vocational training, enhance employment levels, standardize labor management, improve social security, issue unemployment insurance, safeguard legitimate rights and interests, improve living environments, enrich cultural life, develop social resources, expand development space, and other series of service management to ensure that migrant workers with unclear directions can achieve stable employment and stable lives, promote their sense of urban belonging, happiness, and integration, and promote their urban identity and urban integration.

5.2. Promoting the Transformation of Migrant Workers’ Blind Group into Urban Residents and Advancing the Urbanization and Citizenization Process of the Country

The long-term stay of migrant workers’ blind group in cities can lead to extremely adverse unexpected consequences. Therefore, proactive measures should be taken to facilitate the transformation of migrant workers’ blind group into urban residents and returning migrant workers. Especially in the context of the country vigorously promoting the urban settlement of the non-registered urban population, migrant workers have actually become the key group for citizenization. Their migration direction has a global and decisive impact on the new urbanization and citizenization process. Therefore, promoting the transformation of migrant workers’ blind group into migrant workers staying in cities has more prominent practical significance. We should formulate and implement more comprehensive and forceful policies and measures to promote their citizenization, aiming at the main problems that hinder their transformation into urban residents, such as low professional quality, weak employment ability, lack of social resources, shallow urban integration, and unclear development direction. Through improving the mechanism of the employment service system for migrant workers, enhancing their professional skills and abilities, improving their professional levels and employment quality, implementing and improving social insurance subsidy policies, stabilizing and increasing their working income, protecting their labor and social rights and interests, and enhancing their social integration in cities, we can promote migrant workers to fully and equally enjoy basic urban social security and public services, and facilitate the better, faster, and more comprehensive transformation of migrant workers’ blind group into urban residents. Specifically, it is necessary to ensure that migrant workers achieve the “Ten-have” goal based on stable employment, namely “jobs in the city, training before work, labor contracts, guaranteed compensation, access to social insurance, education for children, improved housing, channels for rights protection, cultural life, and clear development goals” (Yang zhiming 2013) [31] thus comprehensively realizing the citizenization and equalization of migrant workers’ blind group in terms of social security, public services, and rights protection, continuously enhancing their sense of belonging, acquisition, and integration into the city, guiding and promoting the rapid transformation of migrant workers’ blind group into migrant workers staying in cities, and laying a solid population foundation for ultimately achieving the goals of new urbanization and citizenization.

5.3. Promoting the Transformation of Migrant Workers’ Blind Group into Returning Migrant Workers and Accelerating Rural Revitalization and Modernization

Migrant workers are an important talent reserve for new rural construction and rural revitalization. Their large-scale return and employment or entrepreneurship in rural areas have a strong driving effect on strengthening the construction of high-quality rural talent teams, cultivating and creating a new type of professional farmers, and promoting rural revitalization and modernization. Therefore, we should focus on promoting the transformation of migrant workers’ blind group into returning migrant workers and providing policy support and institutional guarantees for their return and employment or entrepreneurship. We should combine the general requirements of new rural construction and the implementation of the rural revitalization strategy, stimulate the enthusiasm of migrant workers’ blind group for returning and starting businesses by strengthening policy support and optimizing the entrepreneurial and employment environment, promote their active return and employment or entrepreneurship, and accelerate the development process of agricultural and rural modernization. Firstly, we should strengthen the policy creation and institutional support for migrant workers’ returning entrepreneurship, comprehensively improve their service functions for returning entrepreneurship by establishing an entrepreneurship training system, optimizing the entrepreneurial environment, reducing and eliminating entrepreneurial barriers, building entrepreneurial incubation training bases, cultivating entrepreneurial development platforms, increasing entrepreneurial subsidies and financial services, and strictly implementing tax reduction and fee reduction policies. We should vigorously improve the service efficiency of migrant workers’ returning entrepreneurship, promote their active return and entrepreneurship, and continuously improve their entrepreneurial enthusiasm, stability, and success rate. Secondly, we should broaden the channels and fields of migrant workers’ returning employment and comprehensively enhance the carrying capacity and absorption capacity of counties and townships for returning migrant workers’ employment. Through optimizing the employment environment of counties, townships (towns), and villages, strengthening employment skills training, promoting and improving the combination of production and training, improving employment market intermediary services, improving rural social security and public services, implementing rural employment social insurance subsidies, developing rural characteristic industries and modern industrial systems, implementing rural employment assistance systems, and continuously promoting rural construction actions, we should establish and improve the service system and mechanism for migrant workers’ returning employment, comprehensively enhance the quality and level of their employment services, effectively enhance the ability of rural industries to attract and accommodate migrant workers’ employment and promote new achievements in migrant workers’ returning employment. Thirdly, we should encourage and promote entrepreneurship to drive employment. According to a survey conducted by the Ministry of Human Resources and Social Security, every migrant worker returning home to start a business can create about four new jobs. Therefore, we should give full play to the linkage effect, chain effect, and multiplier effect of migrant workers’ entrepreneurship to drive employment. By implementing policies, such as subsidies for entrepreneurship and employment, tax incentives and reductions, and small-scale guaranteed loans to support migrant workers’ entrepreneurship, we should expand financing channels for migrant workers’ entrepreneurship, optimize the entrepreneurial environment for returning migrant workers, implement service guarantees for migrant workers’ entrepreneurship, build a new development model of migrant workers’ entrepreneurship to drive employment, absorb and drive more agricultural transferred labor force to achieve local and nearby employment and promote the integrated development and comprehensive revitalization of the rural primary, secondary and tertiary industries.

6. Research Limitations and Future Prospects

There are certain limitations to the data acquisition. The core issue of this study focuses more on a logical construction within the migration thinking of migrant workers rather than a “citizenization” calculation of migrant workers from a governmental or policy perspective. This individual-centered logical research paradigm may be prone to distortion in higher-level interpretations. Simultaneously, the survey subjects in this study’s data acquisition are also individual migrant workers. Due to differences in cultural knowledge levels and comprehension abilities, it is difficult to control data and information loss caused by the subjectivity of migrant workers during the collection process. Especially, family-level information is indeed difficult to fully acquire through personal experience narratives, resulting in a certain degree of deviation from the objective truth in the survey results.

There are limitations to measurements based primarily on logical construction. This study is an interpretation of the migration willingness thinking of migrant workers entering cities. The constructed “integration limit” perspective also focuses on the logical influence effects of elemental mid-level dimension indicators. Therefore, the systematicness and standardization of the characteristic variables designed in this study need to be improved, and the precision of the research on the migration willingness of migrant workers needs to be enhanced. Additionally, some migrant workers may hide certain family information to protect their privacy. Hence, many specific measurements can only be simplified and basically measured, making it difficult to reveal some hidden effects deep in the hearts of migrant workers. Future research should explore the multi-subject and multi-dimensional development of migrant workers’ migration willingness, adding various survey units with migrant workers as the core, and further improving field intervention methods and data collection techniques.

In summary, future research should first focus on deep design and refined improvement based on the characteristics of migrant workers. Secondly, when exploring the multi-subject and multi-dimensional development of migrant workers’ migration willingness, various survey units centered on migrant workers should be added, and field intervention methods and data collection techniques need to be further improved. As the flow locations of migrant workers become more nuanced, they will inevitably have clearer and more diverse directions for future migration. For example, this study also mentions that staying in the city can be differentiated into “staying in this city” and “staying in another city”, while returning can also be differentiated into more nuanced migration intentions such as “returning to the countryside” and “returning to the township”. Therefore, exploring the “ternary differentiation” is not sufficient to support a deep interpretation of migrant workers’ migration willingness research in the long run. Future research should conduct refined studies on migrant workers’ migration willingness based on specific locations and build a broader “multiple differentiation” framework. However, the theoretical framework established in this study aims to explore the influencing factors of migrant workers’ migration willingness and analyze the formation mechanism of their final migration willingness. It lacks theoretical support and in-depth analysis of the impact of city clusters and city connectivity on migrant workers’ migration willingness from a meso-level perspective. Therefore, future research should incorporate the role of city connectivity into the scope of influencing factors and shift the perspective of migrant workers’ migration from the individual research level to the macro level. That is, it should explore the differentiating effects of city agglomeration and city connectivity research on migrant workers’ migration willingness.