The Mechanism for Matching the Supply Content and Policy Instruments of Resistive Public Policy

Abstract

In the continuous development of behaviorist public policy theory, nudge tools have gradually become a policy tool of great interest to the academic community, which believes these tools can improve the effectiveness of public policy. However, scholars frequently disregard the impact of policy attributes on the policy effect when the policy content is matched with explicit policy tools and potential policy tools. Most studies have confirmed the application effect of nudges in policies that are receptive, but there is little research on their application in policies that are resistive. Through research on the supply content and promotion mechanism of policy resistance, we determined that the nudge policy tool does not demonstrate significant benefits for the effect of policy behavior, while, among the explicit policy tools, the preaching tool has a significant impact on the effect of policy behavior.

1. Introduction

Policy instruments are the means utilized in the specific action phase under the guidance of policy objectives and in accordance with the classification, structure, and defining characteristics of policy content. Given the existing classification research on policy tools, they can be divided into explicit and potential policy tools.

The term “explicit policy tools” refers primarily to policy tools with obvious intervention means. It is now widely acknowledged in academic circles that Roy Rothwell and Walter Zegveld categorized policy instruments as supply-type, environmental-type, and demand-type [1]. The main measures of supply-based policy tools include direct supply of capital investment, talent introduction, technical support, infrastructure, public services, supporting systems, and other elements. Environmental policy tools primarily refer to the means to influence the implementation of policies via policy publicity and government goal planning. Demand-based policy tools primarily refer to measures requiring the government’s participation, guidance, and coordination. Scholars have conducted most of their research on healthy Chinese policy, hospice policy, and other content from supply-type, environment-type, and demand-type policy tools. The researchers emphasized that the effect of explicit policy tools has a dominant impact when matched with the policy content. Tian Xiujie conducted a quantitative evaluation and an efficiency evaluation of the health industry policy in the three northeastern provinces of China. He also evaluated the structure of the health industry’s industrial policy tools and concluded that the health industry should appropriately reduce environmental policy tools, increase the use of demand-based policy tools, and promote the sustainable development of the health industry [2]. After conducting a quantitative analysis of China’s policy from 2013 to 2020 regarding the integration of medical care and maintenance, Chen Lu et al. discovered that the overall effectiveness of the policy has been steadily increasing. The tools related to environmental policy are the most effective and the most frequently utilized, followed by the tools related to supply policy, and demand policy tools are the least effective [3]. Tang Yong performed a structural analysis of the policy tools used for the hospice care policy. He discovered that environmental policy tools occupy the dominant position in this policy. Tang Yong proposed that the use of demand-based and supply-based policy tools should be increased while continuing to give full play to the effectiveness of environmental policy tools in order to further improve the hospice care policy system [4]. With the rise of behavioral public policy theory, a potential policy instrument, i.e., the nudging policy instrument, has gradually entered scholars’ field of vision. Compared to the explicit policy tool with its obvious intervention means, the nudging tool can be summarized as a potential policy tool that employs less obvious economic or administrative means to change the choice structure and direct the behavior of policy objects based on the premise of ensuring individual freedom of choice. Richard H. Thaler and Cass R. Sunstein [5] were awarded the Nobel Prize in Economics in 2008 for their book Nudge: Improving Decisions About Health, Wealth, and Happiness, which proposed a government regulation tool aimed at influencing (rather than controlling) the predictable development of citizens’ behavior [2]. Initially, the public sectors of the United Kingdom and the United States embraced the concept enthusiastically. In the implementation of foreign government policies, nudges have been widely utilized. The British and American governments, in particular, have made great strides in many policy areas, including medicine and health, energy supply, and philanthropy. Nudges have been investigated by scholars in the fields of economics, psychology, law, and sociology. Richard H. Thaler and Cass R. Sunstein believe that the government can encourage the public to make “better choices” by designing default options with the public’s long-term interests in mind. There is no question regarding the importance of pension insurance for employees, but there are still a significant number of employees who do not participate in insurance because of laziness and other factors. The defined benefit pension scheme in the United Kingdom does not require employees to pay anything, but 49% of people still do not participate. In this regard, the problem can be easily remedied by changing the default setting from “no insurance” to “insurance” [6]. Maya Shanka discovered that the three-step process of informing people to join the “thrift savings plan” substantially increased the number of registrants, which is an example of promoting social welfare by simplifying service processes and limiting service content [7]. Jolien Vandenbroele discussed the efficacy of two specific measures designed to increase meat substitute sales. The experiment determined that product visibility and paired display are the optimal architecture for selecting meat substitutes, which will increase their sales, and validated the role of nudging in promoting the sustainable purchase of meat [8].

Since nudging was introduced to China, scholars in the fields of law and economics have given it considerable consideration. According to domestic scholars, this development appears to have created a third avenue to influence citizens’ behavior and enhance the efficacy of public policy. According to Juhua’s research, the application of nudging, with its gentle methods and the assistance of people’s thinking, can achieve positive results in government purchases [9]. Xiang Fang discovered that the six principles of motivation, trade-offs, default options, prediction errors, feedback, and structural selection played a significant role in encouraging residents of Beijing’s older residential areas to agree to elevator installation [10]. Applying the six nudging principles to the interactive design of knowledge payment applications, according to Liu Shuangshuang, can encourage more users to select knowledge payment applications [11]. Tong Linjie believes that the effective use of nudges can promote the optimization of China’s social moral system based on the psychology of conformity, the call of public figures, self-imagination psychology, and appeal [12]. Shi Lin believes that the consumption aversion theory, herding theory, information burden theory, and heuristic bias theory in the nudging theory can not only adequately explain the causes of violence against medical and health institutions but can also provide viable countermeasures for preventing and reducing violence when combined with the preference reversal theory, the biased view of the current situation, the theory of horizon integration rationality, and the theory of reducing heuristic bias [13].

The policy can be affected by both the tools of explicit policy and those of potential policy. Yet an increasing number of academics are reaffirming the benefits of potential policy tools, including their low decision-making cost, significant policy effect, low policy rebound rate, and compliance with democratic social environments, as well as highlighting their status and role in the policy-making process.

2. Raising Questions

As potential policy tools, the nudging policy tools with the “enlightened paternalism” characteristic are more recognized by academic circles compared with explicit policy tools. Existing studies have generally concluded that nudging policy tools can play a greater role, primarily because the policy content they examine is predominantly one in which the object presents no obvious resistance or opposition, and there is no reversal between its behavior tendency and the nudging direction. When the policy object exhibits obvious resistance to the policy content, can nudging policy tools demonstrate a significant advantage? This is where this article originated. In the face of policy content with which the policy object is in conflict or has obvious objections, how can the nudging policy tools maximize their advantages in order to facilitate the implementation of the policy and the achievement of the policy effect? This is the central issue that this paper seeks to address.

In China, the majority of policies are in line with public desires, but there are a few policies that the government must implement based on public interests and fairness, and the public’s acceptance of these policies is not guaranteed. In the COVID-19 era, countries face numerous social problems, the most significant of which is the employment issue. In 2022, there are 1076 million college graduates of all levels and types, surpassing the million people’s Congress mark for the first time in China. This represents a net increase of 1.67 million compared with 2021. The employment outlook is bleak. In 2016, a number of researchers investigated and analyzed the perspectives of college students on job selection and employment willingness. The survey results revealed that 91.6% of college students preferred to work in large cities such as Beijing, Shanghai, and Guangzhou or in first- and second-tier cities such as those on the eastern coast, while only 2.2% desired to work in rural communities [14]. Against the backdrop of the epidemic, the new first-tier cities (32.40%) accounted for the greatest proportion of new graduates’ employment in 2021, while the proportion of employment in townships and villages remained below 5% [15]. In recent years, approximately 20% of college graduates were employed at the grassroots level, while 2.1% were employed at the grassroots level in rural areas. China’s fundamental social structure is still divided into an urban–rural dual structure. There are significant differences between urban and rural areas in terms of economic development, population quality, and the quality of public services, among others. The majority of graduates do not choose rural grassroots employment based on factors such as pay, working conditions, and advancement opportunities. Even if they choose rural grassroots employment, the majority of graduates use it as a stepping stone rather than attempting to establish themselves at the grassroots level. The remaining portion is eager for employment and chooses it reluctantly because of pressure from society, families, and other factors. It is not difficult to see, in light of the data and current situation research, that the rural grassroots employment policy for college students has characteristics that are in direct opposition to policy objectives. This paper uses the rural grassroots employment policy for Chinese college students as an example to test whether the nudging policy tools still have obvious advantages in matching this policy content and to consider how to achieve the optimal matching effect between policy content and policy tools.

3. Fundamental Concepts and an Analysis of Theoretical Relevance

3.1. Policy Supply Content

Policy supply refers to the allocation of resources by policy makers to address public issues and meet the requirements of policy objects [16]. The content of policies is a vital component of policy supply. Since 2003, the Chinese government has supported four grassroots employment projects, i.e., the “Western Program of College Students’ Voluntary Service”, the “Special Post Teacher Program”, and the “Three Supports and One Assistance Program”, as well as the “College Student Village Official Program”, which began in 2008. Since the 18th National Congress of the Communist Party of China, the central and local governments have provided college students with a greater variety of rural grassroots employment policies. As specified in the Grassroots Growth Plan for College Graduates, a relevant employment guarantee mechanism for college graduates at the grassroots level was established; the “Notice on Further Doing a Good Job in the Employment of Ordinary Higher Education Graduates” encourages college graduates to seek employment at the grassroots level.

Moreover, there are policies that encourage college students to work at the grassroots level by enhancing the welfare of grassroots employees. For instance, the “Opinions on Accelerating the Promotion of Rural Talent Revitalization” issued by the Central Committee of China and the State Council in 2021 stated: “Implement the policy of subsidies for township work and allowances for difficult and remote areas, and ensure that the income of the staff of township organizations is higher than that of the staff at the same level in the county-level organizations”. Relevant policies are classified according to the level of personal development, compensation, and welfare, as well as the level of environmental facilities (see

Table 1. Contents of the rural grassroots employment policy handout for students.

Policy Content	Concrete Content
Personal development level	Provide ability and quality training as well as the necessary time and financial support for the employment of college students at the grassroots level; Post training and talent development plan, establish a reserve personnel selection system at the grassroots level, and prioritize the selection of cadres and talents by superior organizations and institutions, the promotion of posts (grades) of units at the same level, and the evaluation and employment of professional and technical posts (grades); College students working in entry-level positions receive “additional points” or “priority admission under the same terms” when applying to become graduate students or civil servants.
Salary and welfare level	Provide assistance with housing, medical treatment, children’s education, settlement, and professional title declaration and give priority to college graduates who have served at the grassroots level for an extended period of time, are devoted to their jobs, and have demonstrated exceptional performance.
Environmental facility level	Together with relevant departments, the department of human resources and social security shall establish a grassroots growth contact service mechanism for college graduates and conduct regular or irregular discussions, visits, condolences, and other activities. Education, health, agriculture, and other industry departments assist in resolving practical problems for college graduates in this industry.

).

3.2. Behavioral Policy Tools

Policy instruments are the means by which policy implementers allocate resources to achieve policy goals. Traditional public policy instruments are primarily regulatory in nature. With the development of behavioral public policies, public policy researchers and policy implementation practitioners have observed that cognitive and emotional interventions for individuals have clear effects on public policy implementation. This study divided behavioral policy tools into explicit policy tools and potential policy tools based on the classifications of policy tools by Bemelmans-Videc [17], Chengzhe Fu [18], and other scholars, as well as the experience of the British Behavioral Insight Group (

Table 2. Classification and interpretation of policy tools from a behavioral public policy perspective.

Nature of Policy Instruments	Policy Tools	Secondary Tools	Detailed Explanation
Explicit policy tools	Regulatory tools	Define a specific type of behavior as illegal and criminal.	Reduce the behaviors of policy objects that are inconsistent with policy expectations through enforcement and prohibition.
	Incentive tool	Provide convenience policies for housing, medical treatment, children’s education, settlement, and professional title declaration; provide personal long-term development assistance, etc.	Material and spiritual incentives to reinforce policy implementation
	Didactic tool	Education, publicity, etc.	Publicity entails subtly influencing the behavior and selection of policy objects.
Potential policy tools	Nudging tool	Job matching according to employment willingness, personality, ability, etc.	Use less obvious economic or administrative means to alter the choice structure and direct the behavior of policy objects, based on the premise of ensuring individual freedom of choice.

). The regulatory, incentive, and didactic types are categorized as explicit policy instruments, whereas the nudging type is categorized as a potential policy instrument.

The implementation of a policy must not be the result of a single policy tool, but of the comprehensive application of a variety of policy tools. When investigating the rural entrepreneurship policy of college students, Chengzhe Fu discovered that the effect of matching policy content with incentive tools and nudging tools is superior to that of speaking tools in genera [18]. Using Beijing’s environmental governance as an example, Wang Hongmei asserted that the implementation effect of economic incentive policy tools is superior to that of command control policy tools and proposed the reduction of command control policy tools and the development of public participation policy tools [19]. Wang Hongmei believes that command control and market incentives remain the most effective tools for environmental governance in China on a national scale [20]. Liu et al. believe that command control policy tools and reprimand policy tools are effective in promoting energy consumption monitoring technology in public buildings, whereas economic incentive policy tools are ineffective [21].

3.3. Behavior Modification of the Policy Impact

The behavior transformation of the policy effect can be interpreted as the change in the policy object during policy implementation, which can reflect the effect of resource allocation and the degree to which policy objectives are realized. DeHart-Davis proposed, from the perspective of behavioral public policy, that the policy effect is the change in the policy target behavior of individuals, which depends on the effect of the policy mechanism on individual cognition, thereby forming the policy behavior effect, i.e., the policy has an effect on the policy cognition level [22]. Fu Chengzhe and other scholars referred to the five-dimensional conceptual framework of effect, efficiency, political support, institutional support, and citizen support developed by Tummers [23], which identifies the degree of effective transformation of target group behavior and citizens’ recognition of policy as significant determinants of the policy behavior effect [18].

Measuring the degree of policy cognition and the impact of policies on the behavior of target groups, that is, the change in employment tendency, can measure the different policy effects in the application of different policy tools. Therefore, this study comprehensively considered the degree of policy cognition related to the behavior of the target group and the impact of policies on the behavior of target groups as dependent variables to reflect the change in employment tendency and the effect of policy behavior.

3.4. Policy of Reception versus Policy of Resistance

The policy of a state, political party, region, or unit is the code of conduct enacted in order to accomplish the directional and strategic goals of a particular historical period [24]. Active and passive attitudes toward policy implementation exist among policy targets. Existing research on college students’ willingness to start a business in rural areas indicated, for instance, that the majority of college graduates choose to start their own business, which is not a passive action that cannot lead to employment, but rather a deliberate active choice, i.e., an independent choice. Self-selection entrepreneurship accounts for 80% of the total number of entrepreneurs. In the data collected by the questionnaire designed for this paper, only 20% of the subjects chose to work in rural areas in response to the question “Where do you want to work?”. However, more than 50% of the subjects who chose to work in rural areas were forced to because “It is difficult to find a job in one’s own profession”, and passively chose to go to the rural grassroots level for employment.

Different policies demonstrate varying levels of initial subjective will and policy perception support. Some policies have strong subjective will and high policy recognition from the beginning. On the contrary, there are also policies with double low starting points; that is, the subjective will and policy recognition are both low. The policy attribute can be divided into receptive policy and resistive policies based on the policy object’s active versus passive attitude toward policy implementation. The policy object of a receptive policy has a favorable attitude toward the policy’s implementation. Without the formulation and implementation of relevant policies, the relevant behaviors of the policy object will also occur. For the resistive policy, without the formulation and implementation of pertinent policies, the relevant behaviors of the policy object will be uncommon or nonexistent. Implementing the policy of resistance is more difficult. Therefore, the difficulty and focus of policy implementation are precisely the resistive policy. The policy tools that can produce good policy effects in the resistive policy will be more universal. Research on the supply content and promotion mechanism of acceptance-oriented policies, such as college students’ rural entrepreneurship policy, Rural Revitalization policy, environmental protection policy, targeted poverty alleviation policy, and public welfare and charity policy, from the perspective of behavioral public policy, suggests that potential policy tools can encourage policy compliance by increasing the knowledge reserves and ability to distinguish policy intentions, and their policy effects are better than those of traditional explicit policy tools. In the context of receptive policy, the best use of nudging means the policy object taking the initiative to act consistently with the nudging intent. Can nudging, in the context of resistive policy, also produce better behavioral effects?

4. Research Hypothesis

When discussing the issue of inadequate endowment savings and investments in the United States, Seiler and his colleagues proposed the “save more tomorrow” plan. Its essence can be summed up in three points: First, in light of people’s inclination to deviate from the status quo, they should be automatically enrolled in the plan, and if they wish to opt out, they should fill out a form and choose to do so; automatic enrollment will significantly increase participation in the endowment savings plan. Second, allow individuals to choose whether to increase the savings rate “now” or “when the next salary increase occurs”; then require them to continue saving until they opt out of the pension plan or the savings rate reaches the maximum allowed. Linking higher savings rates with higher wages can reduce loss aversion; allowing individuals to make decisions that will not take effect for some time can reduce the impact of limited willpower. Third, the proportion of stock funds to bond funds is carefully calculated so that investors can still achieve the optimal allocation of stocks and bonds even if they employ a simple 1/n diversification strategy. The practice and data of the United States and other nations have demonstrated the plan’s effectiveness in encouraging pension savings and enhancing the investment effect. Thaler and Sunstein discovered that when the government promoted energy-saving and environmental protection measures, compared to promoting the use of these energy-saving and environmental protection measures by providing various subsidies, the policy effect achieved by changing the expression was superior, i.e., the following two phrases were used in the advertising: (1) If you take energy-saving measures, you will save 350 yuan per year; (2) if you do not, you will waste 350 yuan per year. Although both phrases convey the same meaning, slogan (2) employs the loss framework. Field-based research demonstrated that the impact of advertising language (2) is significantly greater than that of advertising language (1) [6]. Duflo et al. confirmed on-site that when faced with a lack of willpower, poor households prioritized immediate consumption after harvesting and delayed the purchase of chemical fertilizers. Finally, when there was no money to buy fertilizers at the time of planting, a temporary emergency could be resolved by giving money to the poor. The following year, however, poverty alleviation institutions were required to give money to the poor once again. When poverty alleviation institutions replaced the direct point of giving money to the poor with a discount when farmers had money after harvest, they more effectively improved the targeted poverty alleviation by purchasing fertilizer back and storing it [25].

Existing research is deficient in that even if the policy content does not use nudging policy tools, because the policy itself is an acceptance policy, the policy objects will still produce behaviors consistent with the policy objectives. Therefore, the nudging policy tools serve as “icing on the cake”, and the effect of the policy is evident. However, in the context of resistive policy, can the combination of nudging policy tools produce such blatant effects on policy behavior? Based on the objectives of investigating the influential mechanism of the match between the supply content of resistive policy and policy tools on the effect of policy behavior, as well as analyzing the optimal solution of the match between resistive policy and policy tools, this study divided the supply content of the rural grassroots employment policy for college students into three levels, i.e., personal development, salary and welfare, and environmental facilities and proposed the following:

Hypothesis 1 (H1):

With respect to the change in college students’ rural grassroots employment tendency, the effects of incentive, preaching, and nudging policy instruments are distinctly manifested.

Hypothesis 2 (H2):

With respect to the change in college students’ rural grassroots employment tendency, the effect of combining personal development policies with nudging policy tools is significantly superior to that of combining didactic and incentive policy tools.

Hypothesis 3 (H3):

With respect to the change in college students’ rural grassroots employment tendency, the effect of the combination of salary and welfare policies with nudging policy tools is significantly superior to that of the combination of didactic and incentive policy tools.

Hypothesis 4 (H4):

With respect to the change in college students’ rural grassroots employment tendency, the effect of combining policies at the level of environmental facilities with nudging policy tools is significantly superior to that of combining didactic and incentive policy tools.

The degree of policy recognition of the target group relative to the rural grassroots employment policy and the degree of policy change relative to the target group reflect the change in college students’ rural grassroots employment inclination.

5. Research Design

5.1. Research Methods and Samples

This study employed investigation and experiment to validate the relationship between the two policy variables and the effect of policy behavior. The investigation experimental method adhered to the logic of the experimental method and collected data in the form of an investigation, compensating for the lack of external validity of experimental research.

In conjunction with previous research, a questionnaire on the implementation and promotion mechanism for college students’ employment was developed and distributed online to the College Students’ Living Group of China University of Geosciences (Wuhan) on 30 September 2021 for preliminary research. A total of 328 questionnaires were recovered, and after excluding those with incomplete responses, 150 valid questionnaires remained (mainly sections of the experimental material design was not answered). The challenge of questionnaire design was to differentiate nudging policy tools from incentive policy tools and didactic policy tools and to minimize the subjects’ obvious interference with nudging policy tools. Therefore, the objective of the preliminary investigation was to determine whether the measurement experimental materials of the nudging policy tools were of sufficient quality. In practice, however, it was difficult to distinguish the three tools precisely. Due to the existence of individual differences, it was impossible to report the expansion of individual freedom in its purest form. Consequently, the experiment could only enhance the experimental design of the nudging policy instrument to the greatest extent.

The following aspects of the questionnaire were optimized based on the results of the preliminary survey: First, in terms of content, the experimental material of the initial nudging policy instrument was designed as “a post-test platform for graduates’ employment links, and each graduate was instructed to conduct a personality ability and post-matching test”. The design of the pre-survey materials overlapped with the use of the other two tools. Therefore, it was modified to “provide an interesting psychological and ability cognition test platform for all students in school, and allow them to participate independently in the matching test”. Because nearly 40% of the pre-survey questionnaires did not completely answer some questions in the experimental design, and in consideration of the possible burden on the subjects caused by the questions in the answer materials, encouraging phrases such as “Come on! The last paragraph of text” and “Answer here, the following questions only require a minute of your time” were added to the questionnaire before the experimental design. Because the subjects of the pre-survey were beyond the scope of college students, this study recruited volunteers from colleges and universities prior to administering the formal survey. Volunteers distributed questionnaires randomly in the dormitories of colleges and universities across the nation, imposed a ban on forwarding the questionnaires, and set a time limit for completing them, thereby excluding non-college students as subjects.

Following this, the questionnaire was distributed with the assistance of the Star online survey platform, which eliminated paper-based forms and increased the survey’s efficiency. The platform supported random scenario generation. Each experimental questionnaire contained random information, and subjects were assigned to different experimental groups at random. The online survey platform finally collected 648 valid responses. The obtained data included college students from 1693 colleges and universities in 28 provinces in China, including Shandong, Henan, and Hubei. According to the enrollment age of Chinese college students, they typically enroll in college at 18 years or older. In general, university study lasts three to four years, master’s study lasts two to three years, and doctoral study lasts three to four or more years. The sample demonstrated that the responses were largely representative of college students. It conformed to the previous data specifications. The sample was demographically representative of the general population in terms of gender ratio, age ratio, and major (

Table 3. Description and statistics of the fundamental sample characteristics (n = 648).

Statistical Variables		Number of Samples	Percentage
Gender	male	315	48.6%
	female	333	51.4%
Age	18–21 years old	273	42.1%
	22–25 years old	198	30.6%
	26 years or older	177	27.3%
Major	Philosophy	53	8.2%
	Economics	85	12.7%
	Law	42	6.5%
	Education	64	9.9%
	Literature	76	11.7%
	History	98	15.1%
	Neo Confucianism	86	13.3%
	Engineering	31	4.8%
	Agronomy	16	2.5%
	Medical Science	18	2.8%
	Management	65	10%
	Art Studies	14	2.2%

).

5.2. Design and Procedure

The questionnaire used in the experiment consisted of six parts:

Part I collected basic information; Part II was the actual policy cognition measurement, which focused primarily on the subjects’ comprehension of the grassroots employment policy for college students. The third section focused primarily on the subjects’ attitudes toward rural grassroots employment prior to reading relevant experimental materials; the fourth section consisted of experimental manipulation materials and situational test questions matched with policy content and behavioral policy tools. This portion of the experimental manipulation materials was modified in accordance with the most recent policy documents. Subjects read various experimental materials and took measurements under varying conditions. This research institute adapted the content into the form of a news report. Examples of policy matching incentive tools at the level of personal development contained the relevant text:

In a recent interview, the Minister of Human Resources and Social Security stated that college graduates are crucial to the revitalization of rural talent. In order to promote college graduate employment in rural communities, the state has also issued a number of support policies. H Province responded actively to the national appeal by launching a series of rural grassroots employment talent training projects and providing financial assistance. W City and J City proposed that graduates who have completed their rural grassroots employment service within the past three years should apply for a master’s degree. Under the same conditions, graduate students can receive additional points on the preliminary examination and preferential admission.

According to the various combinations of policy content and behavior policy instruments, there were nine distinct experimental data points (

Table 4. Experimental material design of the independent variable policy content.

Policy Content	Experimental Material Design
Personal development level	Post training and talent development plan, establish a grassroots reserve personnel selection system and give priority to the selection of cadres and talents by superior organizations and institutions, the promotion of posts (grades) of units at the same level, and the evaluation and employment of professional and technical posts (grades).
Salary and welfare level	Provide support in housing, medical treatment, and other aspects, select “the most accomplished grassroots college graduates”, and publicize their excellent deeds as typical examples.
Environmental facility level	Establish a contact service mechanism for the growth of college graduates at the grassroots level and conduct regular or irregular discussions and visits and provide support to help college students employed at the grassroots level to solve practical difficulties.

and

Table 5. Experimental material design of the independent variable behavior policy tool.

Nature of Policy Instruments	Behavioral Policy Tools	Experimental Material Design
Explicit policy tools	Incentive tool	Ensure sufficient funds for training programs and gradually increase and optimize all-round subsidies for grassroots rural employment.
	Didactic tool	All localities and cities are required to hold at least one publicity and mobilization meeting on the grassroots employment policy for college students every month and to focus on the grassroots employment guidance of college graduates in the graduation season.
Potential policy tools	Nudging tool	Launch the post test platform for college students and conduct post matching according to employment intention and personal ability.

). The fifth section was the measurement of the policy behavior effect, which focused primarily on the subjects’ subjective understanding of rural grassroots employment and their intention to work at the grassroots level after reading the experimental materials.

5.3. Variable Measurement

The independent variables in the experiment were the policy content and the behavior policy tool, while the dependent variable was the policy behavior effect, which was primarily reflected by the degree of policy recognition and the degree to which the policy changed the behavior of the target group. “Which of the following policies do you believe the above rural grassroots employment materials involve?” was the test item for the manipulation effect. “Which of the following policy contents do you think are involved in the above rural grassroots employment materials?” Options corresponded to three types of policy content and behavioral policy tools. The greater the proportion of correct responses for this section of the questionnaire, the greater the manipulation effect of the independent variables.

In the measurement of the policy’s effect on the behavior of subjects, the degree of recognition of the rural grassroots employment policy and the degree to which the policy changed the behavior of the target group were examined.

The experimental control variable was the rural grassroots employment pre-test. In other words, prior to reading the experimental materials, the participants were asked about their attitudes toward rural grassroots employment (Figure 1).

After the test of reliability and validity, Cronbach’s alpha coefficient was 0.843, and the Kaiser–Meyer–Olkin (KMO) value was 0.705.

5.4. Balance Test

The random distribution of materials within each group was examined via analysis of variance of gender, age, major, rural grassroots employment education, rural practice or grassroots internship experience, and actual grassroots employment policy knowledge. The results indicated that there were no significant differences between experimental groups, and the equilibrium was satisfactory (

Table 6. Balance test of experimental groups.

Experimental Group	Sample Size	Gender (M ± SD)	Age (M ± SD)	Major (M ± SD)	Employment Guidance and Education (M ± SD)	Social Practice or Internship Employment Experience (M ± SD)	Realistic Policy Cognition (M ± SD)
Group H1	72	1.61 ± 0.49	1.61 ± 0.49	5.33 ± 3.41	3.69 ± 1.15	3.56 ± 1.31	2.17 ± 0.99
Group H2	72	1.53 ± 0.50	1.53 ± 0.50	5.14 ± 3.01	3.72 ± 1.12	3.50 ± 1.15	2.25 ± 1.21
Group H3	72	1.42 ± 0.49	1.56 ± 0.50	5.89 ± 2.97	3.50 ± 1.19	3.58 ± 1.17	2.47 ± 1.12
Group H4	70	1.40 ± 0.49	1.57 ± 0.49	4.94 ± 2.83	3.71 ± 1.11	3.77 ± 1.05	2.29 ± 1.03
Group H5	72	1.69 ± 0.46	1.83 ± 0.37	5.08 ± 2.37	3.75 ± 1.19	4.00 ± 0.97	2.33 ± 1.03
Group H6	72	1.43 ± 0.49	1.51 ± 0.50	5.90 ± 3.03	3.51 ± 1.17	3.82 ± 1.13	2.29 ± 0.95
Group H7	72	1.47 ± 0.50	1.53 ± 0.50	6.56 ± 3.08	3.53 ± 1.04	3.47 ± 1.39	2.58 ± 1.07
Group H8	74	1.54 ± 0.50	1.59 ± 0.49	6.11 ± 3.21	3.49 ± 1.03	3.73 ± 1.23	2.32 ± 1.04
Group H9	72	1.53 ± 0.50	1.58 ± 0.49	5.81 ± 3.36	3.39 ± 1.09	3.39 ± 1.24	2.44 ± 1.26
F		2.71	2.84	2.22	0.98	1.95	0.99
p		0.06	0.14	0.24	0.45	0.05	0.44

).

The average rate of correct responses to the manipulation test questions in each experimental group was between 57.2 and 76.3%, which demonstrated that the manipulation effect was positive.

6. Research Results

6.1. Descriptive Statistics of the Research Variables

The full score for each item used for the measurement of actual policy knowledge (

Table 7. Realistic policy cognition scale.

Index	Question Options	Item Mean	Dimension Mean
Realistic policy cognition	I know more about rural employment policies	2.35	9.39
	I am very familiar with the rural employment policy for college students.	2.33
	I think the government has formulated many policies to attract college students to rural employment.	2.32
	The school has given guidance on rural employment.	2.39

) was 5, the mean value of the measurement results was 2.3, and the mean value of the dimension was 9.39, indicating that college students’ knowledge of rural grassroots employment policy was low. Therefore, it is necessary to increase the visibility of rural grassroots employment policy among college students.

The total score for each item in the measurement of the policy behavior effect (

Table 8. Measurement scale of policy behavior effects.

Index	Question Options	Item Mean	Dimension Mean
Realistic policy cognition	Support for grassroots employment is provided by the policies in the materials.	2.38	5.93
	Based on the policies in the materials, I will consider becoming a rural grassroots worker.	3.55

) was 5. The average value of the item “Based on the policies in the materials, I will consider becoming a rural grassroots worker” was 3.55, indicating that college students were optimistic regarding the rural grassroots employment policy and the achievement of policy goals.

6.2. Regression Analysis of the Policy Behavior Effect

This study provided preliminary confirmation of the impact of three behavioral policy tools—incentive, preaching, and nudging—on college students’ rural grassroots employment policy. Regression analysis revealed that the aforementioned three policy tools had a positive effect on the college students’ behavior toward the rural grassroots employment policy (the regression of the policy behavior effect on the real policy cognition was obtained, and the coefficient of the real policy cognition was positive and statistically significant (b = 1.558, p < 0.1)). In light of the model’s simplicity, the model’s regression analysis results are omitted. The control variable in the experiment was the pre-test of college students’ rural grassroots employment intent. According to Model (1) (

Table 9. Regression analysis results of the policy behavior effects of the policy content and policy tools.

		Model (1)
		Policy Behavior Effect
		Beta	t
Experimental intervention	Behavioral policy tools (reference group: Incentive Tools)
	Didactic tool	0.226	5.078 ***
	Nudging tool	0.012	0.281
	Pre-test of rural grassroots employment willingness	0.072	1.829 **
	_cons	7.555
	$R^2$	0.004
	N	648

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

), there was a significant difference in the behavior effect between the incentive policy tools and the preaching policy tools (b = 0.226, p < 0.05), but there was no significant difference between the incentive policy tools and the nudging policy tools (b = 0.012, p > 0.1). There was a significant difference between the preaching policy tools and the nudging policy tools in terms of their effect on behavior. In addition, the effect of the didactic policy tools was superior to that of the incentive policy tools and the nudging policy tools (obtained by altering the regression’s reference group). Hypothesis H1 was partially supported.

To further test the effect of the three different behavior policy instruments, Model (1) was constructed with the behavior effect of college students’ rural grassroots employment policy as the dependent variable, and Models (2) and (3) were constructed with the two items in the policy behavior effect scale as the dependent variables (

Table 10. Regression analysis results from a behavior effect model of policy content and behavior policy tools.

		Model (1)		Model (2)		Model (3)
		Policy Behavior Effect		Support for Grassroots Employment Provided by the Policies in the Materials		Based on the Policies in the Materials, I Will Consider Becoming a Rural Grassroots Worker
Experimental intervention		Beta	t	Beta	t	Beta	t
	Interaction item (Reference: incentive tool)
	Personal development level and didactic tools	0.337	4.529 ***	0.186	2.387 **	0.282	3.823 ***
	Personal development level and nudging tools	0.002	0.018	0.143	1.836 *	−0.123	−1.167 *
	Compensation and benefits and didactic tools	0.208	2.671 ***	0.169	2.139 **	0.152	1.937 *
	Compensation and benefit level and nudging tool	0.001	0.10	0.078	0.697	−0.058	−0.520
	Environmental facilities and didactic tools	0.126	1.602	0.156	2.004 **	0.054	0.688
	Environmental facility level and nudging tool	0.042	0.540	0.187	2.402 **	−0.062	−0.790
Pre-test of rural grassroots employment willingness		0.072	1.829 **	0.148	3.817 ***	−0.020	−0.501
	_cons	7.555		4.032		3.523
	$R^2$	0.004		0.020		0.001
	N	648		648		648

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

). The following are the results of the analysis (given the simplicity of the model, only the regression analysis results with the matching incentive policy tools as the control group are presented):

First, at the level of personal development, the policy behavior effect when combined with didactic policy tools differed significantly from that when combined with incentive policy tools and nudging policy tools and was more pronounced when combined with didactic policy tools. Consequently, Hypothesis H2 was not confirmed. Moreover, the combination of policy content at the level of personal development with didactic policy tools significantly enhanced the professional identity of college students and policy support for rural grassroots workers.

Second, with respect to compensation and welfare policy, there was a significant difference between the effect of matching preaching policy tools and nudging policy tools, and the effect of matching preaching policy tools on policy behavior was marginally superior to that of matching nudging policy tools. The validity of H3 was not established.

Third, for environmental facilities, there was a significant difference in policy support when preaching and nudging tools were combined, with the effect being greater when nudging tools were combined. Consequently, Hypothesis H4 was confirmed.

The experiment determined that Hypotheses H1 and H4 were verified based on the above cause-and-effect analysis of the independent variable policy content, behavioral policy tools, and dependent variable policy behavior effects. The effects of incentive, didactic, and nudging policy instruments on the alteration of college students’ rural grassroots employment preferences were distinctly manifested. The policy matching nudging policy tools had a significantly greater impact on environmental facilities than the matching didactic and incentive policy tools. Hypotheses H2 and H3 were not tenable, as they failed to demonstrate that the effect of the combination of the personal development level policy and compensation and welfare level policy with boost-type policy tools had obvious advantages over the combination of didactic-type policy tools and incentive-type policy tools with respect to the change in college students’ rural grassroots employment tendency.

7. Conclusion and Discussion

7.1. What Kind of Policies Are the Nudging Policy Tools Effective for?

Why did the nudging policy tool fail to demonstrate significant benefits in matching personal development, compensation, and welfare level policy? The first reason is that the rural grassroots employment policy for college students selected for this study belonged to the resistance-type policy, and the policy objectives were inconsistent with the subjective desires of the policy objects; consequently, the policy behavior effect of the boost-type policy tools was not readily apparent. The policy’s effect had stages. Generally, the path of development for the resistive policy manifested as the transformation to the acceptance policy. Initially, it displayed a resistive characteristic. With the aid of policy instruments, it gradually demonstrates the acceptance characteristic. Simultaneously, the receptive policy exhibited the tendency to transform into the resistive policy, and the two policies were capable of transforming into one another. There was a distinction between the acceptance type and the resistive type in the policy’s content in the absence of a policy tool. The primary function of policy tools is to transform a policy of resistance into one of acceptance. At the initiation of all policies, the degree of policy recognition and the degree of behavior change among the policy’s target group are distinct. The second reason is that most policies, including those of the acceptance type, have the problem of initial fuzzy cognition of policy objects, which makes the role of nudging policy tools less apparent than that of more explicit incentive policy tools or preaching policy tools. In the initial phase of policy implementation, even if the policy is consistent with the policy object’s subjective will, the policy object’s vague policy cognition may prevent it from comprehending the policy’s intention and cause it to resist the policy. Therefore, explicit policy tools such as incentive policy tools and didactic policy tools are more effective than potential nudging policy tools.

7.2. How to Match Nudging Policy Tools to Give Full Play to Their Advantages

The effective combination of policy content and policy tools is more beneficial for enhancing the effect of policy behavior according to various policy attributes. This study concluded that the degree of the effect of the nudging tool must differentiate between receptive and resistive policies. In the context of the receptive policy, nudging policy tools will match the policy content at all levels effectively. The effect of aligning the policy content of environmental facilities with nudging policy instruments is more evident in the context of resistive policy. The effect of matching the nudging policy tools at the level of personal development, compensation, and welfare lags behind the effect on policy behavior produced by the policy content and the nudging policy tools under the receptive policy. This indicates that the compatibility between the resistive policy content and policy tools must be adjusted based on the policy conditions. The appropriate “precise matching” scheme (see

Table 11. Matching effect of resistive policy content and behavior policy tools.

Policy Content and Policy Tools	Collocation Effect
Personal development and didactic policy tools	Policy behavior effect, policy identity, professional identity
Salary and welfare level and didactic policy tool	Policy behavior effect, policy identity, professional identity
Environmental facility level and nudging policy tools	Policy identity

) can enhance the efficacy of rural community-based employment policies for college students.

Under a resistive policy environment, the policy behavior effect of combining the policy content of the personal development level and salary and welfare level with preaching policy tools is greater than that of combining nudging policy tools. In other words, college students have a higher degree of policy support for rural grassroots employment policy and rural grassroots workers’ professional identity.

The policy effect is produced by the policy matching nudging tools at the level of environmental facilities, that is, college students have a higher degree of policy approval of rural grassroots employment policy identity, and the degree to which the policy changes the behavior of the target group is higher. Thus, a platform for college students’ ability and job matching should demonstrate the rural grassroots working and living environment and other nudging policy tools, which are more effective in improving the behavior of the target group.

Moreover, the benefits of nudging tools cannot be overlooked. Compared to more traditional incentive and preaching policy tools, the nudging tool employs the behavioral public policy principle and has the advantages of low cost and non-compulsion. Despite the fact that it does not demonstrate complete advantages in promoting the implementation of resistive policy, it is still capable of producing behavioral effects. Moreover, it demonstrates that the application of nudging tools must be grounded in reality and emphasizes applicability.

8. Countermeasures and Suggestions

8.1. The Matching of Policy Tools Should Pay Attention to the Distinction of Policy Attributes

In the initial phase of policy implementation, we should differentiate the matching of the policy content with explicit policy tools and potential policy tools. In terms of the content of policy supply and the combination of policy tools, incentive policy tools and didactic policy tools are given priority in the policies of personal development, salary, and welfare. The government, colleges and universities, and rural grassroots cooperatives have formed a rural grassroots employment information-sharing platform for college students, including the rural grassroots employment policy for college students, the rural grassroots employment guidance of college students, and the introduction of the rural grassroots employment living environment, so that college students can effectively obtain rural grassroots employment information and further understand the rural grassroots employment situation. At the same time, through the official WeChat account and microblog for colleges and universities, typical examples of rural grassroots employment of college students are publicized, and the government takes the lead in recommending talent from colleges and universities to establish a rural grassroots employment guidance team for college students [26]. Offline operations such as “College Students’ rural grassroots employment guidance at colleges and universities” are routinely conducted to provide face-to-face guidance and answer questions, so as to reduce college students’ cognitive bias against rural grassroots employment policies and to stimulate college students’ interest in rural grassroots employment [27].

In addition, it remains essential to focus on the combination of nudging policy tools and policy content at the level of environmental facilities by testing the psychological ability and personality of college students during school, providing a post matching test platform and offering career-planning services for rural grassroots employment graduates. On the basis of understanding the needs of college students for rural grassroots employment at the level of environmental facilities, we can use potentially promotional policy tools to dispel their concerns about the employment environment, social cognition, and development opportunities of rural grassroots employment [28,29,30,31,32,33]. College students can then fully understand that they are suited to grassroots employment and believe that they can achieve their life goals at the grassroots level.

8.2. The Matching of Policy Tools Should Pay Attention to the Stages of the Policy

Policies of resistance and acceptance can be transformed into one another, and this transformation should be considered when matching policy instruments. When a policy is in the resistance stage, the policy tools should be precisely matched to the policy’s maturity, an understanding of the policy object, and other factors that aid in its implementation. For college students without a clear understanding of rural grassroots employment and without a plan for it, the use of didactic policy tools should be emphasized. Improving the degree of policy recognition entails continuously enhancing college students’ and the general public’s comprehension of this policy, elucidating the policy’s intent, and implementing incentives tailored to the needs of policy objects. This also facilitates the gradual transition from a policy of resistance to one of acceptance. Therefore, in the later stage of policy implementation, it can be said that the nudging policy tools occupied a dominant position in the transformation of policy types and gave full play to their advantages, such as providing a test platform for college students’ psychology, ability, and personality and giving them the opportunity to better comprehend basic level rural employment and connect themselves with it. Finally, the rural employment policy for college students will be transformed from “nonsense” to “sense” [34,35,36].

In the context of resistive policy, the combination of preaching policy tools plays a greater role in enhancing the behavioral effect of college students’ rural grassroots employment policy. In light of differences in the research on the role of nudging tools in the effect of policy behavior from the perspective of behavioral public policy, this paper questions whether the role of nudging tools in the implementation of receptive and resistive policies is related to the nudging intensity, i.e., whether increasing the nudging intensity of resistive policies can improve the effectiveness of nudging tools in the process of policy implementation and whether the nudging policy tools can significantly enhance the effect of policy behavior when they are used together with other policy tools in the implementation of resistive policies. In subsequent research, we will focus on the aforementioned notions, delve deeper into the promotion mechanism of resistive policy, promote the policy content and policy tools that better meet the needs of policy objects, and improve the efficacy of policy formulation and implementation.