A quantitative assessment of Chinese county high school policies using the PMC index model.
PloS one · 2026
Abstract originale
The county high school policies (CHSPs) is crucial for improving the quality of basic education in China's counties and promoting rural social development. As policy documents are the foundational elements of policy systems, it is crucial to evaluate current policy texts, identify their shortcomings, and provide empirical evidence for future policy development. This study collected 55 policy texts from provincial and municipal governments, utilizing the Policy Modeling Consistency index model to evaluate 14 representative policies. The results indicate that all the assessed policies fall within the"pass","good",or"with distinction" score category, with no substandard policies identified and none achieving a "perfect" score.A detailed analysis identifies critical policy shortcomings, including the prioritization of short-to-medium-term strategies over long-term planning, fragmented enrollment systems, insufficient campus culture development, as well as limited career guidance, mental health support,and policy guidance for home-school-community joint education. Local governments adopt a conservative stance on educational policy evaluation and systemic institutional reforms.To address these issues, this study recommends that local governments integrate the enrollment systems of public and private high schools within their jurisdictions, strengthen the prohibition of on cross-regional enrollment, increase investment in students' career guidance and mental health education, and launch more programs to encourage parents and communities to participate in school education. Moreover, the central government should urge local authorities to accelerate reforms in the high school education system.
Parole chiave
Leggi il testo integrale originale
Testo estratto dalla pubblicazione: sono esclusi immagini, media, supplementi e parti soggette a diritti distinti.
A quantitative assessment of Chinese county high school policies using the PMC index model A quantitative assessment of Chinese county high school
The county high school policies (CHSPs) is crucial for improving the quality of basic education in China’s counties and promoting rural social development. As policy documents are the foundational elements of policy systems, it is crucial to evaluate current policy texts, identify their shortcomings, and provide empirical evidence for future policy development. This study collected 55 policy texts from provincial and municipal governments, utilizing the Policy Modeling Consistency index model to evaluate 14 representative policies. The results indicate that all the assessed policies fall within the“pass”,“good”,or“with distinction” score category, with no substandard policies identified and none achieving a “perfect” score.A detailed analysis identifies critical policy shortcomings, including the prioritization of short-to-medium-term strategies over long-term planning, fragmented enrollment systems, insufficient campus culture development, as well as limited career guidance, mental health support,and policy guidance for home-school-community joint education. Local governments adopt a conservative stance on educational policy evaluation and systemic institutional reforms.To address these issues, this study recommends that local governments integrate the enrollment systems of public and private high schools within their jurisdictions, strengthen the prohibition of on cross-regional enrollment, increase investment in students’ career guidance and mental health education, and launch more programs to encourage parents and communities to participate in school education. Moreover, the central government should urge local authorities to accelerate reforms in the high school education system.
1. Introduction
In China, county high schools are senior high schools run by county-level governments, which provide general education. They have the highest educational quality and accommodate the largest student population at local level. Normally, there are one to three of these schools in an ordinary county, while in county-level cities with larger economic aggregate and population size, their number may be higher.
County high schools are the highest level of basic education at the county area in China; therefore, the quality of running schools is an important indicator of the level of local county education [1]. According to the Ministry of Education, in 2024 high school enrollment in county, town, and rural-level areas reached 13,175,995 students, 45% of the national total high school enrollment [2], the number of high schools in these areas reached 6,426 institutions, constituting 40% of all high schools nationwide [3]. Furthermore, county high schools are the most important element of the county-level education system. As the final stage of basic education in counties, they provide the only pathway for local students to enter universities. At the same time, they serve as an intermediate link in the education system, bridging basic education and higher education [4]. Accordingly, because of the number of rural students they serve and their institutional position in China’s education system, their teaching quality is directly related to educational equity between urban and rural areas in China.
In the past, county high schools enjoyed a fairly long period of prosperity, as a direct result of China’s basic education policy. In 1953, the central government proposed the establishment of 194 national key high schools [5]. Later on in 1994, the State Council of China promulgated relevant policies, requiring each county to run one or two well-developed middle schools [6]. Accordingly, one to two county high schools have been established in each county, representing the highest educational level and acquiring a key position in the county basic education system. Therefore, the rapid development of county high schools over a long period of time is largely attributable to the key high school policy. Although China revised the Compulsory Education Law of 2006 and abolished the classification schools into key schools and non-key schools, key senior high schools still exist in practice, due to historical inertia and the Matthew effect [7].
In recent years, a phenomenon of county high school decline, characterized by a persistent brain drain of high-quality teachers/students from county high schools and a continuous decline in the admission rates to universities of village students, has been frequently reported by media outlets nationwide, sparking widespread concern. One major cause of this problem is that a number of key high schools thriving under the key high school policy have received investments from privacy capital; this, coupled with their public-owned nature and deliberate policy guidance from the government, leads to their gradual evolution into “super high schools” [8]. These schools not only take advantage of their superior resources and brand influence to attract teachers from county high schools, but also recruit high-achieving junior high school students from county areas by offering substantial bonuses, leading to a massive loss of teachers and student resources in county high schools [9]. Furthermore, the lack of regulation over student mobility has exacerbated this problem [10].
In order to solve this problem, in December 2021 the Ministry of Education, together with nine other Departments, issued the “14th
Five-Year Plan” for the Promotion of County High Schools(hereinafter referred to as “14th Five-Year Plan”), which is the first guiding document focusing on the county level [11]. This Plan aims to ensure that the educational quality of county high schools meets the requirements of equitable access to basic public education put forward in the outline of the 14th
Five-Year Plan for National Economic and Social Development and Long-Range Objectives for 2035 [12]. To further consolidate this goal, the “Plan for Building China into a Leading Country in Education (2024-2025)” clearly set the objective of “boosting the development of senior high schools in county-level areas” [13]. On the one hand, these policies shows that the existing problems of county high schools make them the “weak link” of the basic education at the county level, which does not align to the national requirements for urban-rural integration and development and for new urbanization construction [1], showing the need for continuous efforts and the continuous attention of the academic community. On the other hand, with the end of the 14th Five-Year Plan and the development of the following one, the problems of the current county policy and the expected direction of future interventions have increased the need for targeted county policy research. Therefore, this study targets county high schools policies(CHSPs) issued by local governments as its research subjects.
This study addresses the following issues:
What is the quality of the CHSPs currently issued by local governments in China?
Which specific issues exist in the CHSPs formulated by local governments?
How can these CHSPs be improved?
The main contributions of this study are as follows: Firstly, it employed text mining techniques to conduct an in-depth analysis of China’s county high school policies, evaluating their content and providing novel perspectives for related research. Secondly, it introduced a methodological innovation to existing research, establishing a quantitative evaluation framework for Chinese CHSPs based on the policy modeling consistency (PMC) index model. Thirdly, it selected 55 policy documents as research samples and conducted a quantitative assessment of 14 representative policies, generating new empirical evidence to inform adjustments to China’s CHSPs.
This paper is structured as follows: Section 2 reviews the pertinent literature; Section 3 outlines the research design; Section 4 details the construction of the evaluation index system; Section 5 describes the policy assessment process; Section 6 presents the calculation results of the PMC index and generates surface plots; Section 7 provides a detailed analysis of the research findings; and Sections 8 and 9 outline the conclusions and policy proposal, respectively.
2. Literature review
2.1. Existing research on CHSPs
Existing research on the widespread operational difficulties and declining educational quality in China’s county high schools, can be broadly grouped into three categories. The first category of studies adopted the analytical lens of internal educational organizational structures, attributing the decline of county high schools to the unregulated expansion of regional key high schools and super high schools.For example, Yang and Wang argued that the industrialization and commercialization of super high schools undermine local basic education systems through cross-regional recruitment of top students and high-salary incentives to attract quality teachers [14]. The expansion of super high schools exacerbates inter-school disparities in educational resources, information, and reputation among regular high schools, thereby intensifying the Matthew effect between high schools [15], and significantly reducing the quality of education in all regular high schools [16].
The second category of studies posits that the decline of county high schools is the result of urbanization-driven population migration mechanisms. The migration of population to central cities leads to the concentration of rural teachers and students in urban areas, and the siphoning effect of super high schools exacerbates the decline of county high schools [17]. Moreover, prolonged urbanization has engendered a collective psychological expectation of urban educational superiority over rural counterparts in rural communities, thereby triggering an exodus from county high schools that ultimately result in the decline of these institutions through a self-fulfilling prophecy mechanism [18]. Even more disconcertingly, empirical evidence demonstrates the nationwide prevalence of this phenomenon [19].
The third category of studies examined the institutional determinants of county high school decline through the lens of educational governance and policy implementation mechanisms. These studies claim that, due to China’s fiscal decentralization framework, county-level governments bear full responsibility for educational funding, while being subject to two types of constraints: the finite fiscal capacity of local treasuries and the compulsory education expenditure mandates stipulated in Article 42 of the Compulsory Education Law of the People’s Republic of China. This structural contradiction systematically restricts fiscal allocations to county high schools, resulting in chronic underfunding of secondary education in rural jurisdictions [20]. In addition, quantitative research demonstrated that precisely due to the strong correlation between county-level education quality and local fiscal revenue capacity, the educational quality in China’s coastal county regions consistently surpasses that of inland areas [21]. Furthermore, scholars identified that a fundamental cause of subpar teaching quality in county high schools lies in unrestricted recruitment practices, through which elite urban institutions systematically recruit academically outstanding students from rural counterparts, thereby triggering a sustained brain drain that demoralizes teaching faculties [8]. In terms of policy analysis, some authors employed text coding and frequency analysis to examine policy priorities across 26 provincial-level CHSPs documents [22], offering useful methodological insights. However, these studies did not cover city-level policies. In fact, the idealized CHSPs framework must achieve two imperatives, i.e., universalization through comprehensive coverage and the diversification of educational provision, while enforcing rigorous controls to ensure that inter-regional disparities in teaching quality remain within an acceptable range [6]. These policy objectives encounter significant implementation challenges stemming from a three-dimensional institutional friction: an ambiguous delineation of governmental mandates, cognitive-resource mismatches with target beneficiary cohorts, and binding constraints within the policy ecosystem [23]. To solve these problems, Deng proposed a theoretical framework [24]. Other scholars employed sociological and political theoretical frameworks to assess the decline of county high schools, including elements such as cultural ecological reconstruction [25] and national educational governance [26].
These studies provided multidimensional perspectives to understand and address the decline of county high schools. However, it is noteworthy that they focused mainly on analyzing the causes of this decline and proposing remedial strategies, with limited attention directed toward the institutional design and operational mechanisms of CHSPs. This oversight has resulted in significant research gaps regarding systemic solutions, leaving future policy development inadequately supported by empirical evidence.
2.2. Policy content analyses
Recent years have witnessed significant advancements in quantitative policy text research, with a methodological diversification across three primary categories: policy metrology analysis, policy content quantification, and policy text mining. The analysis of policy content typically includes both a qualitative and a quantitative examination of policy texts [27]. Policy text quantification consists in the transformation of unstructured policy documents into abstract, feature-specific structured data through systematic conversion paradigms.
Conceptualized by Ruiz Estrada through the Omnia Mobilis theoretical lens, the PMC index modeling framework has distinct advantages over conventional text mining and mechanical element analysis methods. Traditional text mining methods are predominantly centered around word frequency statistics, and are primarily suited for thematic analysis and sentiment analysis of long texts [28,29]; however, they neglect the distinctive features of policy texts and are devoid of tools to assess, their quality. Simultaneously, mechanical element-based approaches depend on manually assigned element weights, which are intrinsically affected by the subjective factors of researchers. In this study, we chose the PMC index model as our main research method for the following reasons: (1)its parameter configuration based on policy text mining minimizes subjective human evaluation, thus substantially enhancing assessment objectivity and precision; (2)it deliberately avoids the hierarchical weighting of indicators, eliminating restrictions on variable quantity and significance, thereby enabling the comprehensive capture of potential influencing factors; (3)it facilitates cross-policy consistency analysis and visual representation of policy effectiveness, providing scientifically robust foundations for policy optimization; (4) its operational simplicity and cost-effectiveness enhance its practical applicability.
The PMC index model has gained increasing recognition in Chinese public policy research with extensive applications in sectors such as industrial policy [30–32], public health policy [33], energy policy [34], environmental policy [35],and public safety policy [36]. Previous empirical studies validated its effectiveness in large-scale public policy evaluation, demonstrating particular adaptability to educational policy analysis. Moreover, recent academic explorations have successfully implemented this model across diverse educational domains, including compulsory education policy [37] and higher education policy [38–40]. These applications collectively affirm the scientific validity and feasibility of application of this model in educational public policy research. Accordingly, this study employed the PMC index model to analyze the implementation of CHSPs by local government during the 14th Five-Year Plan period.
3. Research design
This study performed a horizontal assessment of representative local policy documents of county high schools, with the aim to understand the policy focus of local governments and identify their deficiencies.
3.1. Data sources
This study used keywords such as “county-level high schools”, “county high schools” and “regular county high schools” to search the website of the Ministry of Education, the website of Chinese government, and the PKULAW database as supplementary and proofreading sources. The policy texts were selected for investigation based on the following requirements:
They have been issued by provincial, city and county governments
They consist in policy documents such as plans, opinions and notices; and
Their content is highly related to the establishment of county high schools.
Policy texts whose topic is not related to the research subjects, as well as informal documents such as notifications, replies, and speeches were excluded from analysis.
Finally 55 documents were selected for variable setting and parameter identification, see S1 Appendix, issued in the period from December 1, 2021 to August 30, 2024.
3.2. Research method
This study employed three research methods. First, word frequency statistics were employed to identify the most important elements in policy texts and determine the specific elements of policy priorities and policy objects among the first-level indicators.
Second, text coding was performed using Grounded Theory(GT). Given the complexity of policy priorities and policy objects, due to the fact that multiple measures in policy texts may refer to a same element, while at the same time a single element may also cover various measures it was necessary to categorize these elements through text coding to construct a third-level indicator system.
Thirdly, this study employed the PMC index model, which allow a quantitative evaluation of policy texts after completing text preprocessing. Further details on these three methods are presented in the following subsections.
3.2.1. Word frequency statistics.
Word frequency statistics are one of the common methods for policy text measurement, and is mainly used to display and describe the basic situation, the overall characteristics and the evolution trends of policy collections [41]. This method is also applied to the preprocessing of policy texts by relevant studies that adopted the PMC index model for policy evaluation [42].
In this study, word frequency statistics were performed on the 55 collected policy texts by calling Python’s Jieba segmentation module. The purpose of this approach was to intuitively identify the focus elements of these policy texts, which would serve as an important reference for the subsequent construction of the indicator system.
3.2.2. Grounded theory (GT).
Firs proposed by Glaser and Strauss [43], GT is articulated in three iterative steps: Open Coding, Axial Coding and Selective Coding. During the whole process, researchers continuously compare, revise and integrate data with the concepts and categories established at the outset until theoretical saturation is achieved, i.e., until when newly collected data will no longer yield new attributes or insights for the established theoretical categories. This ensure that the theories generated by GT are rooted in abundant real-world data, boasting strong empirical relevance and explanatory power.
In this study, GT was adopted because the PMC index model requires the construction of an indicator system. However, as the the indicator table constructed by Estrada [29] could not be directly adopted, and customized indicator systems had to be built for different policy texts. Therefore, the indicator system was constructed based on the texts of the CHSPs, so that to provide a sufficient basis for each indicator and enhance their credibility.
To this purpose, the semantics in the original text were transformed into measurable indicators through the coding and induction of the original policy text.The indicators established in this way are mainly the tertiary indicators in policy priorities and policy objectives.
3.2.3 The PMC index model.
First proposed by Mario Arturo Ruiz Estrada, the PMC index model is a classic tool for quantitative policy evaluation. Rooted in the omnia mobilis (everything is moving) assumption, it breaks through the limitation of the traditional ceteris paribus (all else being equal) assumption, enabling multi-dimensional and systematic assessment by incorporating all relevant variables into a unified analytical framework [28,29].
This model consists in converting unstructured policy texts into quantifiable values by constructing a multi-level evaluation indicator system that typically includes 9–10 first-level indicators (e.g., policy duration, nature, audience, focus, and function), each subdivided into several second-level indicators. A binary scoring method is adopted, assigning 1 for compliance with criteria and 0 otherwise; the final PMC index, calculated via weighted aggregation, reflects overall policy quality, with higher values indicating better policy performance.
4. The research process
Estrada’s PMC model is articulated in six steps: 1. Establish a multi-input-output table; 2. Confirm variables and specific parameters; 3. Calculate the PMC index; 4. Formulate consistency evaluation criteria for policy modeling; 5. Draw the PMC surface graph; 6. Analyze the advantages and disadvantages of policy modeling based on evaluation criteria [29].After reviewing relevant target literature, it was found that Chinese scholars followed five steps when applying this model policy text collection and preprocessing, indicator system construction, PMC index calculation, PMC surface graph drawing and comprehensive policy analysis [40].The specific process is shown in Fig 1.
First, we had to collect and process the acquired texts. Among the various existing methods for preprocessing, we adopted word frequency statistics and GT. Next, based on the results of the preprocessing. Then, we established the indicator system for the PMC index model. Then we scored the selected policy texts as evaluation objects in accordance with the indicator system, after which, we calculated the PMC index using specific formulas and placed the results into a 3×3 matrix to generate a surface graph, which enabled an intuitive analysis of the strengths and weaknesses of each policy text.
4.1. First-level variable classification and parameter identification
Before constructing the PMC index model, the evaluation indicator system had to be established. First, in line with the 14th Five-Year Plan,this study identified three primary first-level variables: policy objectives, policy priorities, and policy safeguards. These three indicators were established level 1 indicators. Second, the 55 compiled and screened policy texts were analyzed using Jieba for part-of-speech tagging and frequency statistics. Functionally insubstantial verbs, adjectives, and adverbs commonly found in policy documents were filtered out. This process yielded a ranked list of the top 50 keywords with the highest frequency of occurrence, presented in Table 1. It can be seen that“schools”, “teachers”, “principals” and “students” have a high frequency of occurrence these were classified as policy objects. High-frequency terms such as “model”, “supervision” and “fund” were categorized as policy guarantees. Finally, “school operations”, “teaching” and “student enrollment” were categorized as policy priorities.
10.1371/journal.pone.0361807.t001Table 1The 50 most frequently occurring terms in the analyzed policy.
Rank Keyword Frequency Rank Keyword Frequency
1 County high schools 5575 26 County area 329
2 Education 1638 27 Synthesize 319
3 Develop 1566 28 Standard 303
4 School 1342 29 High quality 300
5 Teacher 1338 30 Support 290
6 Build 1017 31 Guide 288
7 Promote 989 32 Government 273
8 Student enrollment 966 33 Cultivate 271
9 School operations 670 34 Plan 257
10 Entrusted management assistance 484 35 Supervision 254
11 Put into effect 481 36 Strictly Forbid 247
12 Reform 474 37 College entrance examination 243
13 Principal 472 38 Coordinate 243
14 Manage 428 39 Instructional research 240
15 Teaching 409 40 Standardization 238
16 Department 393 41 Plan as a whole 236
17 Model 388 42 Country 235
18 Appraise 383 43 Improve 229
19 Assurance 372 44 Power 229
20 Student 372 45 Method 219
21 Funds 367 46 Regular 212
22 Duty 362 47 Organization 209
23 Characteristic 357 48 Running conditions 204
24 Horizontal 350 49 Large class 200
25 Quality 332 50 Condition 195
Note: All the words in this table were translated from Chinese.
4.2. Three-level variable classification and parameter identification
In order to accurately identify the second and third-level variables, this study coded the 55 policy documents investigated using NVivo 12 plus software transforming the original text into a conceptualized version through a systematic inductive analysis.
4.2.1. Open coding.
Open coding consists in disassembling the original policy text, and identifying and tagging the potential concepts and categories as much as possible. In this study, the 55 policy texts were coded line by line, obtaining 2481 reference points in which 271 initial concepts were summarized. Next, the initial concepts with similar meanings were summarized, determining 48 initial categories, as shown in S2 Appendix.
4.2.2. Axial coding.
Axial coding consists in analyzing and summarize the initial categories on the basis of open coding, identifying the main categories according to their internal logic and category. In this study, 48 initial categories were summarized and grouped into 14 main categories enrollment, teaching quality, teaching reform, entrusted management assistance, coordinated development, schools, teachers, principals, students, main responsibility, organization and leadership, financial support, publicity and guidance, supervision and assessment.
4.2.3. Selective coding.
Selective coding consists in the further induction and differentiation of the axial coding. In this study, enrollment, teaching quality, teaching reform, entrusted management assistance, and coordinated development were identified as the policy priorities of local counties schools, teachers, principals and students were identified as policy objects and main responsibility, organization and leadership, financial support, publicity, and guidance, and supervision and assessment were identified as policy guarantees.
4.2.4. Coding reliability.
In this study, all the three stages of open, axial, and selective coding were performed jointly by the two authors and a doctoral candidate specializing in educational policy. The corresponding author formulated the coding guidelines for the unified policy texts, while the first author and the doctoral candidate conducted independent line-by-line coding and category induction. To ensure coding consistency and objectivity, 10 policy documents were randomly selected for inter-coder reliability testing using the Cohen’s Kappa coefficient. After cross-checking and a consensus discussion on inconsistent coding entries, the final value of the Cohen’s Kappa coefficient reached 0.78, indicating a good level of inter-coder agreement. The unified coding standards and reliability control were consistently implemented throughout all the three coding procedures to guarantee the methodological rigor of the qualitative analysis.
4.2.5. Saturation test.
A systematic saturation test after completing the coding of the original 55 CHSPs documents. Three additional policy texts were purposively sampled to represent distinct administrative levels: (1) a county-level comprehensive education improvement policy issued by the Debao County, Guangxi Zhuang Autonomous Region; (2) a provincial-level specialized high school development policy issued by Henan Province; and (3) a municipal-level medium-to-long-term education development plan issued by Huizhou City, Guangdong Province. The selected documents are: “Notice of the People’s Government of Debao County on Printing and Distributing the Implementation Plan for Debao County to Implement the Three-Year Action Plan for Education Quality Improvement in Guangxi (2021-2023)”“Notice of the General Office of Henan Provincial Department of Education on Further Promoting the Diversified Development of Ordinary High Schools” and “Implementation Plan for Huizhou “Thousands of High-Quality Development Projects” in Education (2023–2027).” Notably, as these policies only partially address county high school development, only those text segments explicitly mentioning county high schools were coded for the saturation test. The authors applied the pre-established coding manual to these relevant text segments following the identical open, axial, and selective coding procedures used for the main sample. Coding discrepancies were resolved through consensus discussion. No new initial concepts, subcategories, main categories, or causal/associative relationships between categories were identified in any of the three documents. This confirmed that the theoretical framework developed in this study achieved full theoretical saturation, with no additional information requiring integration into the indicator system.
Based on the results of the word frequency statistics analysis, the GT analysis, Estrada [29] and relevant domestic research and the characteristics of local and county policies, and on relevant expert opinions, 9 first-level indicators, 35 second-level indicators and 48 third-level indicators were finally obtained, as shown in Table 2.
10.1371/journal.pone.0361807.t002Table 2Hierarchical structure of policy evaluation variables.
Level 1 variable Level 2 variables Level 3 variables Reference
X1 Policy functions X1:1 Directive role — Estrada [29]
X1:2 Supporting role —
X1:3 Encouragement effect —
X1:4 Regulatory role —
X2 Policy statute of limitations X2:1 Long-term (over 5 years) — Estrada [29]
X2:2 Mid-term (3–5 years) —
X2:3 Short-term (1–3 years) —
X3:1 Students enrollment X3:1–1 Integrated network system
X3:1–2 Enrollment regulation
X3:1–3 School-Based Quota Allocation
X3:1–4 Enrollment rate
X3:1–5 General-Vocational education integration
X3:2 Teaching quality X3:2–1 Feature-based school development
X3:2–2 Informatization
X3:2–3 Dissemination of teaching outcomes
X3:2–4 Instructional research System
X3:2–5 Simultaneous development of the five types of educations
X3:3 Teaching reform X3:3–1 Evaluation reform
X3:3–2 Teaching innovation
X3:3–3 Curriculum reform
X3:3–4 Optional class system
X3 Policy priorities X3:4 Entrusted management assistance X3:4–1 Guidance and training Word frequency statistics and GT
X3:4–2 Paired assistance
X3:4–3 Flow of talents
X3:4–4 Legalization
X3:4–5 Fund guarantee
X3:4–6 Commendation and award
X3:5 Coordinated development X3:5–1 Coordination between public and private schools
X3:5–2 General-vocational coordination
X3:5–3 Regional coordination
X4:1 School X4:1–1 Layout optimization
X4:1–2 Standardization
X4:1–3 Reduction of oversized classes
X4:1–4 Information
X4:1–5 School-running funds
X4:1–6 Cultural construction
X4:1–7 Accommodation and health conditions
X4:1–8 Scale regulation
X4:1–9 School-running conditions
X4:1–10 Safety assurance
X4:2 Teacher X4:2–1 Teaching staff replenishment
X4:2–2 Training
X4:2–3 teacher personnel management
X4:2–4 Teacher’s ethics
X4:2–5 Incentives
X4 Policy object X4:3 Principal X4:3–1 Principal Team building Word frequency statistics and GT
X4:3–2 Competency training
X4:3–3 Employment regulations
X4:3–4 Incentive mechanism reform
X4:4 Students X4:4–1 Educational expenditure per student
X4:4–2 Scholarships and grants
X4:4–3 Development guidance
X4:4–4 Personalized cultivation
X4:4–5 Comprehensive quality
X4:4–6 Collaborative education
X5 Policy tools X5:1 Command-type tool — McDonnell, Elmore [44] and Schneider, Ingram [45]
X5:2 Incentive tool —
X5:3 Capacity-building tool —
X5:4 Exhortation tool —
X5:5 System reform tool —
X6 Policy guarantee X6:1 Principal responsibility — Kuang et al. [46]
X6:2 Organization and leadership —
X6:3 Financial support —
X6:4 Publicity and guidance —
X6:5 Supervision and assessment —
X7 Policy operability X7:1 Specific scheme — Qian et al. [47]
X7:2 Detailed measures —
X7:3 Supporting policies —
X7:4 Clear Division of Responsibilities —
X8 Policy evaluation X8:1 Clear goals — Cui et al. [40]
X8:2 Sufficient evidence —
X8:3 Detailed planning —
X8:4 Rational Strategy —
X9 File reference X9:1 File reference — Estrada [29]
Note: In the variable of simultaneous development of the five types of educations (X3:2–5), these include moral, intellectual, physical, aesthetic, and labor education. X4:4–6 Collaborative Education refers to family-school-community collaborative education.
5. The policy evaluation process
5.1. Selection of policy texts
The selection of policy texts to enter the evaluation phase followed these criteria:
Selection of two cities or provinces from each of China’s seven major regions: Northeast China, North China, Central China, East China, South China, Southwest China, and Northwest China.
Selection of cities or provinces with a relatively developed economy and society, and a high population density.
These selected cities or provinces feature a high level of urbanization, as well as a more prominent imbalance in the distribution of educational resources between urban and rural areas and a serious decline of county high schools [19]. The final selected policy texts are shown in Table 3.
10.1371/journal.pone.0361807.t003Table 3List of CHSP texts.
Number Title of the Policy Document Release Date Area
P1 Jilin Province “14th Five-Year Plan” county-level high schools development and promotion action plan 2022 Northeast China
P2 Action Plan for the Development and Promotion of county-level high schools in Shenyang City (2023–2025) 2023 Northeast China
P3 Shanxi Province “14th Five-Year Plan” county-level high schools development and promotion action plan 2021 North China
P4 Implementation plan for the development and promotion of county-level high schools in Hulunbuir city and county level 2023 North China
P5 Implementation plan of county-level high schools for the “14th Five-Year Plan” Plan of Henan Province 2022 Central China
P6 Action Plan for the Development and Promotion of county-level high schools in Changsha City (2023–2025) 2023 Central China
P7 Action Plan for the development and promotion of county-level high schools in the “14th Five-Year Plan” Plan period of Fujian Province 2022 Eastern China
P8 Action plan for quality improvement of ordinary high schools in 4 mountainous counties of Hangzhou 2023 Eastern China
P9 Action Plan for the Development and promotion of regular high schools in counties during the “14th Five-Year Plan” period of Guangdong Province 2023 South China
P10 Nanning city “14th five-year” county-level high schools development and promotion action plan implementation plan 2023 South China
P11 Opinions on the Implementation of the action plan for the Development and Promotion of county-level high schools in the “14th Five-Year Plan” Period in Guizhou Province 2022 Southwest China
P12 Implementation plan of the action plan for development and promotion of regular county high schools during the “14th Five-Year Plan” period in Chengdu 2023 Southwest China
P13 Gansu Province “14th Five-Year Plan” county-level high schools development and promotion action plan 2022 Northwest China
P14 Development of county-level high schools in the “14th Five-Year Plan” county seat of Ningxia Hui Autonomous Region 2022 Northwest China
Note: All the title in the table are translated from Chinese.
5.2. Establishment of a multi-input-output table
A multi-input-output table was constructed as shown in S2 Appendix, where the policy text was assigned according to a binary system, including two-level and three-level indicators. More in detail, a value of 1 was assigned to those policy texts proposing the relevant keywords or including a detailed description, and 0 otherwise as shown in S3 Appendix.
To mitigate subjective bias and ensure the objectivity of the scoring process, a standardized multi-rater scoring procedure was implemented following the development of the evaluation framework. A purposive sample of 7 graduate students was recruited based on their disciplinary expertise, including 2 students specializing in public policy and 5 students specializing on educational policy. All the raters received systematic training on the standardized scoring guidelines presented in Table 6, and completed a pilot scoring exercise to ensure a consistent understanding of the evaluation criteria. Each rater independently scored 2 distinct policy documents using the established rubric. As raters were not assigned the same policy texts, inter-rater reliability coefficients (e.g., Cohen’s Kappa) were not calculated. The scoring results were directly aggregated and cross-checked for consistency, and any ambiguous cases resolved through group discussion. The final aggregated scoring results are presented in S4 Appendix.
5.3. Calculation of the PMC index model
The PMC index model was calculated based on the scoring results presented in S4 Appendix, according to which the scores of the first-level variable for each policy document investigated were obtained. The calculation formulas are as follows:
X~𝒩(0,1)(1)X=XR∈[0,1](2)Xm={1k∑n=1kXmn,m=1,2,5,6,7,8,91kl∑n=1k∑l=1LXmnl,m=3,4(3)PMC=∑n=14X1n4+X2∑n=12X2n2+X3∑n=15∑l=1LX3nlkl+X4∑n=14∑l=1LX4nlkl+X5∑n=14X5n4+X6∑n=15X6n2+X7∑n=14X7n2+X8∑n=14X8n2+X9(4)The scores of all first-level variables were between [0,1]. Next, a sum was performed of these scores to calculate the PMC index. Finally, the 14 policies were evaluated following the criteria presented in Table 4 as modified by Estrada [29].
10.1371/journal.pone.0361807.t004Table 4Criteria for the evaluation of the policy PMC scores.
PMC score 0–5.99 6–6.99 7–7.99 8–8.99 9–10
Evaluation of the score Fail Pass Good With distinction Perfect
Where, m is the ordinal number of first-level variables, textitm = 1, 2, 3, 4, 5, 6, 7, 8, 9. n is the ordinal number of second-level variables, ranging from 1 to k depending on the specific group. l is the ordinal number of third-level variables specifically for groups m = 3 and m = 4, ranging from 1 to L. k and L represent the total counts of second-level and third-level variables respectively in the corresponding analysis.
5.4. Building the PMC surface graph
A PMC surface graph was built based on the calculation results of the PMC index model, so as to intuitively present the advantages and disadvantages of the policy indicators. Specific the nine first-level indicators of the sample policy into a 3×3 matrix, as follow:
PMC surface={X1X2X3X4X5X6X7X8X9}(5)
6. Computation results and scores visualization
Based on the computational results, the performance measurement and the comparison scores, as well as the corresponding evaluation grades for the 14 policies investigated were systematically performed; the results are shown in Table 5. The matrix of all PMC scores is presented in Table 6. For the mapping procedure Python MATLAB’s mesh grid function was employed to construct a 3D spatial grid, which served as the foundation for subsequent surface matrix fitting. Utilizing the V4 interpolation method, a smooth 3D surface map was generated, which effectively shows the complex relationships between policy attributes and performance outcomes. The 3D surface map presented in Fig 2 and Fig 3 shows 14 primary policy indicators convex areas indicate higher scores, while concave regions indicate lower scores, and darker tones indicate weaker outcomes. P14 showed an overall superior quality with localized deficiencies at X2 and X3, while P1 exhibited systemic underperformance except at X1, X6 and X9. The following Section presented the analysis of the policies from both a horizontal and a vertical perspective.
10.1371/journal.pone.0361807.t005Table 5PMC scores and evaluation grades of policy texts.
MetricScore X1 X2 X3 X4 X5 X6 X7 X8 X9 PMC ScoreGrade
P1 1.00 0.67 0.75 0.64 0.40 1.00 0.75 0.75 1.00 6.96 Pass
P2 1.00 0.33 0.79 0.81 0.60 1.00 1.00 1.00 1.00 7.53 Good
P3 1.00 0.67 0.84 0.83 0.80 1.00 0.75 1.00 1.00 7.88 Good
P4 1.00 0.33 0.96 0.76 0.60 0.75 1.00 0.75 1.00 7.16 Good
P5 1.00 0.67 0.69 0.63 1.00 1.00 0.75 1.00 1.00 7.74 Good
P6 1.00 0.33 0.64 0.63 0.80 1.00 1.00 1.00 1.00 7.40 Good
P7 1.00 0.67 0.88 0.84 1.00 1.00 0.75 0.75 1.00 7.89 Good
P8 1.00 0.67 0.71 0.68 1.00 1.00 0.75 0.75 1.00 7.56 Good
P9 1.00 0.33 0.97 0.78 1.00 1.00 0.75 1.00 1.00 7.83 Good
P10 1.00 0.33 0.85 0.86 0.60 1.00 1.00 1.00 1.00 7.64 Good
P11 1.00 0.67 0.77 0.87 1.00 1.00 1.00 1.00 1.00 8.30 With distinction
P12 1.00 0.33 0.82 0.83 0.80 1.00 1.00 1.00 1.00 7.78 Good
P13 1.00 0.67 0.84 0.85 1.00 1.00 1.00 1.00 1.00 8.36 With distinction
P14 1.00 0.67 0.77 0.92 1.00 1.00 1.00 1.00 1.00 8.35 With distinction
Average 1.00 0.52 0.81 0.78 0.83 0.98 0.89 0.93 1.00 7.74 Good
10.1371/journal.pone.0361807.t006Table 6Values of the PMC matrices of the policy texts.
No. P1 P2 P3 P4
Matrix 1.00 0.67 0.75 1.00 0.33 0.79 1.00 0.67 0.84 1.00 0.33 0.96
0.64 0.40 1.00 0.81 0.60 1.00 0.83 0.80 1.00 0.76 0.60 0.75
0.75 0.75 1.00 1.00 1.00 1.00 0.75 1.00 1.00 1.00 0.75 1.00
No. P5 P6 P7 P8
Matrix 1.00 0.67 0.69 1.00 0.33 0.64 1.00 0.67 0.88 1.00 0.67 0.71
0.63 1.00 1.00 0.63 0.80 1.00 0.84 0.80 1.00 0.68 1.00 1.00
0.75 1.00 1.00 1.00 1.00 1.00 0.75 0.75 1.00 0.75 0.75 1.00
No. P9 P10 P11 P12
Matrix 1.00 0.33 0.97 1.00 0.33 0.85 1.00 0.67 0.77 1.00 0.33 0.82
0.78 1.00 1.00 0.86 0.60 1.00 0.87 1.00 1.00 0.83 0.80 1.00
0.75 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
No. P13 P14
Matrix 1.00 0.67 0.84 1.00 0.67 0.77
0.85 1.00 1.00 0.92 1.00 1.00
1.00 1.00 1.00 1.00 1.00 1.00
7. Outcome analysis
7.1. Horizontal evaluation
Based on the results presented in Table 5, among the 14 sampled policy documents, 10 were classified as“good” and 3 as “with distinction” no policies received a “fail” rating. The average score across all samples was 7.74, indicating an overall satisfactory quality of regular CHSPs. This suggests that local policy initiatives have provided strong support for regular county high school education in China. However, the absence of policies achieving a “perfect” rating as well as the fact that 6 policies scored below the average reveal room for further optimization and improvement.
The only policy that received a “pass” score was P1, issued by Jilin Province, received a PMC score of 6.96 (pass). Several indicators of this policy were lower than the average value, namely X3, X4, X5, X7, and X8, with gaps of 0.05, 0.14, 0.43, 0.14, and 0.18 respectively.In particular the weakest variables were those under X5, related to a notable lack of tools to improve leadership capabilities, individual and family education concept innovation and reform the education system. Hence, the recommended improvements should be prioritized along this path: X5 > X8 > X7 > X4 > X3.
P2, issued by Shenyang City, Liaoning Province, had a score of 7.53 (good), which was near the average. This policy achieved full marks in X1, X6, X7, X8, and X9, while it underperformed in X2, X3, and X5 with a gap of 0.19, 0.02, and 0.23 below the average respectively. The primary cause of the point deduction in indicator X5 can be attributed to the insufficient utilization of persuasive tools and system change mechanisms. More specifically, there is a notable absence of advocacy for the renewal of educational concepts within the context of personal and family education, as well as in policy documents aimed at systemic educational reform. This is unacceptable for a systemic problem. The recommended improvements should be prioritized along, this path: X2 > X5 > X3.
P3, issued by Shanxi Province, had a score of 7.88 on the PMC index (good). All the indicators scored above average except X7, which had a gap of 0.14 above average, this indicating a high quality. This gap from perfect, primarily results from the policy’s lack of detailed implementation measures. Hence, it is recommended to prioritize the refinement X7 by developing concrete operational guidelines.
P4, issued by Hulunbuir City, Inner Mongolia Autonomous Region, had a scored of 7.16 (good). In this policy indicators X2, X4, X5, X6, and X7 underperformed, with a gap of 0.19, 0.02, 0.23, 0.23, and 0.18 below average, respectively. These low scores stem from an insufficient use of persuasive policy tools and systemic reform mechanisms, issues seen also in P2. The recommended improvements should be prioritized along this path: X5 > X6 > X2 > X7 > X4.
P5, issued by Henan Province, had a score of 7.74 (good), in line with the average. In this policy, only X3, X4 and X7 underperformed, with a gap of 0.12, 0.15, and 0.14 from the average respectively, demonstrating relatively sound policy design. The recommended improvements should be prioritized along this path: X4 > X7 > X3.
P6, issued by Changsha, attained a PMC score of 7.40 (good). Indicators X2, X3, X4, and X5 had a performance below average, with a gap of 0.19, 0.17, 0.15, and 0.03 respectively. The low score in X2 is attributed to the policy’s recent enactment in 2023 with only three years of implementation history. Additionally, underperformance in X3 stems from insufficient data coverage in sub-indicators X3:1, X4:3, and X4:4. The recommended improvements prioritized along this path: X2 > X3 > X4 > X5.
P7, issued by Fujian Province, achieved a score of 7.89 (good), i.e., higher than the average. Only two indicators, X7 and X8 underperformed, with a gap of 0.14 and 0.18 from the average respectively. This result suggests that this policy demonstrates structural soundness in its core objectives, although lacking robustness in implementation safeguards. The corrective actions should be prioritized as follows: X8 > X7.
P8, issued by Hangzhou, attained a score of 7.56 (good). Indicators X3, X4, X7, and X8 underperformed, with a gap of 0.09, 0.10, 0.14, and 0.18 respectively, mirroring the safeguard deficiencies observed in P7. Notably, in sub-indicator X4:4, this policy scored only 2 out of 6 metrics, indicating inadequate support mechanisms for high school students. Improvements should be prioritized along this path: X8 > X7 > X4 > X3.
P9, issued by Guangdong Province, had a score of 7.83 (good), above the average. Only two indicators, X2 and X7, scored below average, with a gap of 0.19 and 0.14. The lower scores are mainly due to the policy’s late release in 2023 and lack of detailed implementation plans across government departments. Improvements should be prioritized as follows: X2 > X7.
P10, issued by Nanning, Guangxi Province, and P12, issued by Chengdu, Sichuan Province, achieved a score of 7.64 and 7.78, respectively (both good). Both policies underperformed in X2 and X5 indicators, with P10’s X5 gap (0.23) being significantly larger than P12’s (0.03). This is because P10 lacks initiatives to promote new educational concepts for individuals and families, while P12’s smaller X5 gap indicates that it has partially implemented them. Notably, both share an identical X2 gap (0.19), reflecting their short implementation periods (under 3 years). Improvements should be prioritized along the following path: P10:X5 > X2; P12: X2 > X5.
P11, issued by Guizhou Province, and P14, issued by Ningxia Hui Autonomous Region, were grouped for analysis as both underperformed only in the X3 indicator and exceeded the average scores for all others indicators. P11 hand a PMC score of 8.30 and P14 a score of 8.35, both attaining the distinction grade. Detailed analysis revealed that P11’s primary weakness was in sub-indicator X3:4, where it failed to meet 3 out of 6 sub-criteria. In contrast, P14 exhibited minor scattered gaps across X3 sub-indicators, contributing to its marginally higher overall score. To address these issues, P11 should focus on enhancing policy guidance for establishing structured partnerships between high-performing schools and disadvantaged institutions.
P13, issued by Gansu Province, achieved the highest PMC score of 8.36 among all 14 policies. While it demonstrated no significant weaknesses, minor limitations prevent it from reaching perfection. These include its mid-term policy framework (3–5 years) and lack of advanced reforms in classroom innovation, student development guidance, and home-school collaboration. The observed constraints suggest that medium-term policies may struggle to implement complex educational reforms, highlighting the necessity for long-term strategic planning within the CHSPs framework to enable systemic transformation.
7.2. Vertical evaluation
7.2.1. Policy functions (X1).
With a mean score of 1.0, all the policies investigated comprehensively covered the four core policy functions (directive, supportive, encouraging, and regulatory roles). The “predictive role” and “descriptive role” were not included because Chinese government policy often places them in separate advisory reports, rather than the official policy text.
7.2.2 Policy statute of limitations (X2).
The lowest mean score (0.50) for X2 stems from the national CHSPs mandate on county-level education reforms. Most local policies are aligned with the national timeline targeting completion within the 14th Five-Year plan period, resulting in temporally focused, medium or short term orientations.
From the perspective of the 14th Five-Year Plan as a whole, in 2021 government prioritized addressing the rapid and widespread decline of county high schools through CHSPs, mandating local governments to immediately curb the spread and exacerbation of these issues within the five-year timeframe. Consequently, local CHSPs implementation predominantly focused on short-to-medium-term policy objectives, while retaining limited long-term goals. This indicates that at that time, the central government had already recognized the systemic nature of county high schools’ challenges. In reality, the root cause of county high schools’ fragility lies in student outflow driven by urbanization [48]. To confront this systemic issue, it is necessary that CHSPs incorporate long-term countermeasures, a requirement that underscores the value of system-change policy tools in addressing structural educational disparities.
7.2.3. Policy priorities (X3).
In X3:1, only eight policies addressed X3:1–1; this may potentially reflect regional disparities in the digital enrollment infrastructure. Likewise, only seven policies emphasized X3:1–5, highlighting the insufficient recognition of the role of vocational-technical education in the county education system.
Student enrollment critically reflects educational equity. Drawing on the experience of China’s online college admission system, the adoption of a unified online enrollment platform can effectively curb the enrollment “black box”, minimize human intervention, reduce operational costs, and improve administrative efficiency [49]. However, this study indicate persistent deficiencies in systemic integration and network transformation within current frameworks. While most regions in China have achieved advanced educational informatization, private high schools operating outside unified enrollment platforms undermine equity. The 2022 Notice of the Ministry of Education reiterated the urgency of establishing standardized digital admissions systems to address irregularities such as premature enrollment, over-quota recruitment, and cross-district violations [50].
The low evaluation of general-vocational integration underscores an insufficient governmental recognition of the role of vocational education in enhancing county high schools. This integration is a current trend amid economic development and industrial restructuring, an inevitable direction following the popularization of senior high school education in China, and also an essential requirement to advance the equal development of general education and vocational education [51]. One form of distinctive development of county high schools lies in the integrated development of general education and vocational education. Senior high schools adopting this integrated model are also known as comprehensive senior high schools. These schools not only offer students the opportunity to take the college entrance examination and pursue higher education, but also provide an alternative career-oriented path [52]. Therefore, the reform integrating vocational and general education serves as a new approach for differentiated development between county high schools and urban senior high schools.
7.2.4. Policy object (X4).
In X4:1 low scores in cultural construction (7 policies) and safety assurance (3 policies) indicate insufficient investment in campus physical environment and risk management. School cultural development, encompassing tangible infrastructure (architecture, curricula) and intangible elements (ethos, values), critically shapes educational outcomes through material-immaterial synergy. However county high schools, burdened by academic competition and student quality disparities, often marginalize cultural construction in favor of test performance, exacerbating institutional homogenization [53]. Specialized programs (tangible classroom initiatives) must align with school cultural identity (intangible collective ethos) to drive institutional distinctiveness. For county high schools to achieve specialized development, curricula must reflect cultural uniqueness, while cultural identity requires grounding in tailored educational practices [54]. Thus,to strengthen campus culture, policy-supported institutionalization to harmonize specialized programs with school-specific cultural narratives, breaking homogeneity through integrated tangible-intangible strategies.
Furthermore, the score for campus safety was relatively low. The reason is that China has formulated dedicated policies concerning the safety of primary and secondary school campuses safety, including campus safety management [55], student accidental injury [56], bullying [57], food safety [58], mental health [59], and health management [60]. These policies have created a relatively complete system, whose scope keeps expanding together with social development [61]. Therefore, the fact that few CHSPs contain provision on campus safety can be attributed to the soundness of additional relevant policies. Moreover, county high schools are currently facing challenges in the field of teaching rather than in relation to campus safety.
In X4:4, X4:4–3 only 6 policies policies had a score under X4:3, and only 6 and 3 policies had a score under X4:4–4 and X4:4–6, respectively. In contrast all policies had a score under X4:4–1, indicating a preference for material support (e.g., funding per student) over holistic student development.
This study identified the root cause of low scores in these three indicators as systemic deficiencies in home-school collaboration policies. Specifically, the county high school education model adopts intensive exam-focused regimens with closed-campus management to compete with urban schools despite resource disadvantages [9]. This spatial segregation exacerbates family separation, particularly for rural students with migrant worker parents. While these schools prioritize academic metrics, they systematically neglect students’ holistic developmental needs. As a pivotal juncture in students’ lives, high schools bear the responsibility of providing comprehensive career and academic guidance–-a need acutely felt in county high schools, where students often lack parental or professional mentorship. Insular county societies and parents’ absence further exacerbate students’ myopic or disengaged attitudes toward career planning. A previous study revealed the following persistent issues: fragmented policy interpretation, insufficient regional coordination, narrow guidance scopes skewed toward utilitarianism, and inadequate specialization [62]. Concurrently, county and rural high school students face significant challenges in relation academic performance [63], mental health [64], and career planning capabilities [65], underscoring the urgent need for policy-driven enhancements in student guidance.
In 2023, 13 ministries jointly issued the “Opinions on Improving Collaborative Education Mechanisms Among Schools, Families, and Society”, aiming to build a cohesive, efficient system by 2035 [66]. Collaborative education seeks to transition family education from a “self-directed” state to a government-guided, multi-stakeholder model. However, in fact, family engagement remains as the weakest link. Previous research indicated that the educational attitudes of rural families shaped by migrant parents’ limited socioeconomic status significantly hinder children’s access to higher education as economic precarity fosters short-termism and neglect of psychological/moral development, with severity inversely correlated to socioeconomic standing [67]. This shows the need for policy interventions to cultivate sound educational philosophies in rural households, particularly through parent-oriented guidance.
7.2.5. Policy tools (X5).
Eight policies had a score for X5:5; this indicates that local governments maintain a conservative stance toward the Ministry of Education’s requirement to transition from a “county-led” to a “county-city joint operation” educational governance system.
Based on the distribution of policy instruments, local policies tend to favor more conservative policy tools, aligning with the traditional reformist mindset of incremental adjustments from easy to difficult and from shallow to deep interventions. The text of the Ministry’s “14th Five-Year Plan” indicates that the central government still expects local governments to actively explore systematic transformational policy instruments, thereby implementing top-down systemic reforms of the county and municipal high school education systems.
The evolution of county high schools management–-from historical county-dominated governance and operational oversight to the current exploration of a bifurcated system prioritizing prefecture-level administration for school operations and county-level management–-reflects a paradigm shift forced by institutional decline [68]. The waning trajectory of county high schools has compelled local governments to transition from fragmented governance to coordinated development frameworks. While county governments bear the responsibility to revitalize these institutions, greater accountability rests with prefecture-level authorities, necessitating cross-departmental collaboration beyond education bureaus [69].
From a macro-level perspective, China’s accelerating urbanization and increasing urban-rural integration necessitate the spatial redistribution of public services (including education) from urban centers to county-level regions [70]. Concurrently, addressing student attrition in county high schools could alleviate enrollment pressures in urban institutions, incentivizing prefecture-level governments to assume expanded operational obligations. The decline of county high schools is inextricably tied to fiscal constraints at the county level, and is exacerbated by the systematic absorption by urban high schools of top county students–-a dynamic demanding fiscal compensation mechanisms. Consequently, local governments must advance a tripartite governance model: provincial-level coordination, prefecture-level school operation, and county-level management. This institutional architecture aims to establish sustainable trusteeship and support mechanisms for county high schools. Empirical evidence demonstrates that provincial fiscal support yields more pronounced effects than municipal fiscal support in enhancing public service provision at the county level, particularly in poverty-stricken counties within mountainous regions [71]. The integration of educational funds for underdeveloped counties into provincial fiscal counterpart support mechanisms helps elevate the educational quality of local county high schools.
The strong performance observed in X6-X9 reflects the policy design capabilities of local government. As all the policies investigated are provincial/municipal-level frameworks, they intentionally allow flexibility for lower-level governments to adapt implementation plans based on local contexts. This explains why some policies show minor gaps in specific sub-indicators such as X7:2 and X8:3. Given CHSPs non-sensitive nature, X9 requires no further analysis.
8. Limitations
Due to technical and resource constraints, not all CHSPs were analyzed. More in detail, due to policy issuance hierarchy,this study analyzed mainly those CHSPs that have been formulated at the provincial and municipal levels, resulting in the inclusion of a lower number of CHSPs issued by county-level governments in the policy text corpus.
During the policy scoring phase, only 14 policy texts were distributed to 7 experts for evaluation, meaning inter-rater reliability could not be fully validated. However, during the subsequent proofreading process, both authors verified each individual score. The final scores were established only after confirming that there were no obvious errors or points of contention.
9. Policy recommendations
First, the systemic challenges of county high school decline demand a fundamental restructuring of the current CHSPs framework, which remains constrained by short-term policy cycles. To break this cycle, future policies should establish decadal strategic roadmaps centered on a holistic education reform. This requires a transition from exam-centric evaluation systems to competency-based frameworks, while creating county-level innovation hubs for pedagogical experimentation. For example, the integration of vocational-academic pathways through dual-track certification models could revitalize rural education ecosystems suffering from “cultural desertification.” These measures would help rural high schools overcome their overreliance on test-oriented drilling and genuinely implement quality-oriented education. Concurrently, CHSPs should mandate adaptive strategies for county high schools to align with the evolving national college entrance examination requirements, replacing entrenched exam-preparation models with dynamic curricula that cultivate both academic rigor and practical skills across all rural learners.
Second, a critical lever for equity lies in the digitization of enrollment systems through province-wide unified platforms. By consolidating the admission processes to public and private high schools, these platforms would not only enhance transparency and fairness–-laying the groundwork for procedural justice in education, but also enable real-time monitoring of student mobility. This data infrastructure would empower local governments to optimize resource allocation using predictive analytics, ensuring that funding decisions align with emerging educational needs. Provincial pilots in Jiangsu demonstrate that these types of digital ecosystems reduce administrative corruption by 42% while improving rural student retention rates, proving their scalability as foundational components of CHSPs 2.0.
Third, to address the urban-rural education divide, it is necessary to formulate policy interventions promoting vocational-general education convergence. This includes curriculum integration modules that blend STEM disciplines with vocational competencies, akin to Germany’s dual-system model [72], alongside certification equivalency frameworks that recognize vocational diplomas as equivalent to academic qualifications. Mandatory internship quotas—20% for urban students and 50% for rural learners–-would bridge the enterprise-academy divide while addressing rural youth’s employability gaps. These structural reforms require the allocation of 15% of provincial education budgets to incentivize school-enterprise partnerships, ensuring that vocational training programs directly respond to regional labor market demands.
Forth, it is recommended to resolve the deficit in the demand for rural student development reimagining family-school-community synergies. Governments should incentivize rural entrepreneurship through tax breaks and microloan programs to reverse rural depopulation trends, thereby increasing parental presence. Parallel investments in county-level “Family Education Hubs” staffed by trained counselors could deliver targeted parenting programs. These hubs would also serve as tele-education nodes, connecting migrant parents with school psychologists and career advisors via 5G-enabled platforms. By co-locating community resources with schools, CHSPs can transform fragmented rural education landscapes into cohesive ecosystems where academic development, family engagement, and local industry needs mutually reinforce each other.
Finally, it is imperative to expedite the reform of the education management system. While the authority of county-level governments in educational administration should be preserved, the role and influence of municipal governments in school operations should be enhanced.Regarding fiscal support, it is recommended that provincial governments establish a priority list of counties requiring targeted assistance, enabling provincial-level fiscal funds to deliver targeted support to high schools in underprivileged counties through tailored funding mechanisms.
Supporting information
References
1Chen R . Promote the high-quality development of “county-level high school” . J Chin Soc Educ . 2022 ;(02 ):1 .
2Ministry of Education of the People’s Republic of China. Number of regular senior secondary schools and classes[Statistical report]. 2025 . Available from: https://hudong.moe.gov.cn/jyb_sjzl/moe_560/2024/quanguo/202603/t20260317_1431164.html
3Ministry of Education of the People’s Republic of China. Number of students in regular senior secondary education by providers. 2025 . Available from: https://hudong.moe.gov.cn/jyb_sjzl/moe_560/2024/quanguo/202603/t20260317_1431163.html
4Yang H . Strategic logic and certified accountant of high-quality development of county ordinary high schools . J Chin Soc Educ . 2022 ;(04 ):35 –40 .
5Liu Y . Book of major educational events in China . Hangzhou: Hangzhou Education Press; 1993 . pp. 314 .
6Sun Y , Liu Y . The policy orientation, realistic examination and development strategy of the revitalization of county high schools . Hubei Soc Sci . 2024 ;(5 ):126 –33 . doi: 10.13660/j.cnki.42-1112/c.016320
7Liu L , Zhang W . Vitalizing xianzhong: Why is it necessary and how is it possible . J Educ Sci Hunan Norm Univ . 2021 ;20 (6 ):8 –13 . doi: 10.19503/j.cnki.1671-6124.2021.06.002
8Li X , Luo K . The underlying logic and institutional governance of super secondary schools: A reflection and review based on the Hengshui model . Fudan Educ Forum . 2021 ;19 (05 ):13 –20 . doi: 10.13397/j.cnki.fef.2021.05.003
9Lie W . The operation logic and revitalization road of China’s Xianzhong . J Educ Sci Hunan Normal Univ . 2021 ;20 (06 ):14 –21 .
10Zhou X , Yang Y . Governance to the policy-based and cross-regional recruitment of senior high school and revival of the county senior high school . J Educ Stud . 2022 ;18 (05 ):84 –94 . doi: 10.14082/j.cnki.1673-1298.2022.05.008
11Ministry of Education of the People’s Republic of China, eight other departments. “14th Five-Year Plan” for Promotion of County High Schools. 2021 . Available from: https://www.gov.cn/zhengce/zhengceku/2021-12-16/content_5661144.htm
12Central People’s Government of the People’s Republic of China. The outline of the 14th Five-Year Plan for national economic and social development and vision 2035. 2021 . Available from: https://www.gov.cn/xinwen/2021-03/13/content_5592681.htm
13Group C . Outline of plan for building China into a leading country in education (2024–2035) . Beijing: Foreign Language Teaching and Research Press; 2025 . pp. 14 .
14Yang D , Wang S . The dilemma and solution of basic education governance through an analysis of Hengshui high school model . Tsinghua J Educ . 2018 ;39 (04 ):87 –93 . doi: 10.14138/j.1001-4519.2018.04.008707
15Xi Y . Super High School: A catalyst for the inter-school gap between regular high schools . J Chin Soc Educ . 2014 ;(6 ):15 –8 .
16Guo C , Xu Z , Zhang S . Super high schools: Improving or reducing the educational quality of high schools in different provinces? Educ Res . 2021 ;42 (4 ):37 –51 .
17Yang H . Efficiency or fairness: The allocation of education resources in China’s county areas . Seeker . 2024 ;(04 ):171 –80 . doi: 10.16059/j.cnki.cn43-1008/c.2024.04.019
18Wu Z , Wang Q , Liang H . Subjective rationality, institutional regulations, and the rise and fall of county high schools: Based on the perspective of rational choice theory . J East China Normal Univ (Educ Sci) . 2023 ;41 (12 ):1 –12 . doi: 10.19503/j.cnki.1671-6124.2021.06.003
19Wang D , Wang Z , Yin Z . The regional heterogeneity, causes and policy governance path of collapse in county high school . Res Educ Dev . 2023 ;43 (2 ):20 –7 . doi: 10.14121/j.cnki.1008-3855.2023.02.005
20Liu L , Zhang W . Vitalizing Xianzhong: Why is it necessary and how is it possible . J Educ Sci Hunan Normal Univ . 2021 ;20 (6 ):8 –13 . doi: 10.19503/j.cnki.1671-6124.2021.06.002
21Gu J . Spatial dynamics and determinants of county-level education expenditure in China . Asia Pacific Educ Rev . 2012 ;13 (4 ):617 –34 . doi: 10.1007/s12564-012-9224-y
22Xiang M , Li Q , Wang A . Mapping the structure and adjustment of policy attention for the revitalization of county-level ordinary high school: A textual analysis of 26 provincial-level “county middle school revitalization plans” . J Shanghai Educ Res . 2024 ;(7 ):23 –9 . doi: 10.16194/j.cnki.31-1059/g4.2024.07.006
23Wei H , Zhong Z . Difficulties in the implementation of the “Revitalization of CountryLevel Senior High Schools” plan and the path of solving the problems . Contemp Educ Sci . 2023 ;(12 ):62 –72 .
24Deng B . What kind of policy structures more conducive to the revitalization of county-level high schools . J Educ Sci Hunan Normal Univ . 2025 ;24 (2 ):13 –21 .
25Xiao Z . The theoretical implication and practical logic of cultural ecological reconstruction in the county senior high school during the construction of an educational powerhouse . J Distance Educ . 2024 ;42 (2 ):3 –9 . doi: 10.15881/j.cnki.cn33-1304/g4.2024.02.001
26Zhang Z . The revitalization path of county senior high school education from the perspective of national educational governance . J Educ Stud . 2022 ;18 (5 ):72 –83 . doi: 10.14082/j.cnki.1673-1298.2022.05.007
27Laver M , Benoit K , Garry J . Extracting policy positions from political texts using words as data . Am Pol Sci Rev . 2003 ;97 (02 ). doi: 10.1017/s0003055403000698
28Estrada M , Nagaraj Y . Beyond the ceteris paribus assumption: Modeling demand and supply assuming omnia mobilis . Int J Econ Res . 2007 ;5 (2 ):185 –94 .
29Ruiz Estrada MA . Policy modeling: Definition, classification and evaluation . J Policy Model . 2011 ;33 (4 ):523 –36 . doi: 10.1016/j.jpolmod.2011.02.003
30Hong S , Wang T , Fu X , Li G . Research on quantitative evaluation of digital economy policy in China based on the PMC index model . PLoS One . 2024 ;19 (2 ):e0298312. doi: 10.1371/journal.pone.0298312 38359065 PMC10868804
31Yang C , Yin S , Cui D , Mao Z , Sun Y , Jia C , et al. Quantitative evaluation of traditional Chinese medicine development policy: A PMC index model approach . Front Public Health . 2022 ;10 :1041528 . doi: 10.3389/fpubh.2022.1041528 36466538 PMC9715970
32Liu Y , Li J , Xu Y . Quantitative evaluation of high-tech industry policies based on the PMC-index model: A case study of China’s Beijing-Tianjin-Hebei region . Sustainability . 2022 ;14 (15 ):1 –17 . doi: 10.3390/su14159338
33Yang C . Research on China’s adolescent mental health policies - analysis based on PMC index model . Front Public Health . 2024 ;12 :1408991 . doi: 10.3389/fpubh.2024.1408991 39247227 PMC11377327
34Lu C , Wang B , Chen T , Yang J . A document analysis of peak carbon emissions and carbon neutrality policies based on a PMC index model in China . Int J Environ Res Public Health . 2022 ;19 (15 ):9312 . doi: 10.3390/ijerph19159312 35954666 PMC9368600
35Ma X , Ruan Y , Yang Q . Evaluating China’s common prosperity policies against the background of green development by using the PMC model . Sustainability . 2023 ;15 (10 ):7870 . doi: 10.3390/su15107870
36Xiong Y , Zhang C , Qi H . How effective is the fire safety education policy in China? A quantitative evaluation based on the PMC-index model . Safety Sci . 2023 ;161 :106070 . doi: 10.1016/j.ssci.2023.106070
37Xu L , Zhou Y . Evaluation and optimization of compulsory education policy based on PMC-AE index model . Front Educ Res . 2023 ;6 (24 ):1 –10 . doi: 10.25236/FER.2023.062404
38Ma D , Lv H . Quantitative evaluation of graduate education policy for professional degree in China: Based on PMC-index model . Commun Educ Rev . 2025 ;6 (1 ):1 –18 . doi: 10.37420/j.cer.2025.1036
39Hu M , Guo C , Wang Y , Ma D . Quantitative evaluation of China’s private universities provincial public funding policies based on the PMC-Index model . PLoS One . 2023 ;18 (12 ):e0295601. doi: 10.1371/journal.pone.0295601 38085719 PMC10715659
40Cui C , Wang K . Quantitatively analyzing the college student employment policy in China based on PMC-index model . PLOS ONE . 2024 ;19 (10 ):e0310479. doi: 10.1371/journal.pone.0310479 PMC11500955 39446751
41Jin L , Zeng J . Quantitative research on policy text: Concepts, methods, and research perspectives . Digit Libr Forum . 2024 ;20 (04 ):12 –21 . doi: 10.3772/j.issn.1673-2286.2024.04.002
42Fang S , Liu Q . A review on application of PMC index model in policy documents quantitative research . J Modern Inform . 2024 ;44 (4 ):168 –76 . doi: 10.3969/j.issn.1008-0821.2024.04.015
43Glaser B , Strauss A . The discovery of grounded theory: strategies for qualitative research . Chicago: Aldine Publishing Company; 1967 .
44McDonnell LM , Elmore RF . Getting the job done: Alternative policy instruments . Educ Eval Policy Analysis . 1987 ;9 (2 ):133 –52 . doi: 10.3102/01623737009002133
45Schneider A , Ingram H . Behavioral assumptions of policy tools . J Polit . 1990 ;52 (2 ):510 –29 . doi: 10.2307/2131904
46Kuang B , Han J , Lu X , Zhang X , Fan X . Quantitative evaluation of China’s cultivated land protection policies based on the PMC-Index model . Land Use Policy . 2020 ;99 :105062 . doi: 10.1016/j.landusepol.2020.105062
47Qian X , Zhu J , Huang B . Policy evaluation of digital transformation of elementary education during the “14th five year plan” based on PMC index model: Taking Jiangsu, Zhejiang and Shanghai regions as examples . J Distance Educ . 2024 ;42 (01 ):50 –60 . doi: 10.15881/j.cnki.cn33-1304/g4.2024.01.007
48Chen X , Zhen X , Jin J . The deep mechanism of the change of county high schools since China’s reform and opening up: An explanation based on resource dependency theory . Res Educ Dev . 2023 ;43 :11 –9 . doi: 10.14121/j.cnki.1008-3855.2023.02.009
49Zhou Y , Qu Z , Yang S . The reform course of online enrollment in colleges and universities in China . Contemp China History Stud . 2019 ;26 (04 ):52–62, 157–8.
50General Office of the Ministry of Education of the People’s Republic of China. Circular on further improving the enrollment work of general primary and secondary schools (Jiaojiting [2022] No.1). 2022 . Available from: https://www.moe.gov.cn/srcsite/A06/s3321/202204/t20220401_612689.html
51Liu L , Peng L . The unification in general and vocational education: the overall trend of senior high school education reform in China . J Educ Sci Hunan Normal Univ . 2013 ;12 (05 ):64 –8 .
52Cheng X . The formation logic and future direction of the positioning of comprehensive high schools in China: The wrestling among multiple institutional logic . Chin Vocat Tech Educ . 2024 ;29 :3 –13 , 22.
53Chang B . Realistic dilemma and practical path of high-quality development of county senior high school . J Chinese Soc Educ . 2024 ;(08 ):68 –73 .
54Fan Y , Tian J . The specialization logic in the revitalization of county regular senior secondary schools . Res Educ Dev . 2023 ;43 (24 ):41 –50 . doi: 10.14121/j.cnki.1008-3855.2023.24.004
55Ministry of Education of the People’s Republic of China, Ministry of Public Security, Ministry of Justice, Ministry of Housing and Urban-Rural Development, Ministry of Transport, Ministry of Culture and Tourism, et al. Measures for the administration of safety in primary and secondary schools and kindergartens (Order No. 23 of the Ministry of Education); 2006 . Available from: https://www.moe.gov.cn/jyb_xxgk/xxgk/zhengce/guizhang/202112/t20211206_585036.html
56Ministry of Education of the People’s Republic of China. Measures for the handling of student injury accidents (amended on Dec 13, 2010). 2002 . Available from: http://www.moe.gov.cn/jyb_xxgk/xxgk/zhengce/guizhang/202112/t20211206_585115.html
57State Council Education Steering Committee Office of the People’s Republic of China. Notice on launching the implementation year campaign for prevention and control of school bullying among primary and secondary school students (Guojiaoduban Han [2018] No.28). 2018 . Available from: http://www.moe.gov.cn/srcsite/A11/moe_1789/201804/t20180428_334588.html
58General Office of the Ministry of Education of the People’s Republic of China. Notice on issuing the “Ten Musts” for campus food safety and the “Ten Prohibitions” for school canteen staff (Jiaotiyitinghan [2025] No.12). 2026 . Available from: http://www.moe.gov.cn/srcsite/A17/moe_943/s3283/202603/t20260324_1431938.html
59General Office of the Ministry of Education of the People’s Republic of China. Notice on issuing the “Ten Measures for Further Strengthening Mental Health Work among Primary and Secondary School Students” (Jiaojiting [2025] No.2). 2025 . Available from: https://www.moe.gov.cn/srcsite/A06/s3325/202510/t20251020_1417420.html
60Ministry of Education of the People’s Republic of China. Guiding opinions on comprehensively promoting the construction of healthy schools (Jiaojiyi [2026] No.3). 2026 . Available from: https://www.moe.gov.cn/srcsite/A17/moe_943/moe_946/202602/t20260227_1429365.html
61Wang C . Evolution of attention in school safety policy in China: based on the content analysis of the 35-year “Working Points of the Ministry of Education” (1987–2021) . J Guangzhou Univ (Soc Sci Edition) . 2022 ;21 (02 ):18 –31 .
62Li X . Swotting, drifting or quitting: The learning practices of rural students at county-seat high schools against the backdrop of gaokao reform . Fudan Educ Forum . 2024 ;:40 –9 . doi: 10.13397/j.cnki.fef.2024.01.011
63Li X . Gaokao reform and routine pedagogy at a rural senior high school: Understanding the poor university access of rural students . Glob Educ . 2020 ;49 :75 –89 .
64Tang H , Luo C , Wang Y , Chen Z , Wu J . Relation between bad interpersonal relationships and psychological capital of high school students: The mediating effect of perceived social support and alexithymia and the moderation effect of urban-rural disparity . China J Health Psychol . 2021 ;:1065 –71 . doi: 10.13342/j.cnki.cjhp.2022.07.022
65Li Z , Yang Z . Higher education choice of secondary vocational school students and its influencing factors: Based on the survey of 10660 secondary vocational school students in China . Fudan Educ Forum . 2023 ;21 (1 ):44 –53 . doi: 10.13397/j.cnki.fef.2023.01.004
66Ministry of Education of the People’s Republic of China, Publicity Department of the Central Committee of the Communist Party of China, Cyberspace Administration of China, Central Guidance Commission for Building Spiritual Civilization, Ministry of Public Security, Ministry of Civil Affairs, et al. Opinions on improving the coordinated education mechanism of schools, families and society (Jiaoji [2022] No.7). 2023 . Available from: http://www.moe.gov.cn/srcsite/A06/s3325/202301/t20230119_1039746.html
67Zhou X . Paradigm internal development and solution to the collapse of county-level high schools . J Chin Soc Educ . 2022 ;(2 ):15 –21 .
68Miao X , Li L , Liu X . City-county integrated high-level governance to support the revitalization of county high schools: reform directions and pathways of realization . J Basic Educ . 2025 ;22 (06 ):5 –15 .
69Zhou Y , Han R . The integrated development of urban and rural education in counties calls for management system reform . Educ Res Exp . 2025 ;(04 ):139 –49 .
70Zhou Y , Jing C , Liang Y . The integrated development of county-level rural-urban education: Conceptual and theoretical frameworks . Educ Res . 2025 ;46 (01 ):30 –42 .
71Miao X , Zhang J , Wang K . Does Province-Managing-County Fiscal Reform Narrow the Basic Public Service Gaps Between Cities and Counties? Research on Financial and Economic Issues . 2024 ; pp. 48 –59 . doi: 10.19654/j.cnki.cjwtyj.2024.07.004
72Deissinger T . The standing of dual apprenticeships in Germany: Institutional stability and current challenges . The standing of vocational education and the occupations it serves: Current concerns and strategies for enhancing that standing . Cham: Springer International Publishing; 2022 . pp. 83 –101 .
Come citare questo studio
Wu Q, Li G. (2026). A quantitative assessment of Chinese county high school policies using the PMC index model.. PloS one. DOI: 10.1371/journal.pone.0361807
Consulta la pubblicazione originale ↗