Next Article in Journal
An Interpretive Structural Modeling Approach for Biomedical Innovation Strategy Models with Sustainability
Previous Article in Journal
Hydrothermally Treated Cement Bypass Dust as a Supplementary Cementitious Material
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Text-Mining-Based Evaluation of Data Element Policies in China: Integrating the LDA and PMC Models in the Context of Green Development

School of Public Affairs, Zhejiang University, Hangzhou 310030, China
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(15), 6758; https://doi.org/10.3390/su17156758
Submission received: 10 June 2025 / Revised: 15 July 2025 / Accepted: 22 July 2025 / Published: 24 July 2025

Abstract

In the context of green development, promoting the development of data elements is crucial for advancing the green and low-carbon transition and achieving China’s “dual-carbon” targets. This study quantitatively evaluates China’s data element policies to identify their strengths and weaknesses and to assess their alignment with green development objectives. In this study, we examine 15 representative data element policy texts, evaluating their quality by integrating the Latent Dirichlet Allocation (LDA) topic model with the PMC-Index model. The LDA analysis identifies five core themes within the policy texts: the data element industry, data resource management, data element trading systems, service platform construction, and e-governments. The evaluation results show an average PMC-Index score of 6.03 for the 15 policies, with 9 rated as “Good” and 6 as “Acceptable”. This indicates that while the overall design of the current policy system is acceptable, there remains substantial room for improvement. Based on the average scores for the primary indicators, the policies perform relatively poorly in terms of green development assessment, policy timeliness, policy nature, and policy guarantee. Drawing from these findings, we propose recommendations to enhance China’s data element policies, offering insights for policymakers.

1. Introduction

Following the Paris Agreement (COP21) in 2015 and the subsequent call from COP25 for global carbon neutrality by 2050, the need to promote a green transformation of the economy and society has become a global consensus [1,2]. Concurrently, the rise of Industry 4.0 technologies is ushering global economic governance into a new era of digital transformation [3,4]. Against this backdrop, digitalization and greening have emerged as pivotal themes in contemporary global socioeconomic development. As the world’s largest developing country and internet consumer market, China has ascended to become the second-largest digital economy globally. In 2023, China’s digital economy reached CNY 53.9 trillion, accounting for 42.8% of its GDP [5]. However, China is also the world’s largest energy consumer and carbon emitter. In 2022, its energy consumption accounted for 26.5% of the global total [6], and its carbon emissions comprised 31% of the global total in 2024 [7]. Facing intense pressure on growth, resources, and the environment, the intensive development and green transformation of the economy and society have become national strategic priorities. In 2020, at the United Nations General Assembly, China announced its “dual-carbon” strategic goals: striving to peak carbon emissions by 2030 and achieve carbon neutrality by 2060. In October 2021, the State Council issued the “Action Plan for Carbon Dioxide Peaking Before 2030”, which explicitly advocates for the industrial sector to prioritize the integration of digital and smart transformation with traditional industries to achieve the “dual-carbon” goals.
Green development necessitates economic growth that does not come at the expense of environmental degradation [8]. The energy consumption and pollution inherent in traditional development models are incompatible with the core tenets of green development. As an emerging economic form, the digital economy demonstrates immense potential for sustainable development by optimizing resource allocation patterns and enhancing total factor productivity, particularly through significant improvements in energy utilization efficiency [9,10]. Therefore, the digital economy based on data elements has become a key driver for reducing carbon emissions and realizing green development [11].
Data elements, the key factors of production in the digital economy [12], exhibit environmentally friendly characteristics, such as low energy consumption and minimal pollution during their acquisition and circulation, due to their inherent virtuality and intangibility [13]. Moreover, with features including non-rivalry, non-excludability, and increasing marginal returns, data elements provide substantial additional revenue at a marginal cost of nearly zero, which endows them with the distinct property of delivering high returns from low investment [14]. The integration of data elements into economic activities, therefore, effectively reduces the energy consumption and pollution characteristic of traditional development models, fully embodying the principles of green and sustainable development [15]. It is thus evident that cultivating and expanding data elements is not only an effective pathway to achieving economic growth but also an important instrument for realizing green and low-carbon development.
In this context, China, as the first country to officially recognize data as a factor of production, has introduced a series of policies and plans to promote the application and development of data elements, encouraging stakeholders, including government and enterprises, to leverage data elements for green transformation and environmental governance [16]. Guided by these policies, data elements are being extensively applied in green value-creation fields such as new energy big data, green supply chains, and carbon footprint management [17]. However, in previous research, there has been little evaluation of data element policy texts from the perspective of green development. This raises questions about the rationality of the relevant policies formulated by the government. Are these policies effective? What are their deficiencies? Is a green development orientation reflected in their formulation and implementation? All of these questions need to be answered through an objective and scientific quantitative evaluation of the policies. To measure the effectiveness of China’s data element policy texts and the balance of the policy system, and to improve existing data element policies as well as the formulation of future ones, it is necessary to conduct a multi-dimensional quantitative evaluation of the existing data element policies.
This paper focuses on the following questions: (1) What are the strengths and weaknesses of China’s data element policies? (2) How consistent are China’s data element policies with green development objectives? To address these questions, this study employs the LDA-PMC model as a reliable quantitative framework for evaluating the effectiveness and consistency of policy texts. Our analysis is based on a corpus of 55 policy texts sourced from authoritative Chinese databases. From this corpus, a representative sample of 15 policies was selected for an in-depth assessment of their respective strengths and weaknesses, as well as their alignment with green development objectives. Through this evaluation, policymakers can obtain more precise decision-making support, promoting the effective implementation of data element policies, fostering the development of the digital economy, and contributing to the achievement of the nation’s “dual-carbon” goals.
The structure of this paper is as follows: Section 2 reviews the relevant literature. Section 3 describes the data sources and the design of the LDA-PMC model. Section 4 presents and analyses the empirical results. Section 5 discusses and concludes the study, offers policy recommendations, and outlines directions for future research.

2. Literature Review

2.1. Research on Data Elements in the Context of Green Development

The concept of green development is centered on environmental protection [18]. As both a new philosophy for social development and a new model for economic growth, green development stems from a profound reconsideration of traditional growth models. Its core objective is to overcome the constraints of the natural environment and resolve ecological problems [19]. With the advent of the Industry 4.0 era, digital technologies have begun to play a central role in most strategies for combating climate change [20]. Specifically, driven by digital technology, the digital economy offers opportunities to achieve a form of green development that emphasizes the coordination of economic growth and environmental protection [21].
In this context, the impact of data, as a key factor of production in the digital economy, on green development has become a critical focus of academic inquiry. A substantial body of research has explored the positive role of data elements in the domain of green development. One line of research argues that data elements promote green development by optimizing production methods. Data elements can help enterprises reduce information asymmetry, enhance market insight, and optimize decision-making processes, thereby supporting the intelligent upgrading of traditional industries [22]. For businesses, the application of data elements can effectively promote green innovation by increasing investment in innovation and strengthening external governance factors [23]. Another stream of research contends that data elements influence green development by optimizing the structure of energy consumption. Data elements can reduce dependence on traditional fossil fuels and promote a transition in the energy structure towards low-carbon and intelligent systems [24]. L. Wu et al. proposed that the development of big data can improve energy efficiency by optimizing urban energy structures, thus contributing to energy conservation and emission reduction [25]. A third area of research suggests that the application of data elements provides technical support and intelligent assurance for promoting green development [26], which is conducive to improving the level of ecological monitoring, reducing ecological risks, and increasing the efficiency of environmental governance [19]. Furthermore, government open data have a significant positive effect on the growth of the green economy [27]. The open sharing of public data by governments provides a valuable resource base for corporate green innovation [28,29]. Studies have shown that the open sharing of public data can significantly reduce pollutant emissions at the enterprise level through channels such as promoting corporate digital transformation and lowering operational costs [30,31].
However, some scholars argue that data elements may also hinder green development. The primary reason for this is the energy rebound effect [32]. Kunkel and Tyfield assert that carbon emissions are generated during the production, installation, distribution, and upgrading of digital infrastructure, leading to adverse environmental consequences [33]. Concurrently, the construction and operation of digital infrastructure, such as data centers, consume vast amounts of electricity [34], which, given the current energy structure, leads to a significant increase in carbon emissions [35,36]. Wu and Li, using panel data from 166 countries/regions worldwide, found a “U-shaped” relationship between data elements and the low-carbon development of the manufacturing sector, indicating that the initial application of data elements results in negative environmental externalities [37]. Second, the use of large global data centers and mobile data traffic can generate electronic waste related to manufacturing [38], thereby causing environmental problems [39,40]. It is thus evident that the net environmental effect of data elements is a matter of debate. The reason for this is that the environmental benefits brought about by the development of data elements may only partially offset the negative environmental costs they generate [20]. This conflict in the underlying mechanisms highlights the importance of public policy design, which, through scientifically sound and rational guidance, must ensure that the development of data elements ultimately leads to positive green development outcomes.

2.2. Research on Data Element Policies

Data element policy is defined as the strategies, plans, policies, and pilot projects formulated by governments to ensure the effective promotion and application of data elements [41]. To foster the development of the digital economy, numerous countries have introduced data element policies that encompass national data strategies and plans, open data policies, data sharing policies, and data security and privacy protection policies [42,43,44]. The European Union prioritizes a rules-based approach, emphasizing ethical norms in data governance [45] and defining a macro-level framework for data governance, particularly concerning personal data [46]. In contrast, the policy system in the United States is characterized by fragmentation, being industry-specific yet generally business-friendly [46]. In 2019, the U.S. Office of Management and Budget and the Department of Commerce jointly released the Federal Data Strategy and 2020 Action Plan, which established data as a strategic resource and shifted the focus from the technical level to its capital value [47]. China’s policy framework, in contrast, emphasizes a high degree of government involvement [48] and utilizes its social credit system as a disciplinary mechanism [49].
Scholars have undertaken numerous studies in areas such as framework analysis, comparative analysis, implementation analysis, and textual analysis of data element policies. One line of research focuses on policy framework analysis. Bertot et al. analyzed the challenges within the U.S. information policy framework concerning big data access and dissemination, digital asset management, archival preservation, and privacy security, and they proposed corresponding revisions to the policy framework [50]. A second line of research examines the collaborative aspects of policy implementation. Osifo identified challenges in the collaborative implementation of Finland’s digital policy through interviews and the content analysis of policy documents [51]. A third area of study concentrates on comparative policy analysis. Using a modified evidence-based approach, J. Zhao et al. (2025) conducted a comparative study of the traditional data rights determination policies in the European Union, the United States, and China, providing new perspectives and solutions for establishing a framework for data rights in the circulation of data elements [52]. A fourth stream of research focuses on policy text analysis. Based on an analysis of big data policy documents, Mahrenbach et al. identified potential areas for cooperation in the big data field among emerging “Southern” countries, including Brazil, India, and China [53]. Using semantic network analysis, Jung and Park (2015) identified the core issues and key dimensions of South Korea’s open public data policy, offering guidance for its improvement [54].
The formulation and reform of data element policies face a multitude of challenges. These include a range of technical challenges, such as the determination of data rights and data security protection [55,56], a lack of awareness among stakeholders [51], difficulties in effectively integrating and utilizing limited resources [57], and obstacles at the operational level [58]. To address these issues, the academic community has focused on research and policy recommendations from various perspectives, including clarifying the rights holders and their corresponding rights in the circulation of data elements [52], building partnerships and collaborative networks [51], prioritizing policy support measures [57], and establishing specialized implementation agencies [58].

2.3. Research on Public Policy Evaluation

Public policy evaluation is key to understanding whether public policies are effective and how they can be improved [59], and it provides policymakers with an evidence-based decision-making framework [60,61]. As an analytical tool, public policy evaluation research typically involves investigating a policy program to obtain all information relevant to its performance assessment [62], thereby providing a scientific basis for further policy optimization. Currently, research focused on evaluating data element policies generally falls into two categories. The first category primarily involves the qualitative evaluation of data element policies, employing evaluation frameworks to determine the effects and benefits after policy implementation. For example, Nugroho et al. (2015) established a cross-national comparative framework for open data policies and compared the effectiveness of such policies in the United Kingdom, the United States, the Netherlands, Kenya, and Indonesia [63]. The second category of literature concentrates on the quantitative evaluation of the implementation effects of individual data element policies, mainly using econometric methods to explore the effectiveness or impact of a single data element policy. Hu et al. used the difference-in-differences (DID) method to study the impact of the National Big Data Comprehensive Pilot Zone policy on the digital economy, finding that its introduction increased the level of digital economy development by 2.8% [41]. M. Zhou et al. assessed the effectiveness of the government’s open data policy using panel data from 477 listed companies and found that the OGD policy had a positive impact on firm performance [64].
From the perspective of evaluation stages, policy evaluation can be divided into ex ante, in-process, and ex post evaluation [65]. Among these, ex ante evaluation is an important component of the policy cycle, providing evidence and rigor to the policy process [66]. It is evident that existing evaluations of data element policies are predominantly concentrated on ex post assessments, which focus on the effects or benefits realized after policy implementation, while research on the ex ante evaluation of these policies is relatively scarce. Although ex post evaluation is helpful for understanding the actual impact after implementation, ex ante evaluation can more effectively predict the potential effects and impacts before a policy is enacted [67]. The main ex ante policy evaluation methods currently include Cost–Benefit Analysis (CBA), Multi-Criteria Decision Analysis (MCDA), the Logic Model, and the PMC-Index Model. Cost–Benefit Analysis quantifies and compares the “benefits” and “costs” of a policy in monetary terms [68], providing informational support for the decision-making process [69]. For example, Tol (2012) used CBA to evaluate the EU’s 2020 climate package, pointing out that future policies would need to adopt a lower discount rate to support the 2020 emission reduction targets [70]. Multi-Criteria Decision Analysis is another widely used ex ante evaluation method, and its advantage lies in assisting decision makers in understanding the core criteria of a decision problem and in ranking alternative options accordingly [71]. Browne et al. (2010) used MCDA to evaluate six policy measures related to urban residential heating energy and household electricity consumption in Ireland and identified the most desirable policy measure among them [72]. The Logic Model provides a clear, systematic framework for policy evaluation and is used to identify and measure a project’s inputs, activities, outputs, and outcomes [73]. The use of the Logic Model enables forward-looking and theory-based planning in evaluation analysis and can increase the transparency of policy evaluation [74]. For instance, Petticrew et al. developed a detailed logic model for England’s Public Health Responsibility Deal to articulate its intended outcomes and pathways, laying the groundwork for evaluation planning [75]. In 2010, scholars began to adopt the PMC-Index model proposed by Ruiz Estrada [76], which provides a quantitative framework for the ex ante evaluation of public policy. Compared with other quantitative evaluation methods, the PMC-Index model has the following advantages. First, it is more suitable for ex ante evaluation, as it can effectively assess policies that have not yet produced quantifiable results. In contrast, the CBA method, which requires the precise quantification of costs and benefits in monetary terms, faces predictive uncertainty in the ex ante stage and may, therefore, be more applicable to the evaluation of fully implemented policies. Second, its evaluation is based on existing policy texts. Unlike quantitative models such as Cost–Benefit Analysis (CBA) that rely heavily on numerical inputs, the PMC model achieves greater flexibility by evaluating the policy itself through the integration of qualitative content from policy documents and legal provisions. Third, its evaluation system offers better modularity. The PMC model can be viewed as a universal policy evaluation framework that can be customized for different policy areas while maintaining a unified analytical structure. This feature allows for the development of evaluation frameworks for specific policy domains. Currently, the PMC-Index model has been widely applied in fields such as industrial policy [77], budget management policy [78], land use policy [79], and ecological protection policy [80].
This study applies the PMC-Index model to evaluate China’s data element policies for several reasons. First, the model combines various influencing factors, enabling a comprehensive assessment of policy effectiveness and reducing the risk of evaluation bias. Second, the model determines policy strengths, weaknesses, and internal consistency based on textual content, allowing for the inclusion of more qualitative content in a quantitative evaluation, which facilitates a deeper analysis of the unique characteristics and problems of a policy. Third, the model features flexible evaluation modules, which enabled us to include an assessment of the policy’s green development measures when constructing the evaluation indicator system for data element policies, thereby allowing us to measure the consistency of these policies with green development objectives.
The standard PMC-Index system typically relies on the results of high-frequency words obtained from the literature and text mining as a reference, which can lead to the problem of missing variables in the construction of the evaluation index system [81]. The LDA topic model addresses the potential for omission or oversight that can occur when policy themes are manually induced from high-frequency words by identifying stable and repetitive co-occurrence patterns among words to discover latent themes [82]. Therefore, scholars have improved the PMC-Index model by developing the LDA-PMC model [77]. Compared with other modified PMC models, such as the PMC-AE model, the utility of the LDA-PMC model lies in its ability to supplement variables that may be missing during the construction of the standard PMC-Index system, thereby yielding a more comprehensive PMC indicator evaluation system [81]. The PMC-AE model, in contrast to the traditional PMC, utilizes an autoencoder to generate the PMC-Index score, thus reducing the subjectivity of the index calculation [83]. However, the calculation process of autoencoder technology involves a “black box” problem [84], and the resulting scores may lack interpretability. For this reason, this study adopts the LDA-PMC model to quantitatively evaluate China’s data element policy texts. Specifically, we utilized the LDA topic model to conduct an in-depth analysis of data element policy texts and automatically identified the core themes within them. Subsequently, within the context of green development and based on the results of word frequency and LDA analysis, this study constructed a PMC evaluation indicator system for data element policies. This system was used to identify the strengths and weaknesses of the policies and to assess their consistency with green development goals, providing a more insightful scientific basis for subsequent policy optimization.

3. Research Design

3.1. Data Sources

This study utilizes policy texts concerning data elements as its analytical dataset. The policy texts were primarily sourced from the “PKULAW” database (www.pkulaw.com) and the official websites of China’s provincial governments, retrieved using “data element” as the keyword. The retrieved documents were screened according to the following criteria: (1) duplicate, expired, and informal documents, such as official replies and public notices, were removed. For policies with multiple versions, only the most recent version was retained. (2) The content of the texts was required to be highly relevant to data elements; documents with minimal content on this topic were excluded. After filtering out less relevant texts, a final corpus of 55 valid policy documents was established. The publication dates of the documents in the dataset range from April 2020 to November 2024. The trend of text issuance is shown in Figure 1.

3.2. Word Frequency Analysis of Policy Texts

Word frequency analysis directly reveals the core issues and strategic focus of a policy, offering insights into the internal logic and value orientation of the policymaking process. It helps to identify the primary tools and implementation mechanisms upon which the policy relies. High-frequency words extracted from policy texts can assist in assessing the consistency and comprehensiveness of policy content, providing an initial quantitative examination of the policy documents that lays a solid foundation for subsequent, more in-depth evaluation. This study used Python 3.12.6 to conduct keyword mining on the policy texts, from which representative high-frequency words were extracted. After word segmentation and the removal of stop words, high-frequency terms with no significant impact on policy analysis, such as “data”, “digital”, “promote”, and “strengthen”, were manually deleted. High-frequency words with similar meanings were merged and adjusted, resulting in a final list of 30 valid high-frequency words for data element policies. The specific results are presented in Table 1.
By observing the frequency distribution of these high-frequency words in conjunction with the policy content, several patterns emerge. Words such as “platform”, “infrastructure”, “data center”, “network”, “information”, and “internet” indicate that China’s data element policies prioritize the construction of robust data infrastructure and service platforms. The prevalence of “supervision”, “governance”, and “data security” reflects that policymakers consider data security a fundamental prerequisite for the application and development of data elements. Terms such as “innovation” and “digitalization” show that policies place technological advancement and industrial digital transformation at the core of their strategy. The frequency of “integration” and “collaboration” suggests that policies encourage the integrated application of data across different domains and entities. Words such as “sharing”, “openness”, “data resources”, “public data”, and “big data” highlight a policy emphasis on the supply and management of public data and big data resources, with a focus on enhancing data accessibility. The presence of “cultivation”, “industry”, “sector”, “industrial”, “product”, and “scenario” demonstrates a policy commitment to developing the data element industry, promoting the application of data in specific sectors, especially the industrial domain, and creating more diverse data products based on industry needs. Terms such as “service” and “government affairs” reveal an intention to use data elements to improve public service levels and the efficacy of government administration. Words such as “transaction”, “model”, and “system” reflect active policy exploration towards establishing effective market mechanisms and sustainable business models for data elements. Notably, the appearance of “ecology” indicates that while promoting the application of data elements, China’s policies also stress their harmonious development with the environment, seeking to empower a resource-conserving and environmentally friendly development model through data, thereby supporting the nation’s “dual-carbon” goals and green transformation.

3.3. Identification of Core Policy Themes

This study employs the Latent Dirichlet Allocation (LDA) topic model to further analyze the core themes addressed in the policies. The LDA model, proposed by Blei, Ng, and Jordan in 2003, is a generative probabilistic model that can automatically discover latent topics within a collection of documents [62]. Compared with the manual induction of policy themes from high-frequency word lists, the LDA topic model can eliminate subjective bias and the influence of individual researchers’ cognitive frameworks, thereby facilitating a more objective and comprehensive identification of latent core themes within the policy texts.
We determined the optimal number of topics based on the lowest point of the perplexity curve [85]. Based on the lowest point of the LDA perplexity curve (as shown in Figure 2), we determined the optimal number of topics for the policy texts to be five. The five identified topics and the top ten characteristic words for each topic are presented in Table 2.
The LDA topic model identified five core themes that are the primary focus of China’s data element policies. The first is the data element industry. The characteristic keywords for this theme include “industrial park”, “e-commerce”, and “terminal”. The distribution of keywords within this theme explains two interconnected policy orientations. One is the industrialization of data, which involves creating new, data-centric business models around activities such as data collection, analysis, and application, thereby developing a corresponding industrial supply chain for data elements. The other is the datafication of industry, which emphasizes the deep integration of data elements with the physical economy, particularly by promoting their application in traditional sectors such as manufacturing and agriculture to enhance the digitalization and intelligence of enterprises in these industries.
The second theme is data resource management. Its characteristic keywords include “data processing”, “collection”, “acquisition”, and “public data sharing”. This policy theme has two primary focuses, which are data resource supply and the regulatory framework for data resources. The former concentrates on how to systematically open up access to various types of data generated by public sector entities in the course of their duties or service provision through rational institutional design, thereby expanding the supply of public data resources. The latter focuses on refining the institutional framework governing data access and utilization by formulating regulations for activities such as data authorization, security, and rights to ensure compliance and efficiency in data circulation and use.
The third theme is the data element trading system. Key terms for this theme include “profit distribution”, “data property rights”, and “data authorization operation”. The data element trading system is a central theme in China’s data element policy. This theme comprises three main components, which are a data property rights system, a data trading and circulation system, and a data revenue distribution system. The data property rights system focuses on the clarification and separation of data property rights, particularly on achieving the separation of the “three rights”, which are data resource holding rights, data processing and use rights, and data product operation rights, within a framework that includes individual, corporate, and government data property rights. The data trading and circulation system addresses the advancement of standards for data circulation and access and the regulation of data asset valuation and pricing, which are core components of data element trading. The data revenue distribution system primarily focuses on how to rationally distribute the revenues from the data transaction process to various stakeholders, such as the government, suppliers, demanders, platforms, and data vendors.
The fourth theme is service platform construction. The main keywords for this theme include “online services” and “handle affairs”, reflecting the Chinese government’s policy approach of improving the service environment for market entities by building market service platforms, thereby promoting the application and development of data elements.
The fifth theme is e-government, for which the main characteristic keywords are “administration” and “approval”. This theme focuses on leveraging data elements to reform internal government processes. It emphasizes breaking down departmental silos and reshaping administrative and approval procedures through data sharing and operational synergy to enhance the efficiency of government departments and, in turn, achieve governmental digital transformation.

3.4. Construction of the PMC-Index Model

This study employs the PMC-Index model for the quantitative evaluation of data element policy texts. The PMC-Index is a text-mining-based policy evaluation model primarily used to measure policy consistency and effectiveness. Proposed by Ruiz Estrada, the model is founded on the core assumption of “Omnia Mobilis” [76]. This principle posits that when evaluating a policy, no variable should be analyzed in isolation, as all factors are interconnected and in a constant state of flux. The model utilizes a two-level variable structure: a maximum of ten primary (first-level) variables and an unlimited number of secondary variables, with all variables being weighted equally. This design aims to capture internal policy heterogeneity, thereby enabling a more accurate assessment of a policy’s strengths and weaknesses. The construction of the PMC-Index model in this study involves the following steps (as illustrated in Figure 3): (1) defining policy variables; (2) building a multi-input–output table; (3) calculating the PMC-Index score; (4) generating the PMC surface.

3.4.1. Selection of Representative Policies

The PMC-Index model is designed for the objective examination of all secondary variables and imposes no specific requirements on the unit of analysis. Although the model is capable of quantitatively evaluating any data element policy, the selection of the policy sample should minimize subjective bias [86]. To ensure a scientifically sound and representative sample, this study employed a stratified random sampling method to select representative policy texts. The specific sampling process involved the following three steps.
First, we stratified the nation’s provincial-level administrative divisions into four major economic regions based on China’s official framework for regional economic development: the Eastern, Central, Western, and Northeastern regions. These four regions exhibit significant disparities in economic scale, industrial structure, and the level of development of the digital economy, thereby effectively reflecting the diversity of policy practices.
Second, sample quotas were established based on the number of provincial-level administrative units within each region. We selected 5 from the Eastern region, 3 from the Central, 5 from the Western, and 2 from the Northeastern region, ensuring that the sampling proportion was approximately 50% of the provincial-level administrative units in each region (this evaluation involves 31 provincial-level administrative units). Based on these quotas, we selected 15 representative provincial-level administrative units. The representativeness of the selected provincial-level administrative units is demonstrated in several ways. First, the number of selected units constitutes approximately 50% of the total, ensuring comprehensive coverage. Second, the sample reflects a rational gradient of data element development levels, including leading regions such as Shanghai, Guangdong, and Zhejiang, regions at an intermediate level such as Henan and Jiangxi, and less-developed regions such as Xinjiang and Ningxia. Third, the sample includes three types of administrative units, which are municipalities, standard provinces, and autonomous regions. While all are provincial-level administrative units, they have distinct functional roles and degrees of autonomy, which allows the sample to reflect different modes of policy practice.
Finally, to minimize subjective selection bias, one policy text was randomly drawn from the data element policy database of each selected provincial-level administrative unit to serve as the evaluation sample. Through this sampling procedure, we ultimately obtained a final sample of 15 representative data element policies. The policy sample is detailed in Table 3.

3.4.2. Variable Setting and Parameter Identification

This study constructed the PMC-Index system for data element policies by integrating the results of the high-frequency word analysis and the LDA topic model analysis. The five core themes identified through LDA (see Table 2) reflect the underlying strategic priorities embedded in the policy corpus. These data-driven themes define the key dimensions of policy content and form the basis for the secondary variables under the policy focus (X5) dimension in the PMC framework. This alignment ensures that the evaluation reflects the actual focus of policy discourse, rather than relying solely on theoretical assumptions. The mapping between LDA-derived themes and PMC sub-dimensions is detailed in Table 4.
The final PMC-Index system is composed of 9 primary variables and 41 secondary variables. The primary variables include policy nature (X1), policy timeliness (X2), policy receptors (X3), policy perspective (X4), policy focus (X5), policy tool (X6), content evaluation (X7), green development assessment (X8), and policy guarantee (X9). Detailed descriptions of these variables and the rules for parameter identification are provided in Table 5.

3.4.3. Construction of a Multi-Input–Output Table

The multi-input–output table serves as the analytical framework for the quantitative evaluation of data element policies, storing the data used to score each variable within the PMC-Index framework. Based on the variable definitions and parameter identification rules established for data element policy, this study constructed the multi-input–output table shown in Table 6.

3.4.4. PMC-Index Calculation

Following Estrada’s research [76], the calculation process for the data element policy PMC-Index is as follows: (1) The value for each secondary variable within a policy is determined according to a binary assignment, as shown in Formulas (1) and (2), where 1 indicates the presence of a criterion and 0 indicates its absence. (2) The score for each primary variable is calculated as the average of its constituent secondary variable scores, as shown in Formula (3). This ensures the primary variable score is also within the [0, 1] range. (3) The final PMC-Index score for each policy is calculated by summing the scores of all primary variables, as shown in Formula (4).
X ~ N 0,1
X = X R : 0,1
X t = j = 1 n X t j n ,   t = 1,2 , 3,4
P M C = X 1 i = 1 5 X 1 i 5 + X 2 j = 1 3 X 2 j 3 + X 3 k = 1 5 X 3 k 5 + X 4 l = 1 3 X 4 l 3 + X 5 m = 1 5 X 5 m 5 + X 6 n = 1 8 X 6 n 8 + X 7 o = 1 4 X 7 o 4 + X 8 p = 1 4 X 8 p 4 + X 9 q = 1 4 X 9 q 4
As this study utilizes nine primary variables, the PMC-Index scores for the data element policies range from 0 to 9. Following Estrada’s scoring standard [68], the scores are classified into four grades (as shown in Table 7). A score between 8.0 and 9.0 is considered “Perfect”. A score in the 6.0–7.9 range is “Good”. A score between 4.0 and 5.9 is “Acceptable”. A score between 0 and 3.9 is considered “Poor”, indicating low policy consistency.
To ensure the reliability and replicability of the PMC-Index scoring process, this study implemented a rigorous and systematic protocol designed to minimize subjective bias and enhance the consistency and scientific validity of the secondary variable scores.
First, we developed a detailed scoring manual to supplement the variable definitions provided in Table 5. This manual established explicit inclusion and exclusion criteria for each binary variable. For example, for the variable “X6-1: Financial Support” to be assigned a score of “1”, the policy text had to explicitly mention concrete financial mechanisms such as subsidies or special financial incentives. Vague statements, such as “encouraging investment”, that did not specify operational tools were assigned a score of “0”. These clearly defined rules facilitated a uniform interpretation of all policy texts by the researchers.
Second, the scoring was conducted independently by two raters, the first and second authors. Before the formal scoring commenced, both raters jointly coded a non-sample policy text to calibrate their understanding of the scoring manual and resolve any ambiguities. This process also served to further refine the manual.
Third, the two raters independently scored the 15 representative policy texts without consultation. To formally assess the consistency of the results, an Inter-Rater Reliability (IRR) test was conducted using Cohen’s Kappa coefficient. The calculation, based on 615 data points (15 policies × 41 secondary variables), yielded a Cohen’s Kappa of 0.859, indicating a very high level of agreement [87]. Subsequently, the two raters discussed the secondary variables for which their initial scores differed and reached a consensus, forming a unified set of scores.
Finally, after the completion of the internal scoring, we invited a practitioner from the data element policy field and an authoritative academic with extensive experience in policy evaluation research to independently review and provide a comprehensive judgment on the scoring results. The two experts reviewed the secondary variables with contested scores, referencing the scoring manual and the policy documents, and offered their professional opinions to produce the final PMC-Index evaluation results.

3.4.5. PMC Surface Generation

The PMC Surface is designed to visualize the results of the index calculation, intuitively displaying the characteristics of data element policies and the strengths and weaknesses of each dimension through a three-dimensional surface. The PMC matrix forms the basis for constructing the PMC Surface. Thus, the results of the 9 primary variables are retained, creating a symmetric and balanced 3 × 3 matrix for the PMC Surface. This study uses a third-order square matrix composed of the 9 primary variables, as shown in Formula (5). The PMC Surface for each data element policy is then constructed based on the result of Formula (5).
P M C S u r f a c e = X 1 X 2 X 3 X 4 X 5 X 6 X 7 X 8 X 9

4. Empirical Results and Analysis

4.1. The PMC-Index of the Policies

Following the PMC-Index procedure, the 15 selected data element policies were evaluated using the multi-input–output table to assign scores to their secondary variables, with the results detailed in Table A1. Subsequently, the PMC-Index was calculated for each of the 15 policies using Formula (4). The policies were then graded according to the established standards, as shown in Table 8.

4.2. The PMC-Surface of the Policies

The PMC-Surface presents the PMC scores in a graphical format, which allows for a more intuitive illustration of the strengths and weaknesses of data element policies [76]. Based on the PMC-Index calculation results and Formula (5), we generated the PMC Surfaces for the 15 policies in Python. For a more direct interpretation of these surfaces, and following the practice of existing studies [79,88,89], this section selects P1, which has the highest PMC-Index, and P13, which has the lowest, as analytical examples for the PMC-Surface (see Figure 4). The PMC Surfaces for the remaining policies are available in Figure A1.
The height of the PMC-Surface reflects the policy’s effectiveness [79]. Compared with P13, the PMC-Surface for P1 features a larger area of red-color blocks, indicating a greater overall height. Therefore, the P1 policy exhibits higher effectiveness, whereas P13 has significant weaknesses.
The degree of convexity and concavity of the PMC-Surface indicates the degree of policy consistency. On the surface, convex areas correspond to higher scores for the associated primary variables, while concave areas indicate lower scores [90]. Based on the policy weaknesses revealed by these concave areas, evaluators can propose a series of targeted recommendations for improvement [91]. In the surface chart for P1, the primary variables X4, X5, and X9 form the three highest convex points, showing that these three variables contribute the most to the policy’s final score. Conversely, the surface for P1 shows a sharp decline at primary variable X2, forming a distinct concave point. This indicates that the score for P1 on X2 is much lower than on other primary variables. Therefore, P1 needs to be enhanced with long-term strategic planning to improve on primary variable X2, thereby strengthening the policy’s internal consistency. In the surface chart for P13, primary variables X3 and X7 form two convex points, suggesting that these two variables contribute significantly to P13’s final score. However, the concave points corresponding to primary variables X1, X2, X8, and X9 reveal four prominent weaknesses in the P13 policy. Based on these identified weaknesses, a series of recommendations can be proposed. To improve primary variable X1, the policy should include phased guidance for its objectives and a descriptive analysis of the current status of local data element development. Primary variable X2 could be improved by integrating the policy’s short-term goals with its long-term vision. To address the weakness in X8, the policy should promote the integrated application of data elements in areas such as ecological governance, energy conservation, and carbon reduction. Finally, to enhance X9, the government should actively plan for innovative application pilots for data elements, detail the division of responsibilities among various departments, and establish a regular, high-efficiency mechanism for inter-departmental collaboration.

4.3. Overall Result Analysis

The evaluation results show that the PMC indices of the 15 representative policies range from 4.9 to 7.1, with their grades classified as either “Good” or “Acceptable”. Among them, Guangdong’s data element policy (P1) achieved the highest score of 7.06, earning a “Good” grade. Xinjiang’s data element policy (P13) received the lowest score of 4.93, which, despite being the lowest, still falls within the “Acceptable” grade.
Figure 5 presents a radar chart of the primary variable scores for the 15 policies, intuitively displaying their respective strengths and weaknesses across the nine evaluation dimensions. An analysis of the average scores for the primary variables reveals that policy focus (X5), policy perspective (X4), and policy receptors (X3) demonstrated strong performance with relatively high average scores. In contrast, policy nature (X1), policy timeliness (X2), and green development assessment (X8) showed weaker performance with lower average scores. (1) Policy nature (X1): The average score was 0.60. This relatively low score indicates that there is room for improvement in the detailed articulation of the development status, phased guidance, and specific recommendations within the policy formulation and implementation process. (2) Policy timeliness (X2): The average score was 0.44, reflecting a common tendency towards short-termism in the policies, with a relative lack of long-term strategic planning and visionary goal setting extending beyond five years. (3) Policy receptors (X3): The average score was 0.85, suggesting that the policies generally consider a wide range of stakeholders in their formulation, including government departments, enterprises, the public, research institutes, and service agencies, thus demonstrating broad coverage. (4) Policy perspective (X4): The average score was 0.89, indicating that current data element policies are well designed to encompass macro, meso, and micro levels, possessing a high degree of systematic thinking. (5) Policy focus (X5): With an average score of 0.88, this was the highest-scoring dimension, showing that the policies are generally able to formulate specific measures centered on core themes such as the data element industry, data resource management, data element trading systems, service platform construction, and e-governments. (6) Policy tool (X6): The average score was 0.71, indicating a solid integration of various policy tools, although the application of financial support and tax incentives requires further strengthening. (7) Content evaluation (X7): The average score was 0.73, suggesting that the policy content aligns well with comprehensive evaluation standards. Policies from most provinces performed well regarding the accuracy of objectives, sufficiency of evidence, scientific basis of proposals, and detail of planning. (8) Green development assessment (X8): The average score was 0.28, indicating that the green development orientation of current data element policies is relatively weak, with a notable weakness in applications concerning energy efficiency and the integrated utilization of waste. (9) Policy safeguards (X9): The average score was 0.63, suggesting that while the policy safeguard mechanisms provide foundational support for implementation, they require further improvement. These results highlight the strengths and weaknesses of the data element policies, offering clear directions for enhancing their consistency and effectiveness.

4.4. Analysis of Policy Groups by Grade

Based on the PMC-Index evaluation grades, the 15 policies were divided into two groups for detailed analysis: the “Good” grade group and the “Acceptable” grade group. The first group, “Good” policies, comprises nine policies: P1, P2, P3, P4, P5, P6, P8, P11, and P14. The second group, “Acceptable” policies, includes six policies: P7, P9, P10, P12, P13, and P15.

4.4.1. “Good” Grade Policy Group

P1, with a PMC-Index of 7.06 and a “Good” grade, ranked first. Specifically, the Guangdong Province Action Plan for Market-Oriented Allocation Reform of Data Elements received perfect scores in policy perspective (X4), policy focus (X5), and policy guarantee (X9). This is because the policy actively aligns with national strategies, focuses on building core data hubs, promotes regional synergy and industrial development, and is committed to refining data management standards and circulation rules. The policy sets clear, phased objectives and delineates key tasks into detailed operational plans assigned to specific departments, supported by robust organizational leadership, diverse pilot mechanisms, and regular evaluation and supervision to ensure effective execution. In the green development assessment (X8), the policy integrates the concept of green development into the construction of digital infrastructure, for instance, by increasing the proportion of low-carbon energy use and the efficiency of waste resource utilization in data centers and by optimizing the layout of data centers to improve regional carbon emission management. This provides a feasible path for the synergistic and parallel development of data infrastructure construction and green, low-carbon development. However, its score for policy timeliness (X2) was significantly below the average, as its objectives are primarily focused on short-term implementation within three years, lacking a more extended strategic vision. Simultaneously, the policy tool (X6) also shows room for improvement. Therefore, for P1, the optimization path is X2–X6.
P2, with a PMC-Index of 6.43 and a “Good” grade, ranked fifth. Specifically, the Zhejiang Province Pilot Plan for Advancing the Value Realization of Industrial Data received perfect scores in policy receptors (X3), policy perspective (X4), and content evaluation (X7). The policy features a clear and logical structure that includes general requirements, work objectives, key tasks, and support measures. It involves a diverse range of stakeholders, including government departments, data service providers, research institutions, and industry associations. It also integrates various levels of planning, from industrial development to enterprise cultivation, while aligning with national development strategies. The policy itself has clear objectives, a sound evidentiary basis, scientifically grounded planning, and a detailed implementation roadmap. However, its scores for policy timeliness (X2) and green development assessment (X8) were low, indicating insufficient attention to the green, low-carbon scenarios of data element applications, as well as a tendency towards short-termism in its goal setting. Therefore, for P2, the optimization path is X2–X8.
P3, with a PMC-Index of 6.31 and a “Good” grade, ranked sixth. Specifically, the Opinions of the General Office of the Jiangsu Provincial Government on Accelerating the Release of Data Element Value and Cultivating a Thriving Data Industry received perfect scores in policy receptors (X3), policy perspective (X4), and policy focus (X5). This is because the policy is systemically designed across multiple levels, including regional strategic guidance, industrial ecosystem construction, and market entity empowerment, to comprehensively promote the release of data element value and industrial development. The policy broadly covers government, enterprises, research and educational institutions, and related service organizations, setting forth specific requirements and deployments for all stakeholders under clear and comprehensive objectives. Its score for policy tool (X6) was also significantly higher than other similar policies, indicating a mature integration of diverse policy instruments such as financial support, talent cultivation, technical standards, and resource integration. In the green development assessment (X8), the policy explicitly identifies the “low-carbon economy” as a key development direction for data elements, reflecting its focus on the green industrial development of data elements. Its score on policy timeliness (X2) was low, as it focuses primarily on short-term goals and lacks a longer-term strategic outlook. The score for policy guarantee (X9) was also well below average, with relatively weak content on specific mechanisms for collaborative division of labor and pilot projects. Therefore, the optimization path for P3 is X2–X9.
P4, with a PMC-Index of 6.82 and a “Good” grade, ranked second. Specifically, Tianjin’s Implementation Plan for Deepening the Reform of Market-Oriented Allocation of Data Elements received perfect scores in policy perspective (X4), policy focus (X5), and content evaluation (X7). The policy defines clear roles and responsibilities for a wide range of stakeholders, including government, enterprises, research institutions, and data service platforms. It formulates detailed measures around five core themes: the data element industry, data resource management, the trading system, service platform construction, and e-government. Its scientific policy design framework ensures both its relevance and feasibility. In the green development assessment (X8), the policy encourages enterprises to innovate public data development models centered on green and low-carbon principles and to build public databases to enhance the development of ecological governance scenarios, such as meteorology. This reflects the policy’s green development philosophy of leveraging data elements to improve both ecological governance and corporate carbon emission management levels. However, its score for policy guarantee (X9) was low because its inter-departmental collaboration and division of labor mechanisms are relatively weak, and it fails to establish a clear supervisory mechanism to monitor the completion of key tasks. Therefore, the optimization path for P4 is X9.
P5, with a PMC-Index of 6.73 and a “Good” grade, ranked third. Specifically, the Shanghai Action Plan for Promoting the Innovation and Development of the Data Element Industry (2023–2025) received perfect scores in policy receptors (X3), policy perspective (X4), policy tool (X6), and content evaluation (X7). The policy demonstrates a high degree of maturity in its design, systematically covering a diverse range of market and social entities from data suppliers to end-users. It employs a combination of policy instruments, such as financial support, talent development, and the formulation of technical standards, to create a comprehensive policy support system for the development of data elements in Shanghai. In the green development assessment (X8), the policy proposes creating a trusted data space for the energy sector to provide data support for improving energy efficiency. It also aims to cultivate data-intensive industry chain leaders in fields such as “dual-carbon” initiatives, thereby enhancing the carbon management capabilities of the entire industrial cluster. This reflects the policy’s green development orientation of using data elements to empower the low-carbon transformation of key industries. However, the policy’s scores for policy timeliness (X2) and policy guarantee (X9) were low. As an action plan, it primarily provides guidance and planning for the development of Shanghai’s data element industry during the three-year period from 2023 to 2025, lacking a longer-term strategic vision. The policy text also fails to clearly designate the departments responsible for key tasks and is deficient in its supervision and evaluation mechanisms. Therefore, the optimization path for P5 is X2–X9.
P6, with a PMC-Index of 6.68 and a “Good” grade, ranked fourth. Specifically, the Implementation Plan for the Construction of the Data Element Market in Hubei Province received perfect scores in policy receptors (X3), policy perspective (X4), and content evaluation (X7). This is reflected in the policy’s inclusion of government departments at all levels, diverse corporate entities, data service providers, third-party service organizations, and the public within its framework for data element market construction, clearly defining the roles and responsibilities of each stakeholder. The policy’s overall “1 + 2 + 3 + N” framework reflects both province-wide macro-coordination and a focus on pilot projects in specific regions and industries, showcasing a multi-level, multi-dimensional policy perspective. Based on important national-level guidance, the policy details six key tasks into twenty key work items, forming an exhaustive and highly operational implementation roadmap. In the green development assessment (X8), the policy emphasizes the need to accelerate the cultivation and introduction of specialized, industry-savvy data service providers in key areas such as energy conservation, carbon reduction, and green construction. This approach directly links the development of data service providers with green development concepts such as optimizing energy efficiency and reducing emissions, reflecting the policy’s intent to drive the green transformation of related industries through the construction of the data element market. However, the policy’s focus in terms of policy timeliness (X2) is primarily on short-term planning. Its score for policy guarantee (X9) is also relatively low, mainly due to vague definitions of responsibilities for collaborating departments and the lack of a clear, unified leadership mechanism. For P6, the improvement path is X2–X9.
P8, with a PMC-Index of 6.08 and a “Good” grade, ranked ninth. Specifically, the Henan Province Big Data Industry Development Action Plan (2022–2025) achieved perfect scores in policy perspective (X4) and policy guarantee (X9). The policy not only incorporates national big data strategies and provincial digital transformation plans at the macro level but also focuses on the construction of the big data industrial system and the modernization of the industrial chain at the meso level, and it attends to the introduction and cultivation of leading enterprises at the micro level. Simultaneously, the policy establishes a provincial working group for unified leadership, clearly defines the responsible units for key tasks to reflect collaborative division of labor, plans pilot construction projects including striving for national-level big data industry development pilot demonstration projects, and establishes supervisory and evaluation mechanisms such as key project monitoring, pilot demonstrations, and quarterly progress reporting, which collectively ensure the effective implementation of the policy. In the green development assessment (X8), the plan explicitly proposes to strengthen the “Big Data + Ecological Governance” initiative, combining big data applications with ecological and environmental governance. This provides a feasible path for unleashing the green value of data elements. Despite this, the plan scores low on policy nature (X1), lacking a description of the policy’s implementation status, and with relatively insufficient phased guidance and forward-looking prediction for its main objectives. Its policy timeliness (X2) is also concentrated on short-term planning. This policy’s improvement path is referred to as X1–X2.
P11, with a PMC-Index of 6.18 and a “Good” grade, ranked seventh. Specifically, the Chongqing Action Program for Reform of Market-Based Allocation of Data Elements received perfect scores in policy receptors (X3), policy perspective (X4), and policy focus (X5). The policy’s design demonstrates robustness and a well-structured framework. It not only empowers micro-level market entities by effectively aligning with national top-level strategies and providing clear guidance for regional industrial development, but it also broadly incorporates government, enterprises, and research institutions into its framework. Furthermore, it provides comprehensive and in-depth coverage of key areas such as industry cultivation, resource management, and the trading system. However, the policy’s scores for policy timeliness (X2) and green development assessment (X8) were low because it primarily sets short-term goals within a three-year timeframe, and its content fails to reflect a green development orientation. Therefore, the optimization path for P11 is X2–X8.
P14, with a PMC-Index of 6.15 and a “Good” grade, ranked eighth. Specifically, the Guiding Opinions on the Development of the Big Data Industry in Jilin Province achieved perfect scores in policy focus (X5) and policy guarantee (X9). This is because the policy presents a clear and comprehensive development blueprint with strong implementation support, meticulously planning objectives for industrial growth, data resource management, element market rule construction, and the deepening of e-government. It also ensures effective execution through project leadership, departmental collaboration, overall coordination, and monitoring and evaluation. In the green development assessment (X8), the policy explicitly proposes promoting the application of satellite remote sensing data in areas such as rural land resource surveys, land use status investigations, monitoring and yield estimation of agricultural and forestry pests and crop growth, forest resource inventories, and environmental monitoring. This fully demonstrates the policy’s intent to use big data technology to serve ecological protection and sustainable development. Despite this, its scores for policy timeliness (X2) and policy perspective (X4) were significantly below average, mainly because the policy’s planning and specific indicators are focused on short-term results, and its overall alignment with national development strategies is relatively limited. Therefore, the optimization path for P14 is X2–X4.

4.4.2. “Acceptable” Grade Policy Group

P7, with a PMC-Index of 5.42 and an “Acceptable” grade, ranked twelfth. Specifically, the Jiangxi Province Data Application Regulations received a perfect score for policy timeliness (X2), reflecting its long-term and stable nature as a piece of local legislation. In the green development assessment (X8), the policy proposes the enhancement of the level of data-empowered ecological civilization construction, the strengthening of the digital and intelligent management of carbon emissions, and the use of digital technology to promote the valuation of ecological products, the realization of the “dual-carbon” goals, and smart ecological and environmental governance. This content closely links data elements with national strategies such as ecological civilization construction and the “dual-carbon” targets, promoting green and low-carbon development through data application. Nevertheless, its scores for policy receptors (X3), policy tool (X6), content evaluation (X7), and policy guarantee (X9) were significantly below average. This is because the policy mainly involves the government and general enterprises, with insufficient consideration for innovative and service-oriented entities such as universities, research institutions, and data service organizations and platforms. Furthermore, the policy employs a limited range of policy tools, relying primarily on regulatory controls, with little mention of measures anticipated by market entities, such as direct financial support, tax incentives, robust market promotion, and deep industrial resource integration. Additionally, although the policy clarifies departmental responsibilities, it lacks specific plans for pilot projects, detailed designs for cross-departmental and cross-level collaboration, and a sound mechanism for evaluating and dynamically adjusting its own implementation effectiveness. The optimization path for P7 is X3-X6-X7-X9.
P9, with a PMC-Index of 5.28 and an “Acceptable” grade, ranked thirteenth. Specifically, the Implementation Plan for the Market-Oriented Allocation of Data Elements in Guizhou Province received a perfect score for policy perspective (X4). By aligning with the nation’s top-level design for data infrastructure systems, the policy systematically plans the path for the province’s data element market cultivation, industrial development, and enterprise empowerment, demonstrating a comprehensive consideration of national strategy, regional planning, and market entities in its top-level design. However, its scores in policy timeliness (X2), policy tool (X6), green development assessment (X8), and policy guarantee (X9) were well below average, indicating deficiencies in policy guidance, long-term planning, specific implementation methods, and the detailing of guarantee measures. The optimization path for P9 is X2-X6-X8-X9.
P10, with a PMC-Index of 5.65 and an “Acceptable” grade, ranked tenth. Specifically, the Implementation Plan for the Comprehensive Reform of the Market-Oriented Allocation of Data Elements in Sichuan Province received a perfect score for policy focus (X5), setting clear, comprehensive, and operable policy focal points. In the green development assessment (X8), the plan explicitly proposes to advance intelligent ecological and environmental governance and to thoroughly implement a synergistic action plan for digital and green transformation. This content directly links data elements with ecological environmental governance and synergistic digital–green transformation, reflecting the policy’s emphasis on green development. Nevertheless, its scores for policy timeliness (X2), policy perspective (X4), and policy tool (X6) were well below average. This is because the policy does not effectively embed the tasks of data element marketization reform into the region’s strategic development direction, focuses mainly on short-term goals, and is deficient in the combined use of policy instruments such as taxation, talent, technology, and publicity. Therefore, for P10, the optimization path is X2-X4-X6.
P12, with a PMC-Index of 5.48 and an “Acceptable” grade, ranked eleventh. Specifically, the Implementation Opinions of Ningxia Hui Autonomous Region on Promoting the Development of the Data Element Market received perfect scores in both policy perspective (X4) and policy focus (X5). This indicates that the policy has a clear strategic direction, enabling it to focus on the core tasks of data element development while effectively aligning with broader development goals. However, the policy’s scores in the policy timeliness (X2), content evaluation (X7), and policy guarantee (X9) were all low. Therefore, the optimization path for P12 is X2-X7-X9.
P13, with a PMC-Index of 4.93 and an “Acceptable” grade, ranked lowest among the 15 policies. Specifically, the Measures for Public Data Management in Xinjiang Uyghur Autonomous Region scored below average in every dimension and performed particularly poorly in dimensions such as policy nature (X1), policy timeliness (X2), policy perspective (X4), policy focus (X5), and policy guarantee (X9). This indicates that there is significant room for improvement in the overall quality of the policy’s design. For P13, the optimization path is X1-X2-X4-X5-X9.
P15, with a PMC-Index of 5.17 and an “Acceptable” grade, ranked fourteenth. As a piece of local legislation, the Regulations on the Development of Big Data in Liaoning Province aim to strengthen data governance, fully leverage data utility, and build a healthy ecosystem to accelerate the development of big data in Liaoning. The policy received a perfect score for policy timeliness (X2), reflecting its long-term guiding nature. However, it scored below average in policy nature (X1), policy perspective (X4), content evaluation (X7), green development assessment (X8), and policy guarantee (X9). Specifically, the policy lacks content on the supervision of the data element market and guiding plans for its policy focus, and its integration with the national data element development strategy is not sufficiently deep. Crucially, the policy’s design lacks clear and distinct objectives and a sufficient basis for its measures. At the same time, it pays insufficient attention to social welfare and green development, and its guarantee measures do not mention mechanisms for collaborative division of labor or supervision and evaluation. These factors compromise the realization of the policy’s focus and its implementation effectiveness. The optimization path for P15 is X1-X4-X7-X8-X9.

5. Discussion and Conclusions

5.1. Discussion

This study conducted a quantitative evaluation of China’s data element policies within the context of green development using the LDA-PMC model. The evaluation results indicate that the data element policies demonstrate strong performance in specific domains. Specifically, policy focus achieved the highest average score, suggesting that the policies generally provide clear and comprehensive coverage of the core focal areas for data element applications. Furthermore, most policies successfully incorporate a range of stakeholders, including government departments, enterprises, and research institutions, into their implementation frameworks, reflecting an inclusive approach to policy design. Finally, the policy perspective of most policies exhibits a high degree of systematic thinking, effectively integrating development needs at the macro, meso, and micro levels.
However, the evaluation also reveals several issues in policy design and formulation. The most significant problem is a severe lack of alignment with green development objectives. Among the 15 policies evaluated, 4 (P2, P9, P11, and P15) contained no measures whatsoever for the green development of data elements. The majority of the remaining policies focused only on unidimensional green development measures. This suggests that current policy design prioritizes the economic benefits of data element development over formulating targeted measures to promote the integrated application of data elements in support of green and low-carbon initiatives. Second, most policy objectives exhibit a short-term bias, with a general absence of medium- and long-term goals and planning. This indicates that policymakers have placed greater emphasis on immediately feasible measures and timeliness, with insufficient consideration for the long-term stability of the policy framework. Third, the prescriptive and descriptive content of the policies is weak. This is particularly evident in the inadequate decomposition of policy objectives and the lack of clear guidance on phased implementation pathways. This weakness may create ambiguity for frontline agencies in interpreting policy objectives, potentially leading to deficient policy implementation or the adoption of “one-size-fits-all” approaches. Finally, the support systems for policy implementation are not yet fully developed. Specifically, there is a lack of detailed implementation guidelines, concrete measures, and clear allocations of responsibility, resulting in weak mechanisms for collaboration and division of labor. This may, in turn, undermine the effectiveness of policy implementation.

5.2. Implications for Public Policy

Based on the research results and discussion, the following recommendations are provided for future data element policymakers.
First, the green development orientation of all policies should be specifically strengthened. In policymaking, integrating targeted environmental protection measures with digital economy policies can ensure both economic development and environmental sustainability [92]. Under the guidance of the national “dual-carbon” policy, the development of data elements should support the management of dual-carbon goals, reinforce the crucial role of digital technology in green development, and achieve transformative changes in industrial energy efficiency and carbon emission reduction [93]. Future optimization of data element policies should center on the unique mechanisms through which data elements promote green development. This includes strengthening the application of data elements in ecological governance, energy utilization, waste resource management, and carbon emissions management. It is also essential to actively construct a comprehensive support system that integrates data elements with green, low-carbon development, encompassing policies, standards, talent, research, public participation, and international cooperation. Furthermore, it is recommended that legally binding green development implementation clauses be introduced into data element policies to make green-oriented goals more concrete and actionable. For example, specific resource efficiency standards should be established for new data centers, such as for power usage effectiveness (PUE), water usage effectiveness (WUE), and the proportion of renewable energy use, to mitigate the environmental externality risks of data infrastructure.
Second, long-term strategic planning should be balanced with short-term policy execution. To enhance the coordination and implementation effectiveness of the data element policy system, a dynamic policy framework that balances strategic foresight with operational feasibility should be constructed. Specifically, the commonly adopted three-year action plans for data elements should be organically linked with national digital economy development plans and medium- to long-term “dual-carbon” targets. By establishing a systematic long-term policy framework, a clear development direction, strategic vision, and phased objectives can be defined. On this basis, short-term policies should focus on the implementation of key tasks and the refined deployment of policy tools to ensure that policy objectives are operationally feasible. Concurrently, it is necessary to establish a mechanism for regular evaluation and dynamic adjustment, mandating an assessment and revision of the long-term framework every two to three years. This will ensure that policies can adapt to rapid changes in technology, markets, and the environment, thereby achieving an organic balance between enhancing policy implementation and maintaining policy continuity, stability, and foresight. Additionally, it is important to strengthen the description of the current development status, challenges, and opportunities for regional data elements, breaking down short-term goals into specific tasks based on the actual conditions of social development and providing more actionable implementation guidelines to offer clear direction for stakeholder behavior.
Third, clear mechanisms for inter-agency collaboration and leadership should be established. The “silo effect” in the public sector is a significant cause of policy misalignment, implementation gaps, and resource inefficiency [94]. To enhance policy execution and implementation effectiveness, the policy implementation support section should include special provisions that systematically define the core responsibilities, task boundaries, and collaborative pathways for all relevant government departments, such as those for development and reform, industry and information technology, and the ecological environment. This will avoid functional overlap and ambiguous responsibilities, thereby improving the clarity and controllability of policy execution. Furthermore, institutionalized mechanisms for regular inter-departmental communication, information sharing processes, and dispute resolution procedures should be mandated, and a lead agency must be designated to ensure efficient and coordinated policy execution under a unified leadership structure.
Finally, the application of ex ante policy evaluation tools should be strengthened. Traditional policy evaluation models often rely on ex post assessments, which not only result in costly revisions but also frequently miss optimal opportunities for intervention. To systematically enhance policy quality, the PMC-Index model could be institutionalized as a standard ex ante evaluation tool for use across departments and domains. Before policy drafts proceed to final deliberation, the PMC evaluation tool should be employed for an ex ante assessment. By identifying policy design flaws through multi-dimensional textual analysis and quantitative scoring, this process would compel policymakers to optimize policy design from the outset. This ensures a balanced and internally consistent policy structure and provides methodological support for the construction of a scientific and sophisticated system of evidence-based policymaking.

5.3. Conclusions

This study provides a quantitative evaluation of China’s data element policies within the context of green development. By integrating the Latent Dirichlet Allocation (LDA) topic model and the Policy Modeling Consistency (PMC) Index model, we developed a PMC-Index evaluation framework for data element policies, which comprises 9 primary and 41 secondary variables. Subsequently, this framework was used to evaluate 15 representative policies to quantitatively assess their effectiveness and identify their deficiencies. The analysis yields several key findings. First, the PMC-Index scores of the 15 data element policies ranged from 4 to 8, with an average score of 6.03. Nine policies received a “Good” Grade, and six were assigned an “Acceptable” Grade. Second, the study identified five core thematic areas within the policy texts: the data element industry, data resource management, data element trading rules, service platform construction, and e-government. Finally, the evaluation revealed an imbalance in policy performance. China’s data element policies scored highly on indicators such as policy perspective, policy receptors, and policy focus. In contrast, scores were lower for policy nature, policy timeliness, green development assessment, and policy guarantee, indicating weaker performance in these specific domains. Notably, the green development assessment indicator received the lowest score, which clearly reflects that the integration of a green development orientation in current data element policies is severely insufficient. These findings provide a clear direction for the future improvement and optimization of data element policies.

5.3.1. Contributions of the Paper

The theoretical and practical contributions of our study are summarized as follows.
Theoretical contribution: This study evaluates data element policy texts within the context of green development, addressing a notable gap in the existing research. It is the first study to apply the integrated LDA-PMC model to the quantitative evaluation of data element policies, thereby enriching the body of knowledge and the methodological framework for data element policy assessment.
Practical contribution: This research provides the Chinese government with valuable insights and an empirical basis for supplementing and refining existing policies. The findings have direct practical implications for the sustainable development of data elements and can offer decision support for the implementation, adjustment, and revision of future policy cycles.

5.3.2. Limitations and Future Prospects

This study has several limitations. First, the LDA model cannot capture the semantic nuances and institutional context of policies, while the binary scoring rule of the PMC model is relatively inflexible. These methodological shortcomings may lead to a loss of qualitative policy information. Second, this study presents a static analysis of policy texts, which makes it difficult to fully capture the dynamic evolution of policy. Furthermore, the value assignment in the PMC-Index model scoring relies on the subjective coding of the researchers. Although this study employed a cross-coding validation method to mitigate this issue, the scoring process inevitably retains a degree of subjectivity. Finally, the reliability of this study’s conclusions is highly dependent on the quality of the collected policy texts. If the textual content is overly vague and general, or if documents are missing, the accuracy of the LDA topic extraction and the validity of the PMC scoring can be directly affected.
To address these limitations, the following directions for future research are proposed. First, more advanced natural language processing (NLP) models could be employed for a deeper semantic analysis. Concurrently, exploring weighted or fuzzy scoring systems could refine the PMC-Index to reflect the nuances that are difficult to capture with a binary scoring method. Second, future research could incorporate dynamic analytical methods to investigate the evolution of policies over time, analyzing their developmental trajectories and dynamic effects. Additionally, natural language processing could be combined with text-coding techniques, such as the PMC-TE model, to automate the coding and value assignment of policy texts, thereby enhancing the replicability and validity of the research. Finally, a corpus could be constructed using a more diverse range of textual sources, such as implementation guidelines and official interpretations, to provide a more reliable data foundation for policy text analysis and to produce more comprehensive and robust findings.

Author Contributions

Conceptualization, S.H. and X.W.; methodology, X.W.; software, X.W.; validation, S.H. and X.W.; formal analysis, X.W.; investigation, S.H.; resources, S.H.; data curation, S.H.; writing—original draft preparation, X.W.; writing—review and editing, S.H.; visualization, X.W.; supervision, S.H.; project administration, S.H.; funding acquisition, S.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Key Program of the National Social Science Fund of China (23AZD035).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.

Acknowledgments

The authors would like to thank all of the anonymous reviewers for their constructive comments regarding this article.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Table A1. Multi-input–output table of 15 representative policies.
Table A1. Multi-input–output table of 15 representative policies.
Primary VariablesSecondary VariablesP1P2P3P4P5P6P7P8P9P10P11P12P13P14P15
X1X1-1111111001111010
X1-2101111011101111
X1-3111001111110111
X1-4110111100011000
X1-5010000100000000
X2X2-1000000100000001
X2-2000100100001001
X2-3111111111110111
X3X3-1111111111111111
X3-2111111111111111
X3-3011111100011100
X3-4111011011110111
X3-5111111011111011
X4X4-1111111011011000
X4-2111111111111111
X4-3111111111111111
X5X5-1111111111111011
X5-2111111111111111
X5-3111111111111011
X5-4111111111111111
X5-5101100000111110
X6X6-1001010001110011
X6-2000010000001001
X6-3111111111001110
X6-4111111010011000
X6-5011111110010111
X6-6111111110111111
X6-7111111111111111
X6-8111111011111111
X7X7-1011111011000010
X7-2111111001110110
X7-3111111111111110
X7-4110111010111101
X8X8-1000100110101110
X8-3100011000000000
X8-4100000000000000
X8-5101111100000000
X9X9-1110111011110011
X9-2100000010001010
X9-3111110111110111
X9-4111001110110110

Appendix B

Figure A1. The PMC surface of the remaining 13 policies.
Figure A1. The PMC surface of the remaining 13 policies.
Sustainability 17 06758 g0a1aSustainability 17 06758 g0a1bSustainability 17 06758 g0a1c

References

  1. Webb, J.; de Silva, H.N.; Wilson, C. The Future of Coal and Renewable Power Generation in Australia: A Review of Market Trends. Econ. Anal. Policy 2020, 68, 363–378. [Google Scholar] [CrossRef]
  2. Tanaka, K.; Wilson, C.; Managi, S. Impact of Feed-in Tariffs on Electricity Consumption. Environ. Econ. Policy Stud. 2022, 24, 49–72. [Google Scholar] [CrossRef]
  3. Ghobakhloo, M. Industry 4.0, Digitization, and Opportunities for Sustainability. J. Clean. Prod. 2020, 252, 119869. [Google Scholar] [CrossRef]
  4. Machado, C.G.; Winroth, M.P.; Ribeiro da Silva, E.H.D. Sustainable Manufacturing in Industry 4.0: An Emerging Research Agenda. Int. J. Prod. Res. 2020, 58, 1462–1484. [Google Scholar] [CrossRef]
  5. Research Report on the Development of China’s Digital Economy (2024)—China’s Business Environment Monitoring Service Platform. Available online: http://www.chbeo.org.cn/article_view.aspx?id=6486 (accessed on 14 July 2025).
  6. Energy Economics | Home. Available online: https://www.bp.com/en/global/corporate/energy-economics.html (accessed on 14 July 2025).
  7. Deng, Z.; Zhu, B.; Davis, S.J.; Ciais, P.; Guan, D.; Gong, P.; Liu, Z. Global Carbon Emissions and Decarbonization in 2024. Nat. Rev. Earth Environ. 2025, 6, 231–233. [Google Scholar] [CrossRef]
  8. Chen, L.; Zhao, J.; Wang, J.; Ren, Q.; Liu, W.; Liu, Q. Measuring and Optimizing Pathways for Regional Economic and Low-Carbon Coordination Effects: A Case Study of the Yangtze River Economic Belt. Energy Strategy Rev. 2025, 59, 101725. [Google Scholar] [CrossRef]
  9. Usman, A.; Ozturk, I.; Hassan, A.; Maria Zafar, S.; Ullah, S. The Effect of ICT on Energy Consumption and Economic Growth in South Asian Economies: An Empirical Analysis. Telemat. Inform. 2021, 58, 101537. [Google Scholar] [CrossRef]
  10. Chang, H.; Ding, Q.; Zhao, W.; Hou, N.; Liu, W. The Digital Economy, Industrial Structure Upgrading, and Carbon Emission Intensity—Empirical Evidence from China’s Provinces. Energy Strategy Rev. 2023, 50, 101218. [Google Scholar] [CrossRef]
  11. Zhao, Z.; Zhao, Y.; Shi, X.; Zheng, L.; Fan, S.; Zuo, S. Green Innovation and Carbon Emission Performance: The Role of Digital Economy. Energy Policy 2024, 195, 114344. [Google Scholar] [CrossRef]
  12. Mueller, M.; Grindal, K. Data Flows and the Digital Economy: Information as a Mobile Factor of Production. Digit. Policy Regul. Gov. 2018, 21, 71–87. [Google Scholar] [CrossRef]
  13. Dubey, R.; Gunasekaran, A.; Childe, S.J.; Papadopoulos, T.; Luo, Z.; Wamba, S.F.; Roubaud, D. Can Big Data and Predictive Analytics Improve Social and Environmental Sustainability? Technol. Forecast. Soc. Change 2019, 144, 534–545. [Google Scholar] [CrossRef]
  14. Dong, L.; Zhu, X.; Yang, L.; Jiang, G. Unleashing the Power of Data Element Markets: Driving Urban Green Growth through Marketization, Innovation, and Digital Finance. Int. Rev. Econ. Financ. 2025, 99, 104070. [Google Scholar] [CrossRef]
  15. Balogun, A.-L.; Marks, D.; Sharma, R.; Shekhar, H.; Balmes, C.; Maheng, D.; Arshad, A.; Salehi, P. Assessing the Potentials of Digitalization as a Tool for Climate Change Adaptation and Sustainable Development in Urban Centres. Sustain. Cities Soc. 2020, 53, 101888. [Google Scholar] [CrossRef]
  16. Wang, W.; Xiao, D. Marketization of Data Elements and Enterprise Green Governance Performance: A Quasi-Natural Experiment Based on Data Trading Platforms. Manag. Decis. Econ. 2025, 46, 1686–1700. [Google Scholar] [CrossRef]
  17. Linjia, T.; Chenyang, Z. Green Data Elements: Connotation Characteristics, Impact Effects, and Research Prospects. J. Technol. Econ. 2025, 44, 1–16. [Google Scholar]
  18. Jacobs, M. The Green Economy: Environment, Sustainable Development and the Politics of the Future; UBC Press: Vancouver, BC, Canada, 1991. [Google Scholar]
  19. Tao, C.-Q.; Yi, M.-Y.; Wang, C.-S. Coupling Coordination Analysis and Spatiotemporal Heterogeneity between Data Elements and Green Development in China. Econ. Anal. Policy 2023, 77, 1–15. [Google Scholar] [CrossRef]
  20. Bianchini, S.; Damioli, G.; Ghisetti, C. The Environmental Effects of the “Twin” Green and Digital Transition in European Regions. Environ. Resource Econ. 2023, 84, 877–918. [Google Scholar] [CrossRef]
  21. Xiao, Y.; Wu, S.; Liu, Z.Q.; Lin, H.J. Digital Economy and Green Development: Empirical Evidence from China’s Cities. Front. Environ. Sci. 2023, 11, 1124680. [Google Scholar] [CrossRef]
  22. Yaoteng, Z.; Xin, L. Research on Green Innovation Countermeasures of Supporting the Circular Economy to Green Finance under Big Data. J. Enterp. Inf. Manag. 2021, 35, 1305–1322. [Google Scholar] [CrossRef]
  23. Jia, K.; Chen, Y. Research on the Impact of Data Factors on Enterprise Green Innovation—Evidence from Chinese Manufacturing Enterprises. Sustainability 2025, 17, 2184. [Google Scholar] [CrossRef]
  24. Moşteanu, N.R. Green Sustainable Regional Development and Digital Era. In Green Buildings and Renewable Energy: Med Green Forum 2019—Part of World Renewable Energy Congress and Network; Sayigh, A., Ed.; Springer International Publishing: Cham, Switzerland, 2020; pp. 181–197. ISBN 978-3-030-30841-4. [Google Scholar]
  25. Wu, L.; Sun, L.; Qi, P.; Ren, X.; Sun, X. Energy Endowment, Industrial Structure Upgrading, and CO2 Emissions in China: Revisiting Resource Curse in the Context of Carbon Emissions. Resour. Policy 2021, 74, 102329. [Google Scholar] [CrossRef]
  26. Liu, G.; Liu, Y.; Lee, C.-C. Growth Sources of Green Economy and Energy Consumption in China: New Evidence Accounting for Heterogeneous Regimes. Energy J. 2020, 41, 33–64. [Google Scholar] [CrossRef]
  27. Li, J.; Quan, T.; Han, Q. Open Government Data and Green Economic Growth. Econ. Anal. Policy 2025, 85, 2207–2218. [Google Scholar] [CrossRef]
  28. Jetzek, T.; Avital, M.; Bjorn-Andersen, N. Data-Driven Innovation through Open Government Data. J. Theor. Appl. Electron. Commer. Res. 2014, 9, 15–16. [Google Scholar] [CrossRef]
  29. Jetzek, T.; Avital, M.; Bjorn-Andersen, N. The Sustainable Value of Open Government Data. J. Assoc. Inf. Syst. 2019, 20, 702–734. [Google Scholar] [CrossRef]
  30. Chen, K.; Zhang, S. How Does Open Public Data Impact Enterprise Digital Transformation? Econ. Anal. Policy 2024, 83, 178–190. [Google Scholar] [CrossRef]
  31. Wu, D.; Xie, Y. Unveiling the Impact of Public Data Access on Collaborative Reduction of Pollutants and Carbon Emissions: Evidence from Open Government Data Policy. Energy Econ. 2024, 138, 107822. [Google Scholar] [CrossRef]
  32. Lange, S.; Pohl, J.; Santarius, T. Digitalization and Energy Consumption. Does ICT Reduce Energy Demand? Ecol. Econ. 2020, 176, 106760. [Google Scholar] [CrossRef]
  33. Kunkel, S.; Tyfield, D. Digitalisation, Sustainable Industrialisation and Digital Rebound—Asking the Right Questions for a Strategic Research Agenda. Energy Res. Soc. Sci. 2021, 82, 102295. [Google Scholar] [CrossRef]
  34. Masanet, E.; Shehabi, A.; Lei, N.; Smith, S.; Koomey, J. Recalibrating Global Data Center Energy-Use Estimates. Science 2020, 367, 984–986. [Google Scholar] [CrossRef]
  35. Danish; Khan, N.; Baloch, M.A.; Saud, S.; Fatima, T. The Effect of ICT on CO2 Emissions in Emerging Economies: Does the Level of Income Matters? Environ. Sci. Pollut. Res. 2018, 25, 22850–22860. [Google Scholar] [CrossRef]
  36. Khan, H.; Weili, L.; Khan, I. Examining the Effect of Information and Communication Technology, Innovations, and Renewable Energy Consumption on CO2 Emission: Evidence from BRICS Countries. Environ. Sci. Pollut. Res. 2022, 29, 47696–47712. [Google Scholar] [CrossRef] [PubMed]
  37. Wu, Q.; Li, S. Decarbonization by Digits: How Data Factors Drive Nonlinear Sustainable Dynamics in Manufacturing. Appl. Energy 2024, 374, 123967. [Google Scholar] [CrossRef]
  38. Lennerfors, T.T.; Fors, P.; Rooijen, J.v. ICT and Environmental Sustainability in a Changing Society: The View of Ecological World Systems Theory. Inf. Technol. People 2015, 28, 758–774. [Google Scholar] [CrossRef]
  39. Echegaray, F.; Hansstein, F.V. Assessing the Intention-Behavior Gap in Electronic Waste Recycling: The Case of Brazil. J. Clean. Prod. 2017, 142, 180–190. [Google Scholar] [CrossRef]
  40. Dias, P.; Bernardes, A.M.; Huda, N. Waste Electrical and Electronic Equipment (WEEE) Management: An Analysis on the Australian e-Waste Recycling Scheme. J. Clean. Prod. 2018, 197, 750–764. [Google Scholar] [CrossRef]
  41. Hu, X.; Jiang, Y.; Guo, P.; Li, M. How Does China’s Big Data Policy Affect the Digital Economy of Cities? Evidence from National Big Data Comprehensive Pilot Zones. Heliyon 2024, 10, e24638. [Google Scholar] [CrossRef]
  42. Casanovas, P.; De Koker, L.; Mendelson, D.; Watts, D. Regulation of Big Data: Perspectives on Strategy, Policy, Law and Privacy. Health Technol. 2017, 7, 335–349. [Google Scholar] [CrossRef]
  43. Hashem, I.A.T.; Chang, V.; Anuar, N.B.; Adewole, K.; Yaqoob, I.; Gani, A.; Ahmed, E.; Chiroma, H. The Role of Big Data in Smart City. Int. J. Inf. Manag. 2016, 36, 748–758. [Google Scholar] [CrossRef]
  44. McNeely, C.L.; Hahm, J. The Big (Data) Bang: Policy, Prospects, and Challenges. Rev. Policy Res. 2014, 31, 304–310. [Google Scholar] [CrossRef]
  45. Wang, W.; Zhang, H.; Sun, Z.; Wang, L.; Zhao, J.; Wu, F. Can Digital Policy Improve Corporate Sustainability? Empirical Evidence from China’s National Comprehensive Big Data Pilot Zones. Telecommun. Policy 2023, 47, 102617. [Google Scholar] [CrossRef]
  46. Bodó, B.; Irion, K.; Janssen, H.; Giannopoulou, A. Personal Data Ordering in Context: The Interaction of Meso-Level Data Governance Regimes with Macro Frameworks. Internet Policy Rev. 2021, 10, 2–31. [Google Scholar] [CrossRef]
  47. Wei, J.; Zhang, X. The Role of Big Data in Promoting Green Development: Based on the Quasi-Natural Experiment of the Big Data Experimental Zone. Int. J. Environ. Res. Public Health 2023, 20, 4097. [Google Scholar] [CrossRef] [PubMed]
  48. Broeders, D.; Cristiano, F.; Kaminska, M. In Search of Digital Sovereignty and Strategic Autonomy: Normative Power Europe to the Test of Its Geopolitical Ambitions. J. Common Mark. Stud. 2023, 61, 1261–1280. [Google Scholar] [CrossRef]
  49. Mac Síthigh, D.; Siems, M. The Chinese Social Credit System: A Model for Other Countries? Mod. Law Rev. 2019, 82, 1034–1071. [Google Scholar] [CrossRef]
  50. Bertot, J.C.; Gorham, U.; Jaeger, P.T.; Sarin, L.C.; Choi, H. Big Data, Open Government and e-Government: Issues, Policies and Recommendations. Inf. Polity 2014, 19, 5–16. [Google Scholar] [CrossRef]
  51. Osifo, O.C. A Study of Coordination Challenges in Digital Policy Implementation and Evaluation in Finland. In Proceedings of the 2020 43rd International Convention on Information, Communication and Electronic Technology (MIPRO), Opatija, Croatia, 28 September–2 October 2020; pp. 1402–1409. [Google Scholar]
  52. Zhao, J.; Liu, H.; Zhang, S.; Shu, K.; Fang, Q.; Zhang, W.; Tan, C.; Zhang, X.; Wang, Y.; Qi, Y. A “Five Rights Separation” Framework for Data Rights Confirmation in Data Element Circulation. Humanit. Soc. Sci. Commun. 2025, 12, 205. [Google Scholar] [CrossRef]
  53. Mahrenbach, L.; Mayer, K.; Pfeffer, J. Policy Visions of Big Data: Views from the Global South. Third World Q. 2018, 39, 1861–1882. [Google Scholar] [CrossRef]
  54. Jung, K.; Park, H.W. A Semantic (TRIZ) Network Analysis of South Korea’s “Open Public Data” Policy. Gov. Inf. Q. 2015, 32, 353–358. [Google Scholar] [CrossRef]
  55. Custers, B.; Uršič, H. Big Data and Data Reuse: A Taxonomy of Data Reuse for Balancing Big Data Benefits and Personal Data Protection. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3046774 (accessed on 21 July 2025).
  56. Spiekermann, S.; Acquisti, A.; Böhme, R.; Hui, K.-L. The Challenges of Personal Data Markets and Privacy. Electron Mark. 2015, 25, 161–167. [Google Scholar] [CrossRef]
  57. Kwon, T.H.; Kwak, J.H.; Kim, K. A Study on the Establishment of Policies for the Activation of a Big Data Industry and Prioritization of Policies: Lessons from Korea. Technol. Forecast. Soc. Change 2015, 96, 144–152. [Google Scholar] [CrossRef]
  58. Zuiderwijk, A.; Janssen, M. Open Data Policies, Their Implementation and Impact: A Framework for Comparison. Gov. Inf. Q. 2014, 31, 17–29. [Google Scholar] [CrossRef]
  59. Davies, P. The State of Evidence-Based Policy Evaluation and Its Role in Policy Formation. Natl. Inst. Econ. Rev. 2012, 219, R41–R52. [Google Scholar] [CrossRef]
  60. Head, B.W. Toward More “Evidence-Informed” Policy Making? Public Adm. Rev. 2016, 76, 472–484. [Google Scholar] [CrossRef]
  61. Sanderson, I. Evaluation, Policy Learning and Evidence-Based Policy Making. Public Adm. 2002, 80, 1–22. [Google Scholar] [CrossRef]
  62. Wollmann, H. Policy Evaluation and Evaluation Research. In Handbook of Public Policy Analysis; Routledge: London, UK, 2017; pp. 419–428. [Google Scholar]
  63. Nugroho, R.P.; Zuiderwijk, A.; Janssen, M.; de Jong, M. A Comparison of National Open Data Policies: Lessons Learned. Transform. Gov. People Process Policy 2015, 9, 286–308. [Google Scholar] [CrossRef]
  64. Zhou, M.; Wang, Y.; Huang, X.; Li, G. Can Open Government Data Policy Improve Firm Performance? Evidence from Listed Firms in China. Manag. Decis. Econ. 2023, 44, 2593–2603. [Google Scholar] [CrossRef]
  65. Ma, X.; Ruan, Y. How to Evaluate Green Development Policy Based on the PMC Index Model: Evidence from China. Int. J. Environ. Res. Public Health 2023, 20, 4249. [Google Scholar] [CrossRef]
  66. Turnpenny, J.; Radaelli, C.M.; Jordan, A.; Jacob, K. The Policy and Politics of Policy Appraisal: Emerging Trends and New Directions. J. Eur. Public Policy 2009, 16, 640–653. [Google Scholar] [CrossRef]
  67. Koura, A.; Abdeslam, B.; Mohamed, O. Assessing Public Policies: A Comprehensive Review of Evaluation Methods. 2024. Available online: https://zenodo.org/records/14035228 (accessed on 21 July 2025).
  68. Livermore, M.A.; Revesz, R.L. (Eds.) The Globalization of Cost-Benefit Analysis in Environmental Policy; Oxford University Press: Oxford, UK, 2013; ISBN 978-0-19-993438-6. [Google Scholar]
  69. Dehnhardt, A.; Grothmann, T.; Wagner, J. Cost-Benefit Analysis: What Limits its Use in Policy Making and How to Make it More Usable? A Case Study on Climate Change Adaptation in Germany. Environ. Sci. Policy 2022, 137, 53–60. [Google Scholar] [CrossRef]
  70. Tol, R.S.J. A Cost–Benefit Analysis of the EU 20/20/2020 Package. Energy Policy 2012, 49, 288–295. [Google Scholar] [CrossRef]
  71. Løken, E. Use of Multicriteria Decision Analysis Methods for Energy Planning Problems. Renew. Sustain. Energy Rev. 2007, 11, 1584–1595. [Google Scholar] [CrossRef]
  72. Browne, D.; O’Regan, B.; Moles, R. Use of Multi-Criteria Decision Analysis to Explore Alternative Domestic Energy and Electricity Policy Scenarios in an Irish City-Region. Energy 2010, 35, 518–528. [Google Scholar] [CrossRef]
  73. EvalCommunity. Logic Model: A Comprehensive Guide to Program Planning, Implementation, and Evaluation. Available online: https://www.evalcommunity.com/career-center/logic-models/ (accessed on 21 July 2025).
  74. Langley, T.; Gillespie, D.; Lewis, S.; Eminson, K.; Brennan, A.; Docherty, G.; Young, B. Developing Logic Models to Inform Public Health Policy Outcome Evaluation: An Example from Tobacco Control. J. Public Health 2020, 43, 639–646. [Google Scholar] [CrossRef] [PubMed]
  75. Petticrew, M.; Eastmure, E.; Mays, N.; Knai, C.; Bryden, A. The Public Health Responsibility Deal: How Should Such a Complex Public Health Policy be Assessed? Lancet 2012, 380, S11. [Google Scholar] [CrossRef]
  76. Ruiz Estrada, M.A. The Policy Modeling Research Consistency Index (PMC-Index). SSRN J. 2010. [Google Scholar] [CrossRef]
  77. Li, Y.; He, R.; Liu, J.; Li, C.; Xiong, J. Quantitative Evaluation of China’s Pork Industry Policy: A PMC Index Model Approach. Agriculture 2021, 11, 86. [Google Scholar] [CrossRef]
  78. Zhang, C.; Zhang, J.; Zhang, B. Quantitative Evaluation of Budget Performance Management Policy: Evidence from China. J. Econ. Surveys 2025. Early View. [Google Scholar] [CrossRef]
  79. Kuang, 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. [Google Scholar] [CrossRef]
  80. Dai, S.; Zhang, W.; Zong, J.; Wang, Y.; Wang, G. How Effective is the Green Development Policy of China’s Yangtze River Economic Belt? A Quantitative Evaluation Based on the PMC-Index Model. Int. J. Environ. Res. Public Health 2021, 18, 7676. [Google Scholar] [CrossRef]
  81. Xu, L.; Xi, M. Quantitative Evaluation of Policies for Combining Medical and Nursing Care Based on the LDA–PMC Model: A Comparative Analysis of Typical Chinese Provinces. Public Organ. Rev. 2024, 24, 693–713. [Google Scholar] [CrossRef]
  82. Kherwa, P.; Bansal, P. Topic Modeling: A Comprehensive Review. EAI Endorsed Trans. Scalable Inf. Syst. 2019, 7, 159623. [Google Scholar] [CrossRef]
  83. Shen, H.; Xiong, P.; Yang, L.; Zhou, L. Quantitative Evaluation of Science and Technology Financial Policies Based on the PMC-AE Index Model: A Case Study of China’s Science and Technology Financial Policies since the 13th Five-Year Plan. PLoS ONE 2024, 19, e0307529. [Google Scholar] [CrossRef]
  84. Provan, G. Toward Explainable AutoEncoder-Based Diagnosis of Dynamical Systems. Algorithms 2023, 16, 178. [Google Scholar] [CrossRef]
  85. Griffiths, T.L.; Steyvers, M. Finding Scientific Topics. Proc. Natl. Acad. Sci. USA 2004, 101, 5228–5235. [Google Scholar] [CrossRef] [PubMed]
  86. Hong, 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, e0298312. [Google Scholar] [CrossRef] [PubMed]
  87. Landis, J.R.; Koch, G.G. The Measurement of Observer Agreement for Categorical Data. Biometrics 1977, 33, 159–174. [Google Scholar] [CrossRef] [PubMed]
  88. Liu, 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, 9338. [Google Scholar] [CrossRef]
  89. Geng, R.; Wu, J. Assess the Efficacy of China’s Inter-Provincial Government Services Policy: A Quantitative Evaluation Based on PMC-Index Model. PLoS ONE 2024, 19, e0310491. [Google Scholar] [CrossRef]
  90. Zhang, Y.; Jin, Y.; Xue, X. Evaluation of Construction Industrialization Policy Based on PMC Index Model. In Proceedings of the International Conference on Construction and Real Estate Management 2018, Charleston, SC, USA, 9–10 August 2018; pp. 192–201. [Google Scholar]
  91. Ruiz Estrada, M.A. Policy Modeling: Definition, Classification and Evaluation. J. Policy Model. 2011, 33, 523–536. [Google Scholar] [CrossRef]
  92. Acemoglu, D.; Aghion, P.; Bursztyn, L.; Hemous, D. The Environment and Directed Technical Change. Am. Econ. Rev. 2012, 102, 131–166. [Google Scholar] [CrossRef]
  93. Xie, B.; Liu, R.; Dwivedi, R. Digital Economy, Structural Deviation, and Regional Carbon Emissions. J. Clean. Prod. 2024, 434, 139890. [Google Scholar] [CrossRef]
  94. Lah, O. Breaking the Silos: Integrated Approaches to Foster Sustainable Development and Climate Action. Sustain. Earth Rev. 2025, 8, 1. [Google Scholar] [CrossRef]
Figure 1. Trend of policy text issuance.
Figure 1. Trend of policy text issuance.
Sustainability 17 06758 g001
Figure 2. Perplexity distribution curve of data element policies. The X-axis represents the number of topics, and the Y-axis indicates the perplexity of the LDA model. The lowest point on the Y-axis corresponds to the optimal number of topics, reflecting the best model fit.
Figure 2. Perplexity distribution curve of data element policies. The X-axis represents the number of topics, and the Y-axis indicates the perplexity of the LDA model. The lowest point on the Y-axis corresponds to the optimal number of topics, reflecting the best model fit.
Sustainability 17 06758 g002
Figure 3. Flow chart of the PMC-Index model construction.
Figure 3. Flow chart of the PMC-Index model construction.
Sustainability 17 06758 g003
Figure 4. The PMC surface of the highest-scoring policy (P1) and the lowest-scoring policy (P13).
Figure 4. The PMC surface of the highest-scoring policy (P1) and the lowest-scoring policy (P13).
Sustainability 17 06758 g004
Figure 5. Radar chart of 15 policies and average scores.
Figure 5. Radar chart of 15 policies and average scores.
Sustainability 17 06758 g005
Table 1. Policy text key words’ frequency statistics table.
Table 1. Policy text key words’ frequency statistics table.
No.High-Frequency WordsFrequencyNo.High-Frequency WordsFrequency
1service230316collaboration641
2platform170117openness641
3industry136018network601
4innovation133419infrastructure598
5digitalization132320sector583
6government affairs114121product554
7public data109122system551
8big data106323industrial509
9sharing83524internet504
10supervision77625data resources500
11integration75626model479
12information72527data center464
13transaction70828scenario460
14cultivation69929data security455
15governance68130ecology411
Table 2. Topics with the corresponding terms of data element policies.
Table 2. Topics with the corresponding terms of data element policies.
Topic CodeTopic NameTopic Top 10 High Probability Feature Words
T-1Data Element
Industry
industrial park, e-commerce, terminal, equipment, high-end, digital commerce, agricultural products, satellite, materials, satellite
T-2Data Resource
Management
collection, acquisition, application, administration, public data sharing, data processing, review, materials, regulatory management, legal benefits
T-3Data Element Trading Systemdata trading platform, data authorization operation, profit, data vendor, profit distribution, rights, data infrastructure system, pricing, data property rights, standardization
T-4Service Platform
Construction
cyberspace administration, one-stop, community, law enforcement, handle affairs, construction project, online services, command, handle, licenses and permits
T-5E-Governmentadministration, public affairs, assessment, grassroots, licenses and permits, iteration, agency, approval, prevention and control, official
Table 3. List of data element policy for evaluation.
Table 3. List of data element policy for evaluation.
No.ProvinceRegionPolicy TitleIssue Date
P1GuangdongEast ChinaGuangdong Province Action Plan for Market-Oriented Allocation Reform of Data ElementsJuly 2021
P2ZhejiangEast ChinaZhejiang Province Pilot Plan for Advancing the Value Realization of Industrial DataNovember 2022
P3JiangsuEast ChinaOpinions of the General Office of the Jiangsu Provincial Government on Accelerating the Release of Data Element Value and Cultivating a Thriving Data IndustryNovember 2024
P4TianjinEast ChinaTianjin’s Implementation Plan for Deepening the Reform of Market-Oriented Allocation of Data ElementsSeptember 2024
P5ShanghaiEast ChinaShanghai Action Plan for Promoting the Innovation and Development of the Data Element Industry (2023–2025)August 2023
P6HubeiCentral ChinaImplementation Plan for the Construction of the Data Element Market in Hubei ProvinceAugust 2023
P7JiangxiCentral ChinaJiangxi Province Data Application RegulationsNovember 2023
P8HenanCentral ChinaHenan Province Big Data Industry Development Action Plan (2022–2025)September 2022
P9GuizhouWest ChinaImplementation Plan for the Market-oriented Allocation of Data Elements in Guizhou ProvinceAugust 2023
P10SichuanWest ChinaImplementation Plan for the Comprehensive Reform of the Market-oriented Allocation of Data Elements in Sichuan ProvinceJanuary 2024
P11ChongqingWest ChinaChongqing Action Program for Reform of Market-based Allocation of Data ElementsDecember 2023
P12NingxiaWest ChinaImplementation opinions of Ningxia Hui Autonomous Region on promoting the development of data element marketMay 2024
P13XinjiangWest ChinaMeasures for Public Data Management in Xinjiang Uygur Autonomous RegionMay 2024
P14JilinNortheast ChinaGuiding Opinions on the Development of the Big Data Industry in Jilin ProvinceFebruary 2024
P15LiaoningNortheast ChinaRegulations on the Development of Big Data in Liaoning ProvinceMay 2022
Table 4. Conceptual mapping of LDA themes to PMC policy focus (X5) Variables.
Table 4. Conceptual mapping of LDA themes to PMC policy focus (X5) Variables.
LDA-Identified Topic NameCorresponding PMC Secondary Variable
T-1: Data Element IndustryX5-1: Data Element Industry
T-2: Data Resource ManagementX5-2: Data Resource Management
T-3: Data Element Market Trading RulesX5-3: Data Element Trading Rules
T-4: Service Platform ConstructionX5-4: Service Platform Construction
T-5: E-GovernmentX5-5: E-Government
Table 5. Variable design and evaluation standard.
Table 5. Variable design and evaluation standard.
CodePrimary VariableSub-CodeSecondary VariableReference Standard
X1Policy NatureX1-1PredictionWhether the policy includes predictive content
(1: Yes, 0: No)
X1-2RecommendationWhether the policy contains recommendations
(1: Yes, 0: No)
X1-3SupervisionWhether the policy reflects supervision
(1: Yes, 0: No)
X1-4GuidanceWhether the policy provides guidance
(1: Yes, 0: No)
X1-5DescriptionWhether the policy describes the current development status (1: Yes, 0: No)
X2Policy TimelinessX2-1Long-TermWhether the policy covers more than 5 years
(1: Yes, 0: No)
X2-2Medium-TermWhether the policy covers 3–5 years
(1: Yes, 0: No)
X2-3Short-TermWhether the policy covers less than 3 years
(1: Yes, 0: No)
X3Policy ReceptorsX3-1Government
Departments
Whether the policy applies to government departments (1: Yes, 0: No)
X3-2EnterprisesWhether the policy applies to enterprises
(1: Yes, 0: No)
X3-3PublicWhether the policy applies to the public
(1: Yes, 0: No)
X3-4Universities &
Research Institutes
Whether the policy applies to universities and research institutes (1: Yes, 0: No)
X3-5Service Agencies & PlatformsWhether the policy applies to service agencies and platforms (1: Yes, 0: No)
X4Policy
Perspective
X4-1Macro-LevelWhether the policy focuses on a macro level
(1: Yes, 0: No)
X4-2Meso-LevelWhether the policy focuses on a meso level
(1: Yes, 0: No)
X4-3Micro-LevelWhether the policy focuses on a micro level
(1: Yes, 0: No)
X5Policy FocusX5-1Data Element
Industry
Whether the policy involves the development of the data element industry (1: Yes, 0: No)
X5-2Data Resource
Management
Whether the policy involves data resource management (1: Yes, 0: No)
X5-3Data Element
Trading System
Whether the policy involves the trading rules for data elements (1: Yes, 0: No)
X5-4Service Platform ConstructionWhether the policy involves the construction of service platforms (1: Yes, 0: No)
X5-5E-GovernmentWhether the policy involves e-government
(1: Yes, 0: No)
X6Policy ToolX6-1Financial SupportWhether the policy involves financial support (1: Yes, 0: No)
X6-2Tax IncentivesWhether the policy involves tax incentives (1: Yes, 0: No)
X6-3Talent DevelopmentWhether the policy involves talent development (1: Yes, 0: No)
X6-4Technical SupportWhether the policy involves technical support (1: Yes, 0: No)
X6-5Promotion and
Publicity
Whether the policy involves promotion and publicity (1: Yes, 0: No)
X6-6Regulatory ControlWhether the policy involves regulatory control (1: Yes, 0: No)
X6-7Technical StandardsWhether the policy involves technical standards (1: Yes, 0: No)
X6-8Resource IntegrationWhether the policy involves resource integration (1: Yes, 0: No)
X7Content
Evaluation
X7-1Clear GoalsWhether the policy sets specific goals
(1: Yes, 0: No)
X7-2Sufficient BasisWhether the policy is based on real-world
practice (1: Yes, 0: No)
X7-3Scientific PlanningWhether the implementation plan of the policy is scientific (1: Yes, 0: No)
X7-4Detailed RoadmapWhether the policy includes a detailed timeline
(1: Yes, 0: No)
X8Green Development AssessmentX8-1Ecological Governance EnhancementWhether the policy involves using data elements to improve ecological governance (1: Yes, 0: No)
X8-2Energy Efficiency ImprovementWhether the policy involves using data elements to improve energy utilization efficiency
(1: Yes, 0: No)
X8-3Waste Resource
Utilization
Whether the policy involves using data elements to improve the efficiency of waste resource utilization (1: Yes, 0: No)
X8-4Carbon Emissions ManagementWhether the policy involves using data elements to improve carbon emissions management
(1: Yes, 0: No)
X9Policy GuaranteeX9-1Pilot ConstructionWhether the policy involves pilot construction (1: Yes, 0: No)
X9-2Collaboration and Division of LaborWhether the policy involves collaboration and division of labor (1: Yes, 0: No)
X9-3Overall LeadershipWhether the policy involves overall leadership (1: Yes, 0: No)
X9-4Supervision and EvaluationWhether the policy involves supervision and evaluation (1: Yes, 0: No)
Table 6. Multi-input–output of each data element policy.
Table 6. Multi-input–output of each data element policy.
Primary VariableSecondary Variable
X1X1-1, X1-2, X1-3, X1-4, X1-5
X2X2-1, X2-2, X2-3
X3X3-1, X3-2, X3-3, X3-4, X3-5
X4X4-1, X4-2, X4-3
X5X5-1, X5-2, X5-3, X5-4, X5-5
X6X6-1, X6-2, X6-3, X6-4, X6-5, X6-6, X6-7, X6-8
X7X7-1, X7-2, X7-3, X7-4
X8X8-1, X8-2, X8-3, X8-4
X9X9-1, X9-2, X9-3, X9-4
Table 7. Evaluation grade for policies based on the PMC-Index.
Table 7. Evaluation grade for policies based on the PMC-Index.
PMC-Index0–3.94.0–5.96.0–7.98.0–9.0
Evaluation gradesPoorAcceptableGoodPerfect
Table 8. PMC-Index of 15 representative policies.
Table 8. PMC-Index of 15 representative policies.
X1X2X3X4X5X6X7X8X9PMC-IndexGradeRank
P10.800.330.801.001.000.630.750.751.007.06Good1
P20.800.331.001.000.800.751.000.000.756.43Good5
P30.600.331.001.001.000.880.750.250.506.31Good6
P40.600.670.801.001.000.751.000.500.506.82Good2
P50.600.331.001.000.801.001.000.500.506.73Good3
P60.800.331.001.000.800.751.000.500.506.68Good4
P70.601.000.600.670.800.500.250.500.505.42Acceptable12
P80.400.330.801.000.800.750.750.251.006.08Good9
P90.600.330.801.000.800.500.750.000.505.28Acceptable13
P100.600.330.800.671.000.500.750.250.755.65Acceptable10
P110.600.331.001.001.000.750.750.000.756.18Good7
P120.600.330.801.001.000.750.500.250.255.48Acceptable11
P130.400.330.800.670.600.630.750.250.504.93Acceptable15
P140.600.330.800.671.000.750.750.251.006.15Good8
P150.401.000.800.670.800.750.250.000.505.17Acceptable14
Average0.600.440.850.890.880.710.730.280.636.03//
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hu, S.; Wang, X. A Text-Mining-Based Evaluation of Data Element Policies in China: Integrating the LDA and PMC Models in the Context of Green Development. Sustainability 2025, 17, 6758. https://doi.org/10.3390/su17156758

AMA Style

Hu S, Wang X. A Text-Mining-Based Evaluation of Data Element Policies in China: Integrating the LDA and PMC Models in the Context of Green Development. Sustainability. 2025; 17(15):6758. https://doi.org/10.3390/su17156758

Chicago/Turabian Style

Hu, Shuigen, and Xianbo Wang. 2025. "A Text-Mining-Based Evaluation of Data Element Policies in China: Integrating the LDA and PMC Models in the Context of Green Development" Sustainability 17, no. 15: 6758. https://doi.org/10.3390/su17156758

APA Style

Hu, S., & Wang, X. (2025). A Text-Mining-Based Evaluation of Data Element Policies in China: Integrating the LDA and PMC Models in the Context of Green Development. Sustainability, 17(15), 6758. https://doi.org/10.3390/su17156758

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop