Abstract
The Chinese Government and Statistics Departments have collected huge amounts of data. They use these government open data to conduct applied research and technological studies, including the establishment of a market-based system for data elements. These data are put into e-government to build the construction of digital government and improve the level of government services continuously. This paper uses the perspective of the configurations to discuss the combinations of factors in the marketization of open data of local governments in China with fsQCA method; this paper is based on the TOE framework, and the indices of the data layer of the government open data forest in 16 provinces are taken as the research object. The conclusions are as follows: (1) the market-oriented utilization paths of local government open data are categorized as technology-led, management-resource-led, economic and technology-driven, and location-advantaged. (2) Under certain conditions, when the number of data developers is small, and there is no dedicated government data management agency, optimizing and improving the government open data platform could be an alternative path to promote the effective use of local government open data in the market.
1. Introduction
The Central Committee of the Communist Party of China and China's State Council issued a policy in March 2020 on building a more complete institutional mechanism for the market-based allocation of factors. This policy includes accelerating the cultivation of data factor markets, promoting the open sharing of government data, enhancing the value of social data resources, and strengthening the integration of data resources and data security protection. In May of the same year, another policy was introduced on Accelerating the Improvement of the Socialist Market Economy System in the New Era. It further updated the detailed requirements, such as accelerating the cultivation and development of the data factor market, establishing a data resource list management mechanism, improving the definition of data ownership, sharing, trading circulation, and other standards and measures to bring into play the value of social data resources. At the same time, this policy further requires strengthening the construction of digital government and protecting personal information. Data provides indispensable power support for the development of the digital economy. With the iterative update of data analysis algorithms and technologies, new technologies such as the Internet of Things, artificial intelligence, and blockchain developed based on big data and data technology have had a disruptive impact on social development. E-Government 2.0 has been upgraded to 3.0, and the extended layering of open government data analysis and utilization has accelerated the development of open government technologies. In some developed countries, the development of open data movement started earlier and developed more rapidly. Scholars develop a framework for quality assessment with government portals. They have solid theoretical research and more systematic and complete policies to complement their guarantee [1]. The government’s open data operation and development has formed a more systematic system, which could provide a more refined and efficient approach to social governance, traffic safety, health care, and other public management issues through data [2]. In China, the level of local government open data development is at a high rate of development. According to DMG Lab Fudan University Statistics 2020 Second Half of China Open Data Index Report, it showed that, as of October 2020, China has 142 government public data open platforms at the end of October 2020, which was a 40% increase compared to the year of 2019. Government open data is different from social data due to its attributes and types. If the two are fused and shared, it would produce multiplier effects, which can help cultivate a new industry of data sharing and utilization, realize data flow interconnection and interoperability, and further release the value of the data factor. Compared to developed countries, China, as a developing country, has a different level of digital development. While developed countries are at an advanced stage of development, e-government in developing countries is still in the exploratory stage [3]. The research purpose: the construction of digital government not only determines the level of governance of the government itself, but is also a key variable affecting the development of the digital economy and digital society. Through the massive data sets formed after digital transformation, data of government could use scientific methods to identify the specific impact of various digital services on the accessibility of public services meticulously and the sense of access to livelihood security for different members of society, and to precisely optimize national policy guidance based on these studies. At present, the digitalization of local governments in China has not yet resulted in scalable and replicable innovations in the market-based application of big data. Participants in market transactions include large enterprises and research institutions, among others, who function as toolboxes of high-quality data pools to connect and flow dormant government data elements. Therefore, a market-oriented approach that makes full use of the data collected by government agencies could continuously enhance the value of society's data resources. This study develops government open data according to local conditions and explores development paths to enhance the market-based use of government data, which is conducive to empowering the construction of digital government and improving the digitization and intelligence of social governance. The research question: although China has abundant data resources at all levels of government and statistical bureaus, the data utilization and operation of each region differ due to the uneven distribution of resources and different levels of development in each city and region. Therefore, what factors can prompt the awakening of dormant government data? How these factors can make China's local government open data further utilized by the market is the question to be answered in this paper.
2. Literature Review
2.1. The Urgency of Market-Based Use of Government Open Data
There is a certain urgency to study the market operation and utilization of government open data in China. Currently, there is a great deal of applied research and technological innovation based on data resources in China, which provides a wide space and practice area for many research activities. Data has become a de facto production input factor in China, and how to supply data to the market is a key step in fostering a data factor market [4]. Compared with traditional factors of production, data has unique characteristics, such as nonexclusivity, the economy of scale, sustainable regeneration, and strong permeability. These characteristics have a revolutionary impact on enterprise production, industrial transformation and upgrading, and macroeconomic regulation. Some scholars examine the quality of data, data circulation, value-added data processing, and data foundation services from the asset attributes of data and propose a management framework for government data asset management. Government data governance under the factor perspective should stand on the demand side and emphasize the market transformation of data value and the user benefits of data innovation. The data factor market can be improved by connecting government data with market data through data identification, data optimization, allocation, and cooperation and sharing. Cooperative sharing through data validation connects the government’s data with the market’s data, which can improve the data factor market. Therefore, there is an urgent need for local governments and data management authorities at all levels to develop data standards, which include the identification of data rights and clarification of how data resources are managed and traded. Attention should be paid to normative theories and institutions in big data-driven public services to ensure that these services are more targeted and prospective. In addition, user privacy and security issues are also included.
Many scholars currently have relatively limited research on the use and output of local government data in China but coincidentally view open government data as an ecological whole influenced by multiple factors [5]. Scholars have used sociotechnological systems theory to construct a model of the open government data ecosystem that encompasses policy strategy, data opening and utilization, feedback and communication, benefit generation, and interactions among stakeholders [6]. Many problems have arisen in the development process due to the immaturity of the different local authorities in terms of how to disclose data correctly [7]. Inadequate organization and limited IT infrastructure and skills in some areas make the development of open government data more difficult [8]. A large and authoritative source of data elements in China is mainly collected by government agencies and statistical institutions. These data are a significant source for developing the data factor trading market and exploring the data factor economy. It is necessary to further explore new data standardization scenarios and strengthen scientific data management services to market and effectively transform them into production factors.
2.2. Technology-Organization-Environment (TOE) Framework
Tornatzky et al. proposed the TOE theoretical framework consisting of Technology, Organization, and Environment in 1990 [9]. This theoretical framework was originally used to analyze the impact of technology, organization, and environment on the adoption of new technologies. Since the TOE framework does not specify specific explanatory variables, it is systematic and flexible and has been widely used to study the adoption and implementation process of new strategies, policies, and business models [10]. Scholars have adapted and added to the TOE framework according to different research fields or research objects. TOE framework has examined the factors influencing the construction of public sector websites in China [11], the factors influencing the digital governance of provincial governments and the implementation paths [12], the data management of local government services [13], and government information platforms [14]. These research contents have improved the effectiveness and adaptability of the TOE framework theory.
The technological dimension emphasizes the characteristics of the technology itself as well as other relevant technological factors [15, 16]. It reflects the overall technological resource endowment of a region. It measures the ability of a region to aggregate technology and knowledge and translate that knowledge and technology into new ideas, products, and applications. With the update and development of big data technology, the interaction and integration of data are also increasing. Trading systems as technology infrastructure influence the ability of organizations to adopt innovative technologies [17]. The regional big data trading system provides safe and reliable information system support for data trading, building a credit evaluation system for data from both sides of the transaction [2]. It also increases the flow of data trading in the region and across regions and better connects government open data with market needs [18]. The second influencing factor is the construction of government open data platforms. The Internet may improve governance by facilitating local governments to better serve and engage citizens through platforms such as e-government platforms [19, 20]. The digital platform is an intermediary organization connecting multilateral markets, realizing extensive connection and reorganization of various resources, and realizing value interaction and cocreation of multilateral economic subjects and platform organizers through the rules set by itself [21]. In the era of the information revolution, an interconnected platform can realize the scale effect of data and maximize the advantages of data, which can promote the coupling of data flow, information flow, and value flow to the maximum. The infrastructure and data analysis capabilities of data platforms also greatly influence the paradigm of data collection, analysis, processing, and utilization. How large data platforms filter the intricate and complex types of data, extract key information, and systematically process and publish data is a key technological process for marketable data utilization. The local government’s open data platforms have always been the most authoritative and data-rich public platforms among the various Internet platforms currently available in China. The government open data platform is the most direct platform to realize government data integration, data development, and utilization. As a representative of the level of data development and utilization in a region, the importance of each provincial (municipality directly under the central government) government open data platform is mainly reflected in the coordination work. The impact of the provincial integration situation can be summarized in two ways: on the one hand, the higher the level of integration, the more the high-level coordination forces available to local data management; on the other hand, the higher the level of provincial integration, the higher the overall level within the province. Therefore, the construction of provincial government open data platforms is crucial. So, the technological dimensions include big data trading systems and government open data platforms.
The organizational dimension includes governance, senior management support, and technological competence [13, 17, 22]. Organizational factors are structural safeguards for technological development and can also have a profound impact on the effectiveness of the use of government data. The e-government participants on the supply and demand sides are the suppliers (government) and consumers (citizens), respectively [17]. Strategic management by government managers is important, and top management can provide guidance, support, and commitment to create an enabling environment [23–25]. In April 2020, the UK government established the Data Standards Authority (DSA), which is part of the Government Digital Service (GDS). Its main responsibilities are to build data standards, improve data sharing and utilization across departments, and ensure data quality. Therefore, a dedicated data management agency is the key to the healthy development of open data. As the official formal data management body plays a key role in the organization's use of data [26], the life cycle of data generation-utilization-regeneration constitutes the “data chain,” which goes through multiple stages from generation to completion, including collection, cleaning, review, release, acquisition, development, use, feedback and improvement, and data regeneration. Local governments are the generators of data, while the Government Data Authority is the opener and manager of data [27]. In the data cycle, data openers are mainly responsible for completing the work related to data collection to publication. The restructuring and deployment of government data management agencies provide a solid guarantee of data exploitation. Moreover, government participation plays a key role and helps overcome obstacles in the implementation of OGD. The data developer starts to develop data products and services by acquiring data [25]. Then, the data consumers or employees consume the related products and services and give feedback on their consumption experience [11, 28]. The consumer and developer of data are a collection of organizations such as related companies, universities, media, and social groups [29], which are the main developers of data products and services. The third-party organizations and technology companies related to the data industry could effectively promote the application of open government data [30]. And the consumers of the data element market are mainly third-party institutions and technology business personnel related to the information transmission, computer services, and software industries. Therefore, we have selected government regulatory bodies as formal organizers and people from the information transfer and software industry as informed participants in the organizational dimension.
The environmental dimension focuses on contextual factors that could have an impact on technological capabilities, such as the level of economic development, resources, demand, infrastructure, and other factors. The current digital economy in China has been increasing the level of digital and intelligent development through deep integration with the real economy. In 2020, the size of China’s digital economy is nearly US$5.4 trillion, ranking second in the world; with a year-on-year growth rate of 9.6%, the growth rate is the first in the world, and the level of development of the digital economy in the region has become an important environment to promote the use and development of data. China’s State Council released the Action Plan for Promoting the Development of Big Data on August 31, 2015. It includes the construction of a national zone, which will conduct experimental exploration in big data system innovation, public data opening and sharing, big data innovation and application, big data industry gathering, big data element circulation, data center integration, and utilization, and big data international exchange and cooperation to promote big data innovation and development. On October 8, 2016, the National Development and Reform Commission, the Ministry of Industry and Information Technology, and the Central Network Information Office jointly decided to promote the construction of national big data comprehensive pilot areas in seven regions. These regions specifically include the Beijing-Tianjin-Hebei and Pearl River Delta cross-regional class comprehensive pilot area, four regional demonstration class comprehensive pilot areas in Shanghai, Henan Province, Chongqing, and Shenyang, and the Inner Mongolia big data infrastructure integrated development class comprehensive pilot area. In these regions around the seven major tasks mentioned above to carry out systematic experiments, it is hoped that the pilot zone could form a radiation-driven and demonstration-leading effect.
Based on the TOE framework, this paper constructs an application model for promoting the market use of local government open data among 6 influencing factors in 3 dimensions: technology, organization, and environment from a holistic perspective (Figure 1).
3. Research Methodology
3.1. Qualitative Comparative Analysis
Qualitative comparative analysis (QCA) is an asymmetric data analysis technique that combines the logic and empirical intensity of qualitative approaches that are rich in contextual information, with quantitative methods that deal with large numbers of cases and are more generalizable than symmetric theory and tools. The method combines the idea of aggregation with that of Boolean algebra and uses cases as the research guide to derive the relationships between several variables and different causal effects by comparing the results of different cases under different combinations of conditional variables, which is especially suitable for small sample studies with sample size below 60. Fuzzy-Set Qualitative Comparative Analysis (fsQCA) is considered to be an effective method for exploring combination effects and interactions [31]. It has been widely used in recent years in various areas of management disciplines, and it may produce different results for several different variables.
This paper adopts Fuzzy-Set Qualitative Comparative Analysis mainly based on the following two reasons. Firstly, there are 34 provincial administrative regions in China currently, but the number of provinces that have opened open government data is less than 34. There are only 16 samples that can be extracted in the China Open Data Index Report (second half of 2020). The number of samples is not suitable for large-scale statistical analysis and meets the requirement of small and medium samples for qualitative comparative analysis. Secondly, some scholars argue that government, data users, the general public, and the external environment are indispensable in the process of creating data value, and they constitute an “ecosystem” that jointly determines the ultimate effect of open government data [32]. The modeling framework for the market-based use of open government data is a nonlinear process involving multiple factors. Different factors will have different impacts, so a qualitative comparative analysis is appropriate. The text adopts a fuzzy qualitative comparative analysis, taking into account quantitative analysis and qualitative research, and tries to achieve an appropriate causal complexity while retaining the appropriate causal complexity. The combination of conditional variables is expressed in as simple a form as possible while retaining the appropriate causal complexity.
3.2. Outcome Variable: Data Layer Index
The criteria of data collection are the basis for data element market transactions and are a prerequisite for data trading and pricing of data elements in big data trading. The availability of data is the basis for the timely and effective query and utilization of data by users and is the most fundamental criterion for assessing data quality [33]. Government open data has several conditions to be utilized. The first condition is that data should be easy to extract and be downloaded, which requires the government to open the data platform to provide data that is easy to download [34]. The second condition is convenience [35]. After the platform downloads the data, it should be convenient for third parties to use, which requires that the format of the data is common and recognized by the data analysis software. The market-oriented use of government open data depends to a large extent on the convenient application of government open data. The third condition is high quality. The content of government open data should be comprehensive, real, and diverse to meet the needs of different using subjects. The last condition is updated timeliness [34]. The fourth is updated timeliness, which is one of the criteria for evaluating the activity of the data. The timely updated data can reflect the latest information, while the static and outdated ones are not conducive to value mining [36]. In this study, the outcome variable is taken from the data layer index in the report of the 2020 Second Half of China Open Data Index. The data layer index reflects the indicators of data capacity, data quality, data specification, and open scope of government open data. Data capacity refers to the number of fields (columns) multiplied by the number of articles (rows) of downloadable, structured data sets released in each time batch in a local platform, reflecting the data volume and granularity of downloadable data sets open on the platform; in terms of data quality, high-quality data sets with large open data capacity and high social demand are the focus of open data; data specification mainly refers to the open authorization protocols, which mark the open attributes for the datasets with hierarchical classification; the open scope mainly refers to the thematic distribution of open datasets around the world, which facilitates the subsequent use and processing. The data layer accounts for 40% of the whole data forest index, which shows its important status. The Data Layer Index in the report is a review of key and priority open data areas based on the “Opinions of the State Council of the Central Committee of the Communist Party of China on Building a More Perfect Institutional Mechanism for Market-Based Allocation of Factors” and the requirements of local government policies and regulations. So, this data index is quite authoritative.
3.3. Variable Assignment
Variables and their measurement methods are shown in Table 1. The outcome variable is taken from the data layer index in the report of the 2020 Second Half of China Open Data Index.
3.3.1. Technological Dimension
(1)Big data trading system (abbreviated as Business). This study obtained the sample by counting the big data centers under or led by each provincial government. Provinces with big data transaction systems take the value of 1, while those without take the value of 0.(2)Government open data platform (abbreviated as Platform). This study gets the sample from the platform layer index of the report of the 2020 Second Half of China Open Data Index to indicate the effect of building the platform layer of provincial government open data.
3.3.2. Organizational Dimension
(1)Government data management authority (abbreviated as Administrator). This study will count the government data management authorities of each province through the research of official government documents and websites. The existence of a special government data management agency takes the value of 1, while the absence of it takes the value of 0.(2)The number of employees in the information transmission, computer services, and software industry (abbreviated as Human). In this paper, based on the data of China City Statistical Yearbook 2020, the number of employees in the information transmission, computer services, and software industry in each province is counted, and the total number of statistics in each province and city is taken as the sample.
3.3.3. Environmental Dimension
(1)Regional geographic location (abbreviated as Location). After the official report and website statistics, this paper will take the value of 1 for the provinces or municipalities located in the national big data comprehensive pilot zone, and 0 for the provinces not located in the zone.(2)The level of regional digital economy development (abbreviated as Economic). The Digital Economy Development Index of this paper adopts the evaluation results of “2020 China Digital Economy Development Index (DEDI)” published by CCID Consulting, which is a scientific evaluation index reflecting the development of the digital economy by synthesizing and calculating typical indicators from various dimensions of the digital economy using statistical methods. The indicators of DEDI in this study are taken from this report.
3.4. Data Calibration
In fsQCA, the condition and outcome each are considered as a set, respectively, and each case has affiliation scores in these sets, and the process of assigning set affiliation scores to cases is calibration [37]. Consistent with existing research, in this paper, based on previous experience and theoretical knowledge, the data are converted to fuzzy set affiliation scores using the direct calibration method based on the type of data for each condition and outcome [31]. Calibration had three thresholds: 95% was the threshold for the case data as full membership; 50% was the threshold for the case data as crossover point; 5% was the threshold for the case data as full nonmembership. Crossover point thus qualitatively anchors a fuzzy set’s midpoint between full membership and full nonmembership [38] (Table 2).
3.5. Analysis of the Necessity of Individual Conditions
In fsQCA, a condition is always present when the outcome occurs, then that condition is necessary for the outcome [31]. The purpose of the necessity condition test is to analyze whether a single variable is necessary to explain the occurrence of a particular state of the outcome variable. If a variable is indeed a necessary condition, it shows that the presence of that variable has a necessary role in the emergence of the outcome variable in a particular state, and it is a core element. The core elements indicate a strong causal relationship with the outcome [39].
The calibrated sample data were imported into fsQCA 3.0 for the necessary analysis of individual conditions, and the results are shown in Table 3. Consistency is an important measure of necessity, and when the level of consistency is greater than 0.9, the condition is considered necessary for the outcome [31, 37].
From Table 3, we can see that, in the consistency analysis, the consistency of Government data management authority (Administrator) is greater than 0.9, which satisfies the consistency requirement and can indicate that “ Government data management authority” is a core condition to promote the market-oriented use of local government open data.
3.6. The Configuration Analysis
The configuration analysis attempts to reveal the sufficiency analysis of different configurations of multiple conditions leading to results, in terms of set theory, that is, to explore whether the set represented by the configuration of multiple conditions is a subset of the set of results.
Consistency is also used to measure the adequacy of a configuration, but the minimum acceptable criteria and calculation methods differ from those of the analysis of the necessary conditions. It is generally accepted that the level of consistency for determining adequacy must not be lower than 0.75 [31]. Different consistency thresholds have been used depending on the research context of the body, with some scholars taking values of 0.75 [40], 0.8 [39, 41], etc. As for the determination of the frequency threshold, it needs to be based on the sample size [37], and for small and medium samples, a frequency threshold of 1 is sufficient [42], while for large samples, the frequency threshold should be greater than 1 [43]. In this study, a total of 16 provinces’ sample data were obtained through collation and screening, and in the determination of consistency threshold and frequency threshold, the consistency threshold of this paper was finally determined to be 0.8 by default, and the frequency threshold was 1.
The fsQCA software obtained three solutions after running the data, namely, parsimonious solution, complex solution, and intermediate solution. The intermediate solution is more parsimonious than the complex solution and more reliable than the parsimonious solution [44], so the intermediate solution is the optimal solution.
After running the fsQCA 3.0, the solution consistency of the intermediate solution is 0.93207, which indicates that this result is reliable. Consistent with existing studies, the intermediate solution is reported here [39], supplemented by the parsimonious solution, following the form of Fiss in presenting the results, with solid circles indicating the presence of the condition, circles with forks indicating the absence of the condition, and spaces indicating an ambiguous state; i.e., the condition can be present or absent. “” (the big black circles) indicates the core condition (the core condition is the condition that exists in both the simplex and intermediate solutions), “” indicates the absence of the core condition, and “” (the small black circles) indicates the peripheral conditions. “” indicates the absence of the peripheral conditions, “Blank space” means the condition could be present or absent, and small circles are the auxiliary conditions (conditions that exist only in intermediate solution). In addition, groups with the same core conditions (2A, 2B, and 2C) are grouped and arranged from left to right according to the level of consistency of the groups. Coverage is an important measure of empirical relevance in QCA studies, and raw coverage indicates the extent to which each grouping covers the outcome cases, i.e., the proportion of cases with affiliation in their respective paths, showing the importance of the grouping on the outcome or the empirical tangency [38], similar to R2 in regression [39], and the reported results are shown in Table 4.
The four configurations are presented in the table, where the individual configuration and solution consistency levels are above 0.8, with the solution consistency of 0.93207. The solution coverage is 0.699052, indicating that the solution coverage is 0.699052, indicating that these four configurations can explain 69.9% of the government open data market, which is the same as the QCA study in the field of organization and management. The four configurations in Table 4 can be considered as a sufficient combination of conditions to promote the use of the open data market in the Chinese government.
4. Configurations Analysis and Interpretation
This chapter provides an in-depth interpretation of the above four configurations to actual cases. According to their different combinations, this paper categorizes the paths as technology-led, management-resource-led, economic and technology dual-driven, and location-advantage-led. Three conclusions are then drawn from the above.
4.1. Technology-Led
Technology-led: although this model is not given sufficient support at the organizational and environmental levels, its data utilization of government data is still dominated at the upper-middle level by its superior digital platform technology base. In configuration 1 (∼Human∗Platform∗Administrator∗∼Business∗∼Location), “Local Government Open Platform (Platform)” is the core condition, “Government Data Management Agency (Administrator)” is the peripheral condition, and “Location” is the core condition of absence. The cases covered are Guangxi Zhuang Autonomous Region (0.71, 0.47), Fujian Province (0.56, 0.57). Although Guangxi Zhuang Autonomous Region is not located in China’s national big data comprehensive test area, there is no special big data trading system, and the number of information transmission computer service practitioners is relatively small. The local government department uploads all government data of the autonomous region to the government data sharing and exchange platform of Guangxi Province Autonomous Region in a unified manner, by optimizing and improving the construction of their government open data platforms, categorizing and summarizing government information resources of government systems by the requirements of national government information system integration and sharing. There are now a total of 240 million pieces of data, 5,464 data catalogs, and 505 departments involved. These data are provided to the society through the Digital Guangxi (government data management agency) collaborative dispatching command center, relying on the autonomous region's public data open platform to provide the development and application interfaces of publicly available government data and guide enterprises and society to make reasonable use of the government’s open data. By promoting the comprehensive utilization of “strip data” and “block data,” Guangxi Province taped the value of data and accelerates the cultivation of the data element market. Simultaneously, they deepen the construction of the public data resources system and improve the public data convergence guarantee mechanism, which has a positive effect on promoting the innovative development of the digital economy and fostering the development of the new digital economy model.
The second representative case is Fujian Province. Fujian Province has opened 37 departments, 2224 data directories, 1.2 billion pieces of data, and 1899 data interfaces through the local government data aggregation and sharing platform to push open directories and open data to the provincial government data aggregation and sharing platform. To activate the “dormant” data, the government open data platform carries out government data aggregation and cleaning, classification, desensitization and declassification, quality management, and authorized opening, so that these data can be transformed into market competitive resources, promote the market-oriented use of data, and realize the value-added reuse of open data. Chen Ronghui, head of Fujian Digital Office, said, “The opening of government data is a major initiative to develop service-oriented government, grow the digital economy and sharing economy, and will produce increasing benefits in reducing the cost of obtaining data for enterprises, increasing the enthusiasm of data utilization, and stimulating the vitality of social data utilization.” Combining the above theoretical inquiry and case analysis, we could derive Proposition 1: in the absence of locational advantages, the construction of local government open platforms is crucial to the market-oriented utilization of government open data.
4.2. Management-Resource-Led
Management-resource-led: the relevant condition variables at the organizational, technological, and environmental dimensions drive the high level of provincial government data utilization. These provinces are economically developed, have a high R&D intensity, and value the role of science and technology as breakthroughs and thus have a high technological base; at the same time, provincial governments also recognize the role of the organizational level in supporting the development of open government data and have given sufficient support in terms of human and financial resources and institutional setup. Also, these provinces are strategically located in the national big data comprehensive pilot zones, with a large number of data sources. As a result, they have a high level of big data development, driven by a combination of factors. Examples are Beijing and Shanghai. In configuration 2B (Human∗Administrator∗Business∗Location∗Economic), “information transmission, computer service and software industry employees (Human)” and “government data management agency (Administrator)” are core conditions, “Big data transaction system (Business),” “Location,” and “Digital Economy Development Level (Economic)” could be regarded as peripheral conditions. It covered 2 cases, Beijing (0.93, 0.86) and Shanghai (0.8, 0.83). The core and peripheral conditions of configuration2B are complete relatively. A complete ecological framework for the market-oriented use of government open data has been formed initially. In terms of the political and economic environment, Beijing is the political center of China, and Shanghai is the economic center of China, respectively. Both cities have the top universities, research centers, technology enterprises, and a lot of researchers in China and have certain research and utilization needs for government open data. Also, Beijing and Shanghai have specialized big data management agencies, namely, Beijing Big Data Administration and Shanghai Big Data Center. Next is the introduction of peripheral conditions. In terms of location, Beijing belongs to the Beijing-Tianjin-Hebei cross-regional class comprehensive pilot area, and Shanghai is also a regional demonstration class comprehensive pilot area, both located in China's national big data comprehensive pilot area. The national-level support provides policy support for the development and utilization of local government open data. Secondly, both Beijing and Shanghai have big data trading systems. Under the guidance of the Beijing Municipal Commission of Economy and Informatization, the Big Data Trading Service Platform is constructed and operated by Beijing Software and Information Service Exchange, which is a standardized public welfare Big Data Trading Service Platform with credibility. The services provided by the platform include data transaction service, data validation and valuation service, transaction guarantee service, filing service of validation and transaction, O2O supply and demand matching service, and data financial service. In terms of big data of government affairs, the platform provides services such as data open business rules release, big data open information release, data open docking method guide, enterprise demand service support, social big data sharing support, and technological support for opening big data of government affairs. The objects that can be traded on the platform include four types of data-related products: data products, data tools/technologies, data services, and data talents (Website: http://www.shujushichang.com). Shanghai Data Exchange is developed by the cooperation between government and enterprises. It is a state-controlled mixed ownership enterprise approved by Shanghai Municipal People’s Government and jointly approved by the Shanghai Municipal Commission of Economy and Information Technology and the Shanghai Municipal Commission of Commerce (website: http://www.chinadep.com). Its business scope is to carry out market-oriented utilization and operation of data. Thirdly, in the Local Digital Economy Development Level Index, Beijing is in the second place in the national ranking with an index of 55.0, and Shanghai is in the fifth place with an index of 45.5. Both cities are leading cities in digital economy development, and the solid local industrial foundation provides a broad space for the development and integration of digital technology with the real economy.
4.3. Economic and Technology Dual-Driven
Economic and technology dual-driven: these provinces are not regionally located in the national big data comprehensive pilot zones and do not have a big data trading system. However, the readiness of the local government’s open data platform and the high level of development of the digital economy make the quality of its open government data high, which facilitates market-based operation and utilization. Representative cases are Zhejiang Province and Guangdong Province. In configuration 2A (Human∗Platform∗Administrator∗Economic), the “information transmission, computer service and software industry employees (Human)” and “government data management agency (Administrator)” are the core conditions, and “local government open platform (Platform)” and “Digital Economy Development Level (Economic)” are the peripheral conditions. There are 5 representative cases: Shanghai (0.8, 0.83), Guangdong Province (0.71, 0.87), Zhejiang Province (0.61, 1), Shandong Province (0.51, 0.88), and Sichuan Province (0.51, 0.69). The development of the digital economy has been in the stage of upgrading to an intelligent stage with big data-driven as the main feature. In configuration2A, in addition to having the core conditions, the representative case provinces could be seen to have a relatively high level of local digital economy development. The famous e-commerce Group Alibaba is located in Hangzhou City, Zhejiang Province. Zhejiang Province has become a leader and frontrunner in the development of e-commerce in the country by grasping the development opportunities of e-commerce through commercialization and innovation of information technology. Zhejiang Province ranked first in the country in terms of both government’s open data readiness and data layer in the report of the 2020 Second Half of China Open Data Index. Zhejiang Province launched the first provincial public data opening approach in China, the Interim Measures for the Open and Safe Management of Public Data in Zhejiang Province, in June 2020, to clarify the principle that public data platforms should be open as much as possible in the form of government regulations. This policy puts forward the requirement of data classification and grading opening to build a mechanism of data authorization opening and desensitization processing. At the same time, the government of Zhejiang Province and the Big Data Administration have made clear provisions on the use of public open data, requiring relevant subjects to sign the “public data open utilization agreement,” which stipulates the use of data, and the safety responsibilities to be observed, as well as the corresponding security measures. The content of the local government open data platform has been constantly updated and expanded, with 19,103 data sets, including 9,686 API interfaces, 95,193 data items, and 5780,596,900 pieces of data now open. These measures and regulations make clear provisions for the government's public data management and service departments, the open subjects of public data, and the subjects of public data utilization in the use of public open data. At the same time, the above measures require the relevant subjects to sign the “public data open use agreement,” which stipulates their use of data, the security responsibilities to be observed, and the corresponding safeguard measures.
Guangdong Province was ranked fifth in the report of the 2020 Second Half of China Open Data Index. As a strong economic province on the eastern coast of China, Guangdong ranks among the top in economic development and has strong economic strength, which provides strong financial support for government data openness. In addition, Guangdong Province also has abundant physical resources. On the one hand, Guangdong Province's open data platform “Data Guangdong” is sponsored by the General Office of Guangdong Provincial People’s Government, which has an excellent performance in data discovery, data guidance, and data access, results in submission display and interactive feedback, and has strong platform construction strength due to its outstanding user experience performance. Guangdong Province is the benchmark province in the report of the 2020 Second Half of China Open Data Index. By the end of 2020, the Data Guangdong platform has opened 159.4 million pieces of data, provided a data catalog containing 53departments, and formed 102 application results. Users who need data can complete data application, data correction, and feedback online through the platform, which is the most convenient provincial data open platform for individual registration in China.
4.4. Location-Advantage-Led
Location-advantage-led: location advantage means the comprehensive resource advantage of the location. The locational advantage of a region can be determined by natural resources, labor, industrial agglomeration, geographical location, transportation, etc. The representative case is a province with an unsatisfactory level of economic and resource allocation, but with location advantage, it still has a high performance in the use of open data. The representative case is Henan Province. In Configuration 2C (Human∗∼Platform∗Administrator∗∼Business∗Location∗∼Economic), “information transmission, computer service and software industry employees (Human)” and “government data management agency (Administrator)” are the core conditions. “Location is the peripheral condition,” “Local government open platform (Platform),” “Big data transaction system (Business),” and “Digital economy development level (Economic)” are the peripheral conditions of absence. Henan Province is in 13th place in the report of the 2020 Second Half of China Open Data Index, and the data layer index of the government open platform is not at the top. Henan Province is located in North China and is a largely agricultural province in China with a low level of digital economy development. Henan Province launched a dedicated Henan Big Data Administration website (https://dsj.henan.gov.cn) on December 31, 2019. Although lacking some supporting conditions, Henan Province is the most populous province in China, with a large population base and a large number of employees engaged in information transmission, computer services, and software. Henan Province takes the National Big Data (Henan) Comprehensive Pilot Zone as the starting point and uses 5G to promote the deep integration of big data and the real economy. Zhengzhou Aerodrome Economic Comprehensive Experimental Zone, located in the capital of Henan Province, has collected a huge amount of big data including aviation, railroad, highway, and port information, providing a solid data foundation and momentum for the rapid development of the digital economy in the aerodrome of Henan Province. Currently, 274 enterprises have used identified as big data enterprises, and these enterprises have laid the industry foundation for data development and utilization. Meanwhile, in terms of the setting of government data management authorities, the Henan Provincial Public Data Open Platform is supervised and managed by the General Office of Henan Provincial People's Government and the Development and Reform Commission of Henan Province. Henan Province public data platform is online with 47 departments, 20 fields involved, 3,567,200 data items, 806 data sets, and 1609 API interfaces, with excellent performance in data discovery, data guidance, and data access and results in display and interactive feedback. It has a good user experience. Compared with provinces such as Guangdong, which are driven by both economic and technological tracks, Henan Province has generated good data opening performance despite its unsatisfactory economic and organizational resource allocation level with its advantageous location, rich Internet infrastructure resources, and good data opening platform construction level.
Combining the case studies of “management-resource-based,” “economic and technology-driven,” and “location-driven,” we could derive Proposition 2: the market-oriented use of open government data by information transfer computer service and software industry practitioners and government data management agencies is crucial.
To facilitate comparison and understanding, a summary of the factors influencing the operation of the government open data market is presented based on the above analysis (Table 5).
4.5. Alternative Paths for Market-Based Development of Government Open Data
After comparing the four paths, we can see that the “management-resource-based” development path is undoubtedly the better path to promote the open market use of local governments. Due to the relative differences between the central region, western region, coastal region, and inland of China, some provinces do not have location advantages and are in the nonnational big data comprehensive pilot zone. At the same time, these provinces are located inland, with fewer universities and technology enterprises. Compared with the areas of provinces with richer economic research, there is a significant shortage, resulting in the number of employees in the information transmission, computer services, and software industry (Human) less, and the main force of data development is insufficient. As seen in Configuration 1, if the development level of government open data platform in the region is high, even in the absence of the core conditions of “number of employees in information transmission, computer services, and software industry (Human)” and “government data management agency (Administrator)” as in Configurations 2A, 2B, and 2C, Configuration 1 is an alternative path to promote the development of government open data (Administrator). Configuration 1 is an alternative path to promote the development of open government data. Therefore, provinces and regions in the same situation can refer to the development approach of Path 1, starting with optimizing and improving the government open data platform (Platform), and doing the basic work of data development and utilization, to promote the effective use of open government data in the market. Combining the above findings, we could derive Proposition 3: under certain conditions, when the number of data development majors is small, and there is no dedicated government data management agency, optimizing and improving the government open data platform could be an alternative path to promote the effective use of government open data in the market.
5. Analyses for the Robustness
This paper conducts robustness tests on the antecedent grouping of factors influencing the market operation of government open data. Robustness tests were conducted using adjusted consistency levels, increasing the consistency level from 0.8 to 0.85 and conducting the histories again, with the frequencies held constant [37]. Compared with the results before the improvement, the overall consistency improved from 0.8 to 0.85, which was still higher than the minimum acceptable consistency standard, and the overall coverage was still 0.699052, with an overall consistency of 0.932069 (the result in the previous paper was 0.93207). The findings of this paper are not substantially changed after increasing the consistency threshold.
6. Conclusion and Discussion
6.1. Conclusion
Data-driven government services have improved China’s government services and given the nation a sense of happiness and security during the epidemic. This paper uses fuzzy qualitative comparative analysis and the TOE theoretical framework to study the “combination effects” of six factors on the market-oriented use of government open data: the number of employees in the information transmission computer service and software industry, government data management agencies, government open data platforms, big data trading systems, geographic location, and the level of digital economy development at three levels, using group thinking.
The theoretical contributions are as follows. The paper proposes an analysis of the pathways to open data utilization by local governments in China based on the TOE framework using the fsQCA approach. It may enrich the case of fsQCA empirical analysis to a certain extent and facilitate the application and expansion of qualitative comparative analysis in e-government theory. Through theoretical and empirical analysis, the organizational dimension of government data management authority could be seen to have a greater impact on the utilization of government open data in the necessary analysis and the grouping analysis than the technical and environmental factors. At last, by drawing on the TOE theory, this paper provided a more standardized theoretical perspective on the complex mechanism of market-oriented government open data utilization and operation. It also enriched the practical cases of the TOE framework.
The practical findings of this paper are as follows: firstly, there are geographical differences in the marketization of government open data. Beijing, Shanghai, and Guangzhou have been exemplary frontrunners in the development of government open data, with competent institutions, supporting facilities, R&D personnel, and market environment due to other regions. Inland areas, although not as advantageous as the regions in the north, still can be developed in other ways for the marketization of government open data. Under certain conditions, when the number of data development majors is small, and there is no dedicated government data management agency, optimizing and improving the government open data platform could be an alternative path to promote the effective use of government open data in the market. Secondly, the role of the government open platform is mostly used to collect and organize data, most of the information is still in a “dormant” state, and the analysis and processing of these data and information still need professional big data analysis experts and teams to complete. In practice, the implementation of open data around the world is mostly “outsourced,” relying on the development and operation teams of software platforms, while the data governance of government departments is in a state of absence, lacking specialized governance teams and operable technological frameworks, resulting in low quality and interoperability of metadata. Thirdly, the current data property rights are not clear, and the regulation system is not sound. Some provinces have prominent information barriers, weak data security management, and an imperfect industrial ecosystem, making some useful data unable to connect with the market, resulting in the waste of data.
6.2. Limitation and Prospects
This paper has some shortcomings inevitably: the 2020 Second Half of China Open Data Index Report published by DMG Lab Fudan University (http://ifopendata.fudan.edu.cn/) is available for a limited number of provinces. Some of the platforms are inaccessible for unknown reasons, or the platforms are online but not open data, and only 16 provinces (municipalities directly under the central government) have rating indices. This is one of the reasons why fsQCA had to be used in this study. Therefore, the sample data selected for this paper is limited to cases that are available on Digital Forest, and data for provinces that cannot be identified are not included. Corresponding analyses could be continued in this area later to further enrich the research results.
Data Availability
All data used to support the findings of the study are included within the article.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This study was sponsored by Ministry of Science and Technology of the People’s Republic of China (no. B1190800). Name of the funding: 2021-2035 National mid- and long-term science and technology development plan to carry out research on major issues.