China’s Policy Environment’s Development and Path from the Perspective of Policy Sustainability: A Visual Analysis Based on CNKI and WOS

Abstract

In this paper, we examine China’s policy environment from the perspective of policy sustainability. Through our use of the CiteSpace and NVivo software, the literature on the policy environment is visualized using a knowledge graph analysis and a content analysis of highly cited papers. The analysis shows that: the number of published papers on “policy environment” is on the rise, and the research prospect is broad. There is a lack of communication and cooperation between domestic authors and a close relationship between international authors. The main research countries included the USA, Switzerland, Finland, Ireland, and Austria, etc. Domestic research focuses on vehicles for high-quality economic and social development, while international research focuses on theoretical model discussion. The domestic optimization path focuses on the transformation from “Sending blood” (“ShuXieShi”) to “Making blood” (“ZaoXieShi”), while the international research focuses on the main line of “public health and environmental protection”. Domestic research focuses on fostering a sound business environment, while international research focuses on the public, political and professional aspects of health and medical policy environment. On the whole, in the future, we should improve the adaptability of the policy environment, optimize the policy implementation environment, and establish the interaction mechanism of policy environment optimization to achieve policy optimization.

1. Introduction

The “policy environment” generally refers to the sum of all the elements of conditions that affect policy. It not only provides a theoretical perspective and a theoretical model, but also provides a practical condition, which runs through the whole lifecycle of a policy. In-depth analysis of the environment that a policy faces or will face is of great significance for promoting policy reform and making policies more forward-looking, scientific and effective. Research on the policy environment is conducive to enriching policy science and promoting the optimization and improvement of the policy environment. Global human and social environments vary greatly, and the policy environment varies from country to region. Overall, China’s current policies are numerous, the policy environment is complex and different policies are rooted in different environment backgrounds. The implementation of a policy will create new ecological policies, based on specific policy research on the policy environment, or from the perspective of the policy environment regarding current policy.

In recent years, Chinese scholars have focused on key issues. Li Linzi et al. (2022) carried out a stakeholder analysis through a policy environmental impact assessment [1]. Li Yongning et al. (2022) analyzed the theoretical evidence formation and improvement path through a policy environmental impact assessment [2]. Cao Li et al. (2022) evaluated the environmental effects of ecological transfer payment policy based on quasi-natural experiments [3]. Liao Qi et al. (2022) analyzed the urban policy environment for promoting youth development [4]. Gu Yichun et al. (2022) analyzed the policy environment for the development of healthcare big data in China based on the PMC index model [5]. Yuan Xin et al. (2022) studied the longevity dividend in China through population opportunities, the policy environment and the development direction [6]. He Da et al. (2022) analyzed the policy environment and the current situation in respect of the construction of regional medical centers in Shanghai [7]. Zheng Liping et al. (2022) discussed the monetary policy environment and the choice of enterprise investment behavior from the perspective of credit term structure [8]. Feng Qingyu et al. (2022) studied the application of the agricultural policy environment extension model in respect of agricultural non-point-source pollution in China [9]. Ruan Rongping et al. (2022) analyzed rural poverty alleviation enterprises during the COVID-19 pandemic based on survey data for 1269 rural poverty alleviation enterprises in China; production and operation, poverty alleviation drive and the policy environment were analyzed [10]. Huang Shaoqing et al. (2022) discussed the concept, influence mechanism and policy environment of a contract-intensive economy [11]. Yu Lihua (2022) reviewed the thinking and strategy regarding DRG implementation in medical institutions in the context of medical reform policy [12]. Xu Zhijun (2021) introduced the development characteristics, experience and enlightenment of the leisure sports policy environment in developed countries [13]. Zhang Tieming et al. (2021) analyzed the policy environment for the high-quality development of science and technology journals in Chinese universities [14]. An Xiaoling et al. (2021) evaluated and analyzed the risk identification and assessment of innovative drug clinical trials in the policy environment [15]. Guan Haoyuan et al. (2021) conducted a comparative study on the development strategy and policy environment in respect of artificial intelligence in China and Europe in the new period [16]. Wu Yunyong et al. (2021) analyzed the impact of the policy environment on China’s education development and the overall requirements of future reform [17]. Peng Ruxia et al. (2021) analyzed the impact of the innovation policy environment on the location choices of foreign direct investment through case studies of six cities in the core region of the Pearl River Delta [18]. Zeng Ming et al. (2020) studied the business models and development strategies of energy storage power stations from the perspective of the policy environment [19]. Zhou Jingjing et al. (2020) discussed the transformation of the rural micro-governance structure under the new policy environment [20]. Li Qiushi et al. (2020) studied the construction of a policy environment to promote China’s open and scientific development [21]. Jin Chun (2020) studied the impact of fiscal decentralization on environmental pollution based on the threshold effect of the policy environment [22]. The practical work of promoting the policy environment must be based on the implementation standards of the policy environment. However, in terms of the development level and standard system construction of policy environment implementation, there are not many relevant dissertations, mainly including the following literature: ‘Evaluation of Urban Policy Environment for Promoting Youth Development’ [23], ‘Comparison of Science and Technology Policy Environment in Provinces Along the Silk Road under the Dual Cycle Pattern: A Case Study of Shaanxi, Sichuan and Chongqing’ [24], ‘Evaluation of Smart City Development Potential Based on Entropy Weight and Cloud Model: A Case Study of 13 Cities in East China’ [25], ‘Research on the Impact of State-owned Equity on Private Property Rights Protection Level’ [26], ‘Research on the Optimization of Entrepreneurial Policy Environment for University Students Based on Publicity Perspective’ [27], and ‘Research on the Introduction and Service Policy Coordination of Overseas High-level Talents in China’ [28].

In summary, there is an abundance of research regarding the theoretical analysis framework interpretation of the policy environment and current policy environment evaluation and optimization. However, there is still a lack of comprehensive sorting and systematic discussion of macro-policy-environment research, which needs to be studied and strengthened. Therefore, the analysis of China’s policy environment research in this paper is of value.

To sum up, the specific objectives of this paper are as follows: based on the basic concept of “policy environment”, this paper systematically sorts out and analyzes the key literature on policy environment included in CNKI and WOS, strives to create a visual analysis of the research status of “policy environment”, and compares and analyzes the differences, theme distribution and emotional trend of Chinese and foreign scholars on “policy environment”. From the perspective of text mining, this paper analyzes the differences in the development paths of “policy environment” between China and foreign countries, and tries to present the overall form of research and development of “policy environment”. Through the overall presentation of the research and development of “policy environment”, this paper hopes to promote the theoretical research on policy environment and promote the development and maturity of policy environment theory.

2. Research Design

2.1. Research Methodology

In this paper, CiteSpace software was used for quantitative analysis and atlas analysis of the literature [29]. CiteSpace5.8R3 (http://bi.zbrj88.cn/Citespace/1.html, accessed on 4 November 2022) was used to draw the knowledge map, and the hot spots, frontiers, core authors and cooperative networks of policy environment research were visually analyzed. There has been high recognition of this research paradigm in China [30,31]. CiteSpace articles are often published in foreign journals. For example, see [32,33,34]. In this study, NVivo 12 Plus (http://www.strong-vinda.com/, accessed on 4 November 2022) was used for content analysis, targeting highly cited papers on the subject of policy environment research, and we aimed to determine the basic value, main framework and main content in respect of the policy environment [35]. Foreign journals have also published articles on NVivo as research software, for example, see [36,37,38]. In general, CiteSpace software and Nvivo12plus software can not only carry out quantitative analysis of literature, but also realize visual presentation of analysis results. Therefore, we choose these two software types for the study.

2.2. Data Sources

In order to expand the breadth of research as much as possible while ensuring the accuracy of research, SSCI and SCI databases were selected for international research, and CSSCI and Peking University core databases were selected for comparative analysis of the policy environment for domestic research. In terms of domestic research, CSSCI and PKU core index journals can show the highest level and most cutting-edge research results in China. According to the search type “(title = policy environment)” in the CNKI database, the source types were set as “CSSCI” and “Peking University core”. A total of 2533 related articles were retrieved (2 February 2022). The retrieval results were artificially filtered, excluding essay notices, book reviews, news reports, meeting notices, names of researchers leading the academic research, repeated articles and clearly unrelated literature. In total this resulted in 2444 records of domestic research for analysis of sample data. The data were downloaded and transformed into a data format that could be processed using CiteSpace. Advanced retrieval was conducted in the Web of Science core collection database according to the search term “(TS = (“policy environment”)) NOT TS = (“environment policy”) AND DT = Article”. The database types were SSCI and SCI. A total of 1331 related articles were retrieved (24 August 2022).

2.3. Study Process

The research process of this paper mainly includes three steps: first, download the software and install and debug it. This section describes how to download CiteSpace and Nvivo12plus and install and debug them. Second, search, download research data and import software. This includes searching and downloading research data from CNKI and WOS, and importing them into the software. Thirdly, analyze literature and write papers. This mainly includes data analysis, chart making, text expression, theory extraction and literature annotation. The entire process can be represented in Figure 1.

3. Visual Analysis

3.1. Post Analysis

Article analysis involves analyzing the quantity, trend and composition of published literature in a research field from different dimensions. The analysis of published articles in a research field can help researchers grasp the total number of published articles in the research field and the annual number of published articles, and predict the trend of published articles.

Based on the number of articles published in the CNKI and WOS databases, a trend chart of articles published in the research field was drawn (Figure 2). From Figure 2, it can be seen that in terms of the earliest published time and the total number of published articles, the published domestic policy and environment literature started in 1992. From 1992 to the retrieval date, 2533 related articles were published in the CNKI database. Internationally published policy and environment literature started in 2004, and from 2004 to the retrieval date, 1331 related articles were published in the WOS database. From the overall publication trend, the number of international and domestic policy environment publications showed an upward trend. In terms of the annual number of articles published, the average annual number of domestic articles published was 81.71, and the average annual number of international articles published was 70.05. In general, both at home and abroad, great importance is attached to the policy environment. From the trend of publication, the development speed of the international policy environment is faster than that of the domestic one. On the whole, policy environment research is very promising.

3.2. Analysis of Research Subjects and Cooperation Networks

Progress in policy environment research cannot be achieved without the unremitting efforts of relevant researchers and teams. Analysis of the structural characteristics of research authors and their cooperative networks can reflect the core authors and their cooperative relationships in this field.

“Core authors” refers to scholars with higher academic levels and more scientific research achievements in a certain field. The analysis of core authors enables understanding of the research status and progress in a field. According to Price’s law, the number of core authors can be calculated as follows:

$$\mathrm{MP}=0.749\sqrt{NP\max }.$$

MP represents the minimum number of papers published by the core authors, and NPmax represents the cumulative number of papers published by the authors with the most papers in the research time-interval. If the number of stable core authors accounts for 50% of the total number of papers, it is considered that the field has formed a core author group.

In China, the retrieval information in respect of 2444 articles was imported into CiteSpace for visual analysis of author status, and it was found that the scholar Zhanfeng Dong published the most articles (17 articles in total). According to the calculation formula, the MP value was 3.09; that is, an author with four or more published articles can be regarded as the core author in the field of domestic policy and environment research. As shown in

Table 1. Comparison of the number of articles published by research authors (Top 10).

Serial Number	CNKI			WOS
	Author	Count	Year	Author	Count	Year
1	Zhanfeng Dong	17	2010	Ziming Xuan	5	2015
2	Jiali He	16	1992	Anne Marie Thow	5	2017
3	Yanguo Xu	9	2010	Alex Collie	4	2014
4	Yuanjing Wang	8	2001	Pauline Zardo	4	2014
5	Chazhong Ge	7	2019	Shahadat Uddin	4	2017
6	Ting Liu	7	1992	Agnes Erzse	4	2021
7	Rong Shao	4	2017	Timothy S Naimi	4	2015
8	Feng Long	4	2019	Kathryn Quissell	3	2016
9	Jianjun Xu	3	2010	Hans Peter Schmitz	3	2016
10	Jinhua Cheng	3	2021	Ayse Bugra	3	2007

, there are eight core authors in respect of policy environment research: Zhanfeng Dong, Jiali He, Yanguo Xu, Yuanjing Wang, Chazhong Ge, Ting Liu Rong Shao, Feng Long, Jianjun Xu and Jinhua Cheng. In general, in terms of the number of articles published, the field of policy environment research has outstanding authors represented by Zhanfeng Dong, Jiali He, Yanguo Xu, Yuanjing Wang, and the number of core authors published is higher than that of other authors. Zhanfeng Dong has a total of 17 articles published, followed by Jiali He with 16, Yanguo Xu with 9 and Yuanjing Wang with 8. Of these authors, Jiali He and Ting Liu are the two core authors focusing on the policy environment research that started in 1992. Research over 30 years shows that the four core authors are continuously paying attention to the policy environment, and have made a series of achievements. Yanguo Xu, Zhanfeng Dong and Yuanjing Wang started to pay attention to the policy environment after 2000, while Chazhong Ge, Rong Shao and Feng Long started to pay attention to the policy environment in 2017, indicating that Chazhong Ge, Rong Shao and Feng Long are authors who began to pay attention to the policy environment in recent years and have produced high-quality research. According to the statistics of the literature authors, the core authors published 72 articles in total, which accounted for 17.06% of the total literature in respect of policy environment research. This was lower than the standard of the core authors, indicating that the core authors of domestic policy environment research have not yet been formed.

From the international perspective, the retrieval information in respect of 1331 articles was imported into CiteSpace for visualization analysis of the author status, and it was found that Ziming Xuan, the author of the most published literature in WOS related to the policy environment, published only five articles. According to the calculation formula, the MP value was 1.67. That is, authors with two or more published articles can be regarded as core authors in the field of international policy environment research. There are 77 core authors in the field of international policy and environment research, and Table 1 lists only the top 10. The total number of core authors is 177, accounting for 50.14% of the total number of international policy and environment studies; this is higher than the standard of the core authors, indicating that the core authors of international policy and environment studies have been initially formed.

CiteSpace was used to obtain the author cooperation network map (Figure 3 and Figure 4), in which the tree-ring nodes represent the author, the node size represents the centrality, and the node connection represents the existence of a cooperative relationship. It can be seen from the analysis that, in the domestic aspect (Figure 3), the number of network nodes N = 231 and the number of connections E = 73. The network density D = 0.0027, indicating that the distribution is relatively scattered. From the international perspective (Figure 4), the number of network nodes N = 253, the number of connections E = 97, and the network density D = 0.003, indicating that the distribution is relatively scattered and the research network has not yet been formed. In general, there is a certain degree of cooperation among authors, but the concentration is still low and the cooperation relationship is relatively loose. To be specific, the authors in the policy environment research field show the characteristics of large dispersion and small aggregation. Most of the authors are in the state of independent research, only a few of the authors have established cooperative relations, and the cooperation network has not yet been formed.

In order to explore the cooperation of research institutions in the field of policy and environment research, a co-occurrence analysis was conducted on the institutions cited in the sample literature, and a cooperation network map of institutions was obtained (Figure 5 and Figure 6). The size of annual rings was proportional to the number of publications, and the connection and thickness between nodes represented the cooperation relationship and cooperation frequency between institutions, respectively. As shown in Figure 5, the number of nodes in the cooperation network of domestic research institutions is N = 202, the number of connections is E = 35 and the network density is D = 0.0017. In other words, 202 institutions and 35 links between institutions were selected from the co-occurrence knowledge graph in the field of policy and environment research in China. Institutions were identified (

Table 2. Comparison of numbers of papers published by research institutions (Top 10).

CNKI			WOS
Institution	Count	Year	Institution	Count	Year
Central South University Business School	103	1992	Univ Melbourne	19	2010
Shanghai Ocean University	20	1992	London Sch Hyg & Trop Med	18	2008
Shanghai Institute of Electrical Engineering	17	1992	Univ Sydney	18	2012
Chinese Academy of Environmental Sciences	14	1992	Univ CalifSan Francisco	14	2014
Environmental Planning Institute of Ministry of Ecology and Environment	14	2019	Boston Univ	13	2015
School of Economics and Management, Harbin Engineering University	12	2007	Harvard Univ	12	2007
Environmental Planning Institute of Ministry of Environmental Protection	9	2012	Univ Toronto	10	2007
Institute of Strategic Consulting, Chinese Academy of Sciences	7	2018	Univ Oxford	8	2008
University of Chinese Academy of Sciences	7	2017	Univ Pretoria	8	2015
School of Economics, Jinan University	6	2014	Monash Univ	8	2011

) in terms of the number posts for the Central South University Business School (103 articles), followed by the Shanghai Ocean University (20 articles) and Shanghai Institute of Electrical Engineering (17 articles). As can be seen from Figure 5, the number of links is less than the number of nodes, and the density is relatively low, which indicates that there is a lack of close cooperation between institutions that have made important contributions to this field in China, and the degree of communication is relatively low. In the future, all institutions should strengthen cooperation, exchange theoretical frontiers, and promote research and development. As shown in Figure 6, the number of nodes in the cooperation network of international research institutions is N = 257, the number of connections is E = 177, and the network density is D = 0.0054. In other words, a total of 257 institutions and 177 connections between institutions were selected from the co-occurrence knowledge graph in the field of international policy and environment research. In terms of the number of articles published (as shown in Table 2), the University of Melbourne had 19 articles published, followed by the London School of Hygiene & Tropical Medicine and the University of Sydney, both with 18 articles published. Again, for the University of San Francisco, California, 14 articles were published. On the whole, the number of articles published by international influential institutions is less than that of domestic institutions, but the international influential institutions are closely connected with each other and have formed a certain scale of cooperation network.

The analysis of the publication situation, cooperation network and publication trend of different countries in a research field can help researchers to quickly grasp the situation in respect of national attention, cooperation, publication trends and national influence. In order to explore policy research institution cooperation in the field of environmental research, a co-occurrence analysis was carried out on the sample document-cited country, using the national post number and centricity comparison table (

Table 3. Comparison of number of national publications and centrality.

	Count (Need)			Centrality (Need)
Serial Number	Country	Count	Year	Country	Centrality	Year
1	USA	447	2004	USA	0.6	2004
2	England	162	2007	Switzerland	0.54	2008
3	Australia	140	2008	Finland	0.47	2008
4	People’s R. China	127	2010	Ireland	0.31	2008
5	Canada	73	2008	Austria	0.27	2009
6	South Africa	57	2008	France	0.2	2008
7	England	53	2004	People’s R China	0.2	2004
8	Netherlands	41	2008	Indonesia	0.18	2012
9	Germany	31	2009	Nepal	0.17	2016
10	Switzerland	28	2008	Denmark	0.15	2009

) and the national co-operation network graph (Figure 7). The ring size is proportional to the number of node connections between partnerships and cooperations, and the frequency is represented by the thickness of the representative lines.

As shown in Figure 7, the number of nodes in the national cooperative network is N = 135, the number of connections is E = 176 and the network density is D = 0.0195. In other words, a total of 135 countries and 176 links between countries were selected from the co-occurrence knowledge graph in the field of policy and environment research in different countries, indicating that a large number of countries are concerned about policy and environment globally, and there has been relatively close cooperation between countries. In terms of the number of national publications, the top five countries were USA (447 publications), England (16 publications), Australia (140 publications), People’s Republic of China (127 publications) and Canada (73 publications). This indicates that scholars in these countries pay close attention to the policy environment, and a large number of academic papers have been published. From the perspective of national intermediation (heart), the top five countries ranked by strong and weak centrality were the USA (0.6), Switzerland (0.54), Finland (0.47), Ireland (0.31) and Austria (0.27). This indicates that the USA, Switzerland, Finland, Ireland and Austria are countries with relative influence in the field of policy environments. Overall, both the United States and Australia ranked among the top five countries in terms of the number of publications and centrality, indicating that the United States and Australia have great national influence in the field of policy environments. In particular, the United States ranked first in terms of the number of studies and centrality, indicating that the United States has a dominant position in the field of policy environments.

3.3. Analysis of Research Hotspots

3.3.1. Keyword Co-Occurrence Analysis

Keywords are words selected by the author to summarize the topic of the article, comprising the author’s highly summarized and refined academic thoughts, research themes and research contents of a specific research. Therefore, keywords can be used as a means by which to analyze the research topic. At the same time, by examining the frequency of keywords in a field, we can understand the research hotspots in that field and judge the update speed of the research content and the vigor of the research.

Through the analysis of a knowledge graph, the research topics and research hotspots of a certain field can be obtained through the analysis of keywords. In the keyword co-occurrence network picture constructed using CiteSpace, each node represents a keyword, and the size of the node represents the frequency of keywords. We ran CiteSpace and set the node type as Keyword. We set the time ranges as 1992 to 2022 for China and 2004 to 2022 for the rest of the world. We imported the research data and used the “Keyword” function of CiteSpace. A keyword co-occurrence map (Figure 8 and Figure 9) and a keyword frequency and centrality schematic table (

Table 4. Keyword frequency comparison table.

CNKI (Need)				WOS (Need)
Keyword	Count	Centricity	Year	Keyword	Count	Centrality	Year
policy environment	296	0.28	1992	impact	100	0.19	2005
policy	80	0.08	2000	policy	76	0.17	2008
environmental protection	45	0.1	2000	care	47	0.22	2005
environmental regulation	41	0.08	2011	model	45	0.26	2006
environment	37	0.17	2000	health	40	0.09	2005
influencing factors	35	0.08	2011	consumption	40	0.21	2010
monetary policy	30	0.01	2003	United States	36	0	2006
public policy	29	0.36	2002	management	33	0.17	2010
small and medium-sized enterprises	28	0.39	1999	governance	33	0.16	2010
policy suggestion	21	0.12	2003	politics	31	0.1	2008

) were obtained. As can be seen from the figures, the map has a total of 240 domestic nodes, E = 185 lines, network density = 0.0065 and N = 202 international nodes. E = 249 lines, network density = 0.0123. In terms of the domestic results, as shown in Table 4, the top keywords are policy environment (296), policy (80), environmental protection (45), environmental regulation (41), environment (37), influencing factors (35), monetary policy (30), public policy (29), small and medium-sized enterprises (28) and policy suggestion (21). This indicates that domestic research on the policy environment mainly focuses on how to provide a good business environment for the high-quality development of private economy, especially for small- and medium-sized enterprises. On the international side, as shown in Table 4, the top keywords are impact (100), policy (76), care (47), model (45), health (40), consumption (40), United States (36), management (33), governance (33) and politics (31), indicating that international research on the policy environment mainly focuses on the impact of the policy environment, plus the optimization and governance of the health and nursing policy environment. From the above, it can be seen that research on the policy environment in China places more emphasis on economic development, while international research focuses more on social security and welfare.

3.3.2. Keyword Clustering Analysis

According to the network structure and the clarity of clustering, CiteSpace measures the effect of mapping using two indicators: Q value and S value. The Q value (modularity) represents module value, and its interval is [0, 1]. Q > 0.3 means that the divided community structure is significant. The S value represents the weighted mean silhouette, which means the average contour value. S > 0.5 means the clustering is reasonable and S > 0.7 means the clustering is efficient and convincing (

Table 5. Clustering summary comparison.

Database	Label	Number of Nodes	Outline of the Value	Annual	Keywords
CNKI	0	31	1	2011	Policy environment (111.88, 10−4); policy (15.04, 0.001); environmental regulation (14.19, 0.001); independent innovation (13.01, 0.001); environmental protection (12.5, 0.001)
	1	15	0.954	2005	Monetary policy (43.69, 10−4); China (32.98, 10−4); external environment (21.66, 10−4); environment (16.95, 10−4); policy environment (14.31, 0.001)
	2	14	0.977	2014	Influencing factors (34.86, 10−4); business environment (24.8, 10−4); innovation performance (24.8, 10−4); industrial policy (19.56, 10−4); policy environment (15.34, 10−4)
	3	14	0.965	2016	Environmental regulation (76.18, 10−4); technological innovation (21.97, 1.0 × 10−4); difference-in-difference (15.23, 10−4); policy environment (14.29, 0.001); corporate performance (13.03, 0.001)
	4	13	0.969	2007	Public policy (24.84, 10−4); environmental tax (22.5, 10−4); reform (16.66, 10−4); innovation (12.29, 0.001); environmental regulation (11.08, 0.001)
	5	13	0.939	2009	SMEs (44.41, 10−4); development environment (22.64, 10−4); policy recommendations (17.7, 10−4); entrepreneurial environment (16.66, 10−4); climate change (11.08, 0.001)
	6	12	0.918	2004	Private investment (17.13, 10−4); private investment (17.13, 10−4); private economy (13.69, 0.001); financing channels (11.39, 0.001); investors (11.39, 0.001)
	7	12	0.963	2007	Environmental protection (61.79, 10−4); eco-environment (17.16, 10−4); policy environment (11.54, 0.001); economic policy (10.96, 0.001); environmental monitoring (10.96, 0.001)
	8	10	1	2007	Policy (58.25, 10−4); policy implementation (24.52, 10−4); environmental effects (21.58, 10−4); institutional environment (12.9, 0.001); policy environment (10.34, 0.005)
WOS	0	19	0.884	2012	Having (9.18, 0.005); Policies (9.18, 0.005); The policy design (9.18, 0.005); Aid (5.58, 0.05); The productivity (5.58, 0.05)
	1	17	0.929	2010	The model (24.49, 1.0 × 10−4); Uncertainty (9.76, 0.005); Choice (9.76, 0.005); The strategy (8.29, 0.005); Information (6.88, 0.01)
	2	16	0.941	2014	Nutrition (8.41, 0.005); Iycf (7.18, 0.01); Physical activity (5.59, 0.05); Indicators (5.42, 0.05); Adaptation policy (5.42, 0.05)
	3	16	0.753	2011	Capabilities (8.83, 0.005); The governance (8.83, 0.005); Disasters (8.83, 0.005); Drought (8.83, 0.005); European Union (5.24, 0.05)
	4	16	0.989	2011	USA (13.02, 0.001); Pa (6.49, 0.05); Displaced person (6.49, 0.05); Clinical practice (6.49, 0.05); C 33 (6.49, 0.05)
	5	15	0.889	2010	Newborn (9.95, 0.005); Care (7.15, 0.01); Uganda (6.31, 0.05); Access (6.31, 0.05); Women’s health (4.97, 0.05)
	6	14	0.949	2014	Consumption (16.08, 1.0 × 10−4); Drinking (14.64, 0.001); Alcohol control policy (10.97, 0.001); Substance use (7.29, 0.01); Alcohol policy environment (7.29, 0.01)
	7	14	0.962	2010	Monetary policy (8.6, 0.005); Hospitality stocks (8.6, 0.005); The discount rate (8.6, 0.005); China (6.1, 0.05); The health insurance (5.03, 0.05)

). In the domestic context, as shown in Figure 10, the Q value of the keyword clustering map is 0.8517, so its structure is significant; the S value is 0.9699. In the international context, as shown in Figure 11, the Q value of the keyword clustering map is 0.7488 and the S value is 0.887. This is not only larger than the reasonable average contour value of 0.5, but is also larger than 0.7, indicating that the clustering analysis in this paper is efficient and convincing, and that the clustering results, with significant structure and efficiency, are helpful measures by which we can analyze and grasp the overall characteristics and development trends of policy environment research. As shown in Figure 10, the top nine clusters in China are “#0 policy environment”, “#1 monetary policy”, “#2 influencing factors”, “#3 environmental regulation”, “#4 public policy”, “#5 small and medium-sized enterprises”, “#6 private investment”, “#7 environmental protection” and “#8 policy”. As shown in Figure 11, the top 7 clusters in the international rankings are “#0 human rights”, “#1 model”, “# 2 nutrition”, “#3 capabilities”, “#4 USA”, “#5 newborn”, “#6 consumption” and “#7 monetary policy”.

A keyword cluster summary table was drawn to further explore the number of keywords contained in each cluster, the closeness of the cluster itself, the average year of keyword distribution and the main keywords included in the cluster. As can be seen from the above table, the mean silhouette (S value) of each cluster is greater than 0.7, indicating that each cluster is efficient and convincing. In the domestic context, the label number of the cluster is 0 (policy environment), with 31 nodes. We then have Cluster 1 (monetary policy), with 15 nodes; Cluster 2 (influencing factors), with 14 nodes; Cluster 3 (environmental regulation), with 14 nodes; Cluster 4 (public policy), with 13 nodes; and Cluster 5 (small and medium-sized enterprises), with 13 nodes; Cluster 6 (private investment) and Cluster 7 (environmental protection), each with twelve nodes; Cluster 8 (policy). From the international perspective, we have Cluster 0 (human rights), with 19 nodes. Then, from the most to the least number of nodes in each cluster, we have Cluster 1 (model), with 17 nodes; Cluster 2 (nutrition), Cluster 3 (capabilities) and Cluster 4 (USA), all with 16 nodes; Cluster 5 “Newborn”, with 15 nodes; and Cluster 6 (consumption) and Cluster 7 (monetary policy) each with 14 nodes. The keywords extracted according to the weighting algorithm are listed in each cluster. The top five keywords in each cluster are ranked from left to right in order of their importance.

3.3.3. Keyword Timing Analysis

Timeline Analysis

The CiteSpace software was used to draw the timeline map of keywords in the field of policy environment, and the duration and evolution trend of research hotspots could be obtained, as shown in Figure 12. The keyword evolution graph is arranged chronologically from left to right, and the size of circular nodes in the graph is proportional to the frequency of corresponding keywords. CiteSpace software was used to analyze the evolution of research hotspots in the topic field in the selected literature. In domestic terms, the time parameters were set as follows: time period = 1992–2022, year per slice = 1, g-index (k = 5), node type = keyword. The running results of topic term clustering evolution are shown in Figure 11. A total of 240 nodes and 185 lines were generated in the map. In this case, Q = 0.8517, indicating that the divided structure was significant. Internationally, the time parameters were set as 2004–2022, year per slice = 1, g-index (k = 10), and node type = keyword. The running results of the inscription clustering evolution are shown in Figure 12. A total of 202 nodes and 249 lines were generated in the international map. In this case, Q = 0.7488, indicating that the partition structure was significant.

Domestically, as shown in Figure 12, highlighted keywords on the timeline experienced ups and downs in the larger stage of the timeline and changed from few to fewer, indicating that the development of policy environment research has experienced a steady rise. The steady development and fine development have occurred in three stages, therefore the domestic researchers on the policy environment have experienced ups and downs. In the figure, the node of the keyword policy environment is the largest, while other nodes are not prominent, indicating that the research hotspots in the field of policy environment are not prominent. In addition to policy environment, automobile regulation and policy is an obvious node in the whole field of policy environment research, reflecting the importance of automobile regulation and policy research for the development of policy environment research. The nature of this keyword also determines that it is studied by scholars as a central keyword. Moreover, this keyword is located at the front of the time axis, indicating that the domestic policy environment field began to be explored based on this keyword. Since then, many related studies on policy environment research have been associated with this keyword. This shows that the domestic policy environment research not only interprets the policy environment for the policy as the sum of all conditions, but that it also understands the policy environment in respect of the effects of policy on the environment, especially in terms of major economic policies, which often involve long investment periods, large investments, high resource utilization, high strength, and wide ranges; these factors have a significant impact on the ecological environment. Obviously, auto-regulation and policy are very typical, with the high-speed development of China’s economic and social care laws, complicated regulations and policies related to environmental impacts, the need to coordinate the relationship between automobile industry development and environmental protection, scientific and reasonable methods by which to evaluate the environmental impact of policy on the automobile industry, and improving the positive environmental effect of industry policy. Therefore, automobile regulations and policies have become a hot topic in the research in respect of the policy environment. The right side of the graph shows the keyword cluster label, which is divided into policy environment, monetary policy, influencing factors, public policy, small and medium-sized enterprises, private investment, environmental protection, policy and other clusters in the evolution process. In general, from the perspective of time axis analysis, the keywords related to policy environment are not prominent, the correlation of keywords is increasing year by year, and the keyword clustering is forming, which fully shows that the research on policy environment has attracted the attention of Chinese scholars year by year.

On the international side, as shown in Figure 13, the prominence of keywords in this timeline has experienced a wave-like development process from less to more, then to less, then to more and then to less, indicating that the development of international policy environment research has experienced a gradual development process. Although there are certain fluctuations, the overall development is good. In the figure, health, care, impact, model, politics, policy and consumption are keywords with larger nodes, indicating that these keywords are hotspots in terms of international policy environment research. This reflects the importance of the terms health, care, impact, model, politics, policy and consumption in terms of international policy environment research. The nature of these keywords determines that they are studied by scholars as central keywords. Moreover, there are many links between the above keywords and other keywords, indicating that international scholars have conducted a series of studies on the above research hotspots. This is because international scholars are more familiar with policy theory research, the implications of an aging society, and the need for public administration research to pay more attention to social security and welfare. The right side of the graph shows the keyword cluster label, which can be divided into human rights, model, nutrition, capabilities, USA, newborn, consumption, monetary policy and other clusters in the evolution process. In general, from the perspective of time axis analysis, the hot and hot keywords, keyword correlation and keyword clustering of international policy environment research are not stable, which is quite different from the domestic comparison, which indicates that policy environment research has not attracted enough attention from international scholars.

Time Zone Analysis

The research hotspots are dynamic and vary over time. CiteSpace software can provide a “Timezone View” presentation mode for a literature co-citation network. Keywords are set in the time zone in which they appear for the first time, and the time series are arranged in the order from far to near. Then, the keyword co-occurrence map is obtained by adjusting and enhancing the image. In this way, the research process for hot topics in the time dimension can be clearly displayed. On the basis of the keyword co-occurrence map, we set the time slice to 1 year, and left the other settings unchanged. We selected “Run” to obtain the keyword co-occurrence map. We then selected “Layout” on the shortcut control board of the visualization interface, and selected “Timezone View” under “Visualizations” to obtain the original map. By adjusting the parameters to further enhance the map, the time zone map of domestic research hotspots relating to policy environments from 1992 to 2022 was obtained, as shown in Figure 14, and the time zone map of international research hotspots relating to policy environments from 2004 to 2022 was obtained, as shown in Figure 15.

Domestically, as shown in Figure 14, graph nodes represent keywords. The keywords in a particular year are used to collect data for the first time in all years; because of the time slice of 1 year, these can only be determined to be the keywords in that year. After that year, keywords appear again, and these will be in the position of the first occurrence frequency accumulation, appearing several times. When the frequency is accumulated several times, the circle becomes larger accordingly. Therefore, the large circle for the “policy environment” node does not mean a high frequency of the occurrence in the current year, but the high total frequency of the keyword in the collected data, indicating that the collected data are all discussed around the policy environment. The line represents the connection between the keywords. If the keywords appear in a paper at the same time, there will be a line between the two keywords, and the connection between the two years will also be generated. If the two keywords appear in multiple papers at the same time, the line will be bold.

Overall, the domestic policy environment research has reached a certain scale, and the center of policy environment research has moved from an early focus on the regional policy and economic environment to systems and mechanisms used to introduce the advanced experience of developed countries. These, in turn, reflect the development of highly technical and high-quality innovation environment optimization studies of the evolution. This shows that, for domestic policy environment optimization paths that directly involve money and investment (“ShuXieShi”—equivalent to a blood transfusion) or encourage innovation entrepreneurship (“ZaoXieShi”—equivalent to hematopoiesis), the overall research trend is to build a high-quality development environment. It is clear that with the proposal of China’s innovation-driven development strategy and high-quality development strategy, the current research hotspots are likely to continue for some time.

International aspects are shown in Figure 15. Nodes in the graph represent keywords, and the keywords in a particular year are used to collect data for the first time in all years; due to the time slice of 1 year, we can only determine the key words in that year. After this year keywords appear again; these will be in the position of the first occurrence’s frequency accumulation, appearing several times. If the frequency is accumulated several times, the circle will become larger accordingly, and the relevant lines will represent the connection between the keywords. If the keywords appear in one paper at the same time, a line will appear between the two keywords, and a connection between the two years will also occur. If the two keywords appear in multiple papers at the same time, the line will be bold.

Overall, the policy environment does not appear as a major node in the collection of data, showing that international research did not represent the policy environment as a whole when discussing the research object. The impact is shown as a circle, and one of the biggest nodes shows that policy environment impact studies represent a hot issue in the field of international policy environment research; this appears on the far left of the timeline. This means that international studies on policy environments focus on the impact of policies on the environment at the beginning. After 2008, research began to pay attention to the policy itself and the political, governance and management issues directly related to the policy environment. On the whole, policy environment research mainly focuses on public health and environmental protection.

3.3.4. Keyword Emergent Analysis

“Emergent keywords” refers to keywords that are used with a sudden increase in frequency within a certain period, reflecting the importance of a certain keyword in that period. Chaomei Chen defined the research frontier as a set of emergent dynamic concepts and potential research questions, which can accurately reflect the frontier fields of related disciplines. Emergent keywords are used to explore the emergent dynamic concepts and potential research issues in policy environment development research, explore the reasons behind them, reflect active or cutting-edge research nodes and assist in predicting research hotspots and trends in the future. The basic principle of emergent word detection is that the word frequency of a certain keyword variable surges in a short period of time and suddenly becomes a research hotspot. This can be represented as the “Baidu index” of academic circles. Since the emergent state of emergent words usually has time continuity, with a continuation period of 2 years or more, it can be used to assist in predicting research hotspots and trends in the future. At the same time, emergent word detection can be used to review which keywords have become hotspots in which time period.

When using the burstiness to represent the keyword occurrence, domestically, the gamma value was set to 1, and the parameters of the minimum duration were set to 1. We obtained the graph shown in Figure 16. Internationally, the gamma value was set to 1, the parameters of the minimum duration were set to 2, and we obtained the graph shown in Figure 17. In the maps, “Begin” represents the year when the keywords first appeared in the study time category; “End” represents the year when the keywords last appeared; “Strength” represents the intensity of emergence; the blue block represents the unit annual time slice and the red block represents the emergence period.

In Figure 16 and Figure 17, the keywords with a high outburst intensity in terms of CNKI policy environment research are “environmental regulation” (13.81), “influencing factors” (8.54), “environment” (5.61), “monetary policy” (5.57), “small and medium-sized enterprises” (5.45), “individual ability” (5.33) and “business environment” (5.29), indicating that these keywords correspond to domestic researchers. Frontier topics paid more attention to the time period. Keywords with a high emergence intensity in terms of WOS policy environment research are: “system” and “framework” (5.42 and 5.18), “institution” and “challenge” (4.38 and 4.17), “industry” and “organization” (4.03 and 4), etc. This indicates that the keywords mentioned above are frontier topics to which international researchers pay more attention in their corresponding time periods. From this point of view, the emergent span CNKI policy environment shows a long duration: “individual ability” (10 years), “policy” (9 years), “innovation performance” (9 years), “influencing factors” (8 years) and “monetary policy” (8 years). These keywords have been the subjects to which domestic scholars have paid close attention for some time. Some of them represent hot topics. In terms of the WOS policy environment, “AID” (10 years), “growth” (5 years) and “developing country” (3 years) have been the focus of international scholars for a long time. According to the time ranking, it can be found that frontier keywords are constantly changing with the passage of time, and the overall evolution shows stages. Therefore, in this study, we divided the research frontiers in the field of policy environment according to the time stage, and selected keywords with a high emergence intensity in this period for analysis. In the domestic context, the period from 1992 to 2001 represented the early stage, and the keywords with high emergence intensity were “individual ability s(5.33)”, “environment (5.61)s”, ”policy (4.8)”, “private economy(5.17)” and “private investment (3.84)”. The period 2002–2014 represented the middle period, and the keywords with a high emergent intensity were “mall and medium-sized enterprises(5.45)”, “environmental protection(4.36)”, ”public policy(3.76)”, ”independent innovation(4.88)”, ”financial crisis(3.59)”, ”influencing factors(8.54)”, and “America(3.53)”. The period 2015–2022 represented the recent period, and the keywords with high emergent intensity were “ecological environment(4.86)”, “innovation performance(4.25)”, “monetary policy(5.57)”, “innovation(3.72)”, ”environmental regulation(13.81)”, ”policy tools(4.8)”, ”business environment(5.29)”, ”institutional environment(4.05)”, ”environmental governance(3.69)”, ”difference-in-difference(3.69)”, ”policy implementation(3.82)”, and “industrial policy(3.78)”. From an international perspective, the period from 2004 to 2012 represented the early stage, and the keywords with high emergent intensity were experience (3.64), HIV (3.57) and prevention (3.53). The period 2013–2015 was the middle period, and the keywords with high emergent intensity were challenge (4.17), industry (4.03) and organization (4). In the recent period of 2016–2022, the keywords with higher emergent intensity were framework (5.42), system (5.18) and institution (4.38). Over recent years, domestic research on the policy environment frontier has not appeared. This will emerge with the high-speed development of China’s economic and social environment, and as more domestic scholars publish in international journals, studies of the impact on China’s policy environment will become a future WOS database research hot topic. In general, although there are differences in keywords and subject terms between domestic and international scholars in policy environment research, the correlation between domestic and international scholars is increasing, which helps policy environment research to reach a consensus and form a systematic theory as soon as possible in theory construction.

3.4. Analysis of Highly Cited Literature

Papers in a research field are not isolated, and they gradually form a citation space network that is interconnected, connecting the past and the future and cross-integrating between disciplines through the standard citation behavior among scholars. In view of this, we used the restricted citation frequency method in our study. In order to improve the accuracy of the research literature, we conduct further screening on the existing basis and select the literature that contains the policy environment in the title. In China, 309 articles retrieved from the CNKI database were sorted using the same method, and papers with no less than 50 citations were intercepted for high citation analysis, as shown in

Table 6. Highly cited CNKI articles from 1992 to 2022.

Serial Number	Article	Author	Citation (Times)
1	Research on the Behavior of Farmers’ Production in Different Policy Environments—Application of Farmer System Model [39]	Linxiu Zhang, Xiaoming Xu	214
2	Policy Environment, Financial Structure and Credit Technology: A Systematic Solution to Solve SME Loan Problem [40]	Research Group, School of Finance, Hubei University of Economics, Institute of Finance and Trade, Chinese Academy of Social Sciences, Dexu He	101
3	Analysis on development performance and Policy Environment of Agricultural Science and technology Demonstration Park [41]	Yankun Du	83
4	Preliminary Study on Environmental Impact Assessment of Major Economic Policies—Environmental Impact Assessment of China’s Automobile Industry Policies [42]	Wei Li, Zhifeng Yang	72
5	Research on the cultivation of innovative core scientific and technological talents and policy environment: an empirical analysis based on 625 questionnaires in Jiangsu Province [43]	Shuming Zhao, Qianwen Li, Xufan Zhang	66
6	Environmental regulation policy, Environmental law enforcement and Industrial green innovation capability Improvement [44]	Chao-jun Yang, Ruo-Qing Hu, Zhi-jun Feng	60
7	Competition, Capital Regulation and Efficiency Optimization of Commercial Banks: Also on the Influence of Monetary Policy Environment [45]	Jingjing Yu, Dexu He, Feifei Tong	57
8	Policy environment for nearby employment, poverty alleviation and migrant workers returning home to start businesses [46]	Liang Zhang, Yajun Li	57
9	Sme Growth: Policy Environment and Firm Performance: Empirical data from 309 smes in 23 provinces and cities in China [47]	Chen Hao	55
10	Trade liberalization, vertically linked Markets and Strategic Environmental Policies: The significance of environmental taxes for developing green trade [48]	Fei Xing, Huanlang He	52

. From the international perspective, the 95 articles retrieved from the WOS database were ranked in descending order of citation frequency, and the papers with a citation frequency of at least 25 times were selected for high citation analysis (see

Table 7. Citation of high WOS in the period 2004–2022.

Serial Number	Article	Author	Citation (Times)
1	Academic identity and autonomy in a changing policy environment [49]	Henkel, M.	390
2	Farm management indicators and farm typologies as a basis for assessments in a changing policy environment [50]	Andersen, E.; Elbersen, B.; Godeschalk, F.; Verhoog, D.	123
3	A New Scale of the US Alcohol Policy Environment and Its Relationship to Binge Drinking [51]	Naimi, T.S.; Blanchette, J.; Nelson, T.F.; Nguyen, T.; Oussayef, N.; Heeren, T.C.; Gruenewald, P.; Mosher, J.; Xuan, Z.	99
4	‘Social evils’ and harm reduction: the evolving policy environment for human immunodeficiency virus prevention among injection drug users in China and Vietnam [52]	Hammett, T.M.; Wu, Z.; Duc, T.T.; Stephens, D.; Sullivan, S.; Liu, W.; Chen, Y.; Ngu, D.; Jarlais, D.C.D.	81
5	The Alcohol Policy Environment and Policy Subgroups as Predictors of Binge Drinking Measures Among US Adults [53]	Xuan, Z.M.; Blanchette, J.; Nelson, T.F.; Heeren, T.; Oussayef, N.; Naimi, T.S.	55
6	Structural Change, the Social Policy Environment and Female Employment in Turkey [54]	Bugra, A.; Yakut-Cakar, B.	51
7	Assessment of the health system and policy environment as a critical complement to tracking intervention coverage for maternal, newborn, and child health [55]	Cavagnero, E.; Daelmans, B.; Gupta, N.; Scherpbier, R.; Shankar, A.	44
8	Identifying structural barriers to an effective HIV response: using the National Composite Policy Index data to evaluate the human rights, legal and policy environment [56]	Gruskin, S.; Ferguson, L.; Alfven, T.; Rugg, D.; Peersman, G.	36
9	Assessing the value of diverse cropping systems under a new agricultural policy environment in Rwanda [57]	Isaacs, K.B.; Snapp, S.S.; Chung, K.; Waldman, K.B.	34
10	Towards comprehensive sexuality education: a comparative analysis of the policy environment surrounding school-based sexuality education in Ghana, Peru, Kenya and Guatemala [58]	Panchaud, C.; Keogh, S.C.; Stillman, M.; Awusabo-Asare, K.; Motta, A.; Sidze, E.; Monzon, A.S.	25

).

In Table 6, the most widely cited literature [39] uses the theory of farmer economics to analyze the production behaviors of farmers in Zhangjiagang and Xinghua under different policy environments and the effects of farmers’ behaviors on the effect of national policy implementation. It is one of the earliest examples in the literature to study “policy environment” in China. One study [40] clearly defines the concept of “policy environment” and believes that it refers to the rules and conditions that affect financial institutions’ ability to provide credit to smes, including information environment, legal environment, social environment and regulatory environment. Another study [44] analyzes the mechanism of optimizing the efficiency of commercial banks by means of competition and capital regulation under the monetary policy environment. This paper specifically expands the “policy environment”. However, the concept of “policy environment” is not clearly defined. The authors in [47] constructed a policy environment evaluation system and designed a policy environment evaluation index system composed of five specific evaluation indicators specifically for the growth of small- and medium-sized enterprises. This paper is a key document of policy environment research. Based on a full analysis of the background, constraints and operational performance of the construction of agricultural science and technology demonstration zones, one study [41] put forward a preliminary conception of the choice of policy environment. Another study [42] analyzed the potential impact of major economic policies on the environment. The authors in [43] studied the impact of environmental regulation policies and environmental law enforcement on industrial green innovation ability. One study [45] analyzed the policy environment required for the training of scientific and technological talents. Another study [46] analyzed the policy environment of migrant workers returning home to start businesses from two aspects: policy demand and policy supply. The authors in [48] study the impact of trade liberalization on balanced environmental policies and social welfare. On the whole, these 10 highly cited studies in the literature systematically analyzed the concept of policy environment in theory and put forward targeted suggestions on the optimization of policy environment in practice.

In Table 7, the authors in [49] discuss the impact of British policy changes on academic identity from the perspective of communitarianism theory. Another study [50] considers the policy environment to be one of the assessment indicators of farm management, and considers that typology is a necessary tool for comprehensive environmental assessment. According to one study [51], the composite measure(s) of the alcohol policy environment have internal and construct validity. Another study [52] reviews the evolution of government policies in China and Vietnam regarding harm reduction interventions for China and Vietnam human immunodeficiency virus (HIV) prevention. The paper argues that ongoing challenges and lessons learned include the persistence of tensions between drug control and harm reduction that may have negative effects on programs until a fully harmonized policy environment is established. The authors in [53] examined the relationships of the state-level alcohol policy environment and policy subgroups with individual-level binge drinking measures. Another study [54] examined labor market relations and the social policy environment in Turkey, and their impact on the demand for and supply of female labor. The paper argues that current changes in the prevailing policy environment are considered to highlight a certain contradiction in the attitudes of decision makers who seem to be faced with a trade-off between continuing adherence to conservative patriarchal values and the objectives of increasing labor force participation and combating poverty. Another study [55] presents a systematic approach to assess the wider health system and policy environment needed to achieve positive outcomes for maternal, newborn, and child health. The authors in [56] agree, the existence of laws and policies which present obstacles to accessing HIV services for key populations. According to one study [57]: agricultural policies need to be based on a better understanding of smallholders’ objectives and constraints. Efforts to improve farming systems require innovative and inclusive approaches that enable adaptation to the socio-ecological context. The study in [58] is a key document of policy research, which systematically analyzes the policy environment of education development. The paper argues that the successful implementation of comprehensive sexuality education (CSE) programs in schools depends on the development and implementation of strong policy in support of CSE. This paper offers a comparative analysis of the policy environment governing school-based CSE in four low- and middle-income countries at different stages of program implementation: Ghana, Peru, Kenya and Guatemala. Based on an analysis of current policy and legal frameworks, key informant interviews and recent regional reviews, the analysis focuses on seven policy-related levers that contribute to successful school-based sexuality education programs. The levers cover policy development trends; current policy and legal frameworks for sexuality education; international commitments affecting CSE policies; the various actors involved in shaping CSE; and the partnerships and coalitions of actors that influence CSE policy. The paper argues that all four countries benefit from a policy environment that, if properly leveraged, could lead to a stronger implementation of CSE in schools. Latin American and African countries show notable differences in the development and evolution of their CSE policy environments, providing valuable insights for program development and implementation. On the whole, these 10 highly cited studies in the literature use different research methods to make targeted analysis of the policy environment in different fields, which is of enlightening significance to the deepening of the policy environment theory.

3.4.1. Research Topic Analysis

By analyzing the research topics of highly cited research papers in the literature of the research field, we can identify the main topics of the research field and the main content categories of the knowledge base.

In respect of China, the full text of 10 highly cited papers was imported into NVivo 12PLUS, and the function of “identifying themes” under the function of “automatic coding” was run to obtain the summary of themes, node numbers and reference points of domestic reclaimed water utilization policies (as in

Table 8. Summary of topic nodes and reference points of CNKI automatic coding.

Serial Number	Node	Reference Points	Serial Number	Node	Reference Points
1	Commercial bank loans	69	11	The developed countries	[48]
2	Level of economic development	68	12	Bank profit efficiency	[48]
3	Policy suggestions	65	13	Small cost	[47]
4	A certain degree	63	14	The theoretical study	[45]
5	High risk	59	15	Policy banks	[43]
6	Corporate credit guarantee	57	16	Dynamic mechanism	[42]
7	Commercial Banks	54	17	Cluster of scientific and technological talents	[42]
8	Technical renovation	51	18	Development space wage	[41]
9	Industrial transformation	49	19	Risk resource allocation	[41]
10	Commercial bank costs	49	20	Small businesses	[40]

), and a hierarchical diagram of nodes and reference points (as shown in Figure 18). As shown in Figure 19, “Environment (291)”, “Enterprise (268)”, “Policy (245)”, “Economy (181)”, “Development (169)”, “Science and Technology (148)”, “Banking (148)”, “Country (142)”, “Impact (138)”, “And (132)”, “Research (132)”, “Market (131)”, “Technology (128)”, “Risk (111)”, “Finance (111)”, “Talent (108)”, “Level (108)”, “Loans (107)”, “Production (102)” and “Entrepreneurship (97)” were the topics in the highly cited domestic literature. As shown in Table 7, the top ten nodes ranked by the number of reference points under those topics are “commercial bank loan”, “economic development level”, “policy proposal”, “certain degree”, “high risk”, “enterprise credit guarantee”, “commercial bank”, “technological transformation”, “industrial transformation” and “commercial bank cost”. This shows that the domestic policy environment mainly focuses on how to cultivate a good business environment in order to provide convenient conditions for enterprises’ technological and industrial transformation and reduce the risks of enterprise entrepreneurship and operation.

From the international perspective, the full text of 10 highly cited articles was imported into NVivo 12PLUS, and the function of “identifying topic” under the function of “automatic coding” was run. In addition to research subject, only the topic node and reference point hierarchy maps composed of connectives such as “and”, “to”, “in”, “for” and “that” were obtained. This indicates that the topics in highly cited papers in the WOS database are not prominent, or that the international literature on the policy environment does not take it as the research object, or does not involve systematic thematic discussion on the policy environment even though it mentions it. We ran the NVivo 12PLUS word frequency search function, set the display words as “100”, the minimum length as “2” and the group as a synonym. We added the above connectives into the list of stopped words, exploring other topics except connectives, and obtained

Table 9. Summary of WOS word frequency queries.

Serial Number	Word	Percentage (%)	Serial Number	Word	Percentage (%)
1	policy	0.89	11	employment	0.27
2	health	0.71	12	alcohol	0.26
3	policies	0.56	13	level	0.25
4	hiv	0.48	14	environment	0.25
5	education	0.40	15	government	0.25
6	countries	0.37	16	crop	0.25
7	national	0.32	17	system	0.24
8	systems	0.32	18	data	0.24
9	social	0.30	19	services	0.24
10	cse	0.28	20	based	0.23

and Figure 19. As the table shows, the WOS word frequency search is sorted by percentage, and the top ten words are “policy”, “health”, “policies”, “hiv”, “education”, “countries”, “national”, “systems”, “social” and “cse”. This shows the international discussion in respect of the healthcare policy environment and the public, political and professional aspects of the policy environment.

3.4.2. Emotional State Analysis of Research Literature

The highly cited literature in CNKI and WOS was analyzed, and an automatic condition-recognition sentiment analysis was carried out using NVivo. A statistical table of coding points, a 3D bar chart and a hierarchical summary chart of the research literature, compiled according to sentiment, were obtained. From

Table 10. Summary of sentiment results of CNKI highly cited literature automatic coding.

Serial Number	Document Name	Very Negative	More Negative	More Positive	Very Positive
C01	Research on Farmers’ Production Behavior under Different policy Environments—Application of farmer system model	8	10	29	0
C02	Policy Environment, Financial Structure and Credit Technology—A systematic solution to the difficulty of SME lending	33	50	62	0
C03	Analysis of development performance and policy environment of agricultural science and technology demonstration park	10	16	41	0
C04	Preliminary Study on Environmental Impact Assessment of Major Economic Policies—Environmental Impact Assessment of China’s Automobile Industry Policies	23	7	20	2
C05	Research on the cultivation of innovative core scientific and technological talents and policy environment: An empirical analysis based on 625 questionnaires in Jiangsu Province	20	14	51	5
C06	Environmental regulation policy, environmental law enforcement and industrial green innovation capacity improvement	19	22	26	0
C07	Competition, Capital Regulation and Efficiency Optimization of Commercial Banks—Also on the influence of monetary policy environment	14	47	45	1
C08	Policy environment for nearby employment, poverty alleviation and return of migrant workers to their hometowns to start businesses	18	14	55	1
C09	Sme Growth: Policy Environment and Firm performance—Empirical data from 309 smes in 23 provinces and cities in China	6	12	23	3
C10	Trade liberalization, vertically linked Markets and Strategic Environmental Policies: Implications of environmental taxes for developing green trade	63	34	83	1
Total		214	226	435	13

, the statistics show that in the domestic context, “very negative” has 214 coding nodes; “relatively negative” has 226 coding nodes; “relatively positive” has “435 coding nodes; and “very positive” has 13 coding nodes. From a comprehensive analysis of Table 10 and Figure 20, at present, scholars’ attitudes towards the policy environment are still complex, but tend to show that existing policies have made certain achievements and there is a significant development space and development prospects. From Figure 21, it can be seen that the maximum values of very negative and very positive correspond to literature numbers C10 and C05, indicating that the academic community holds a strong negative attitude towards the environmental policy of trade liberalization and a strong positive attitude towards the policy environment of innovative core scientific and technological talent training. From Figure 22, we can see that the “neutral” and “hybrid” attitudes were the most frequent. From the international perspective, the above operation was performed using the NVivo software, and no-emotion coding results were obtained, indicating that the international research on policy environments is not taking the policy environment as the whole research object in terms of the emotional orientation elaboration, and that future research in this aspect should be strengthened.

4. Conclusions

Policy environment has attracted the attention of scholars at home and abroad because of its strong influence on policy formulation, implementation and effect. In recent years, in terms of theoretical research, domestic and foreign scholars have generally paid more attention to the policy environment, and the published papers show a trend of increasing year by year. Domestic and foreign scholars have different concerns, which is closely related to the different development of the policy environment. In practice, compared with developed countries, China, as a developing country, has more realistic expectations for the optimization of its policy environment. Specifically, it is manifested in the following ten aspects:

(1)

The number of “policy environment” articles at home and abroad shows a steep upward trend, scholars both at home and abroad attach great importance to policy environment research, and policy environment research has broad prospects.

(2)

There is a lack of communication and cooperation among domestic “policy environment” research subjects, and a close and benign interaction cooperation network has not yet been formed. Compared with domestic research, international “policy environment” research subjects are closely connected, the core authors have been initially formed, and research institutions have formed a certain scale of cooperation network. There is no fixed contact network between international research subjects and domestic research topics, and international seminars on policy research are few or not large-scale, leading to the lack of relevant dialogue and contact mechanisms at home and abroad.

(3)

Countries with great influence in the field of policy environment research are the USA, Switzerland, Finland, Ireland and Austria. The international influence of China’s policy environment research needs to be improved.

(4)

The research hotspots in terms of the domestic policy environment mainly focus on encouraging high-quality economic and social development, paying less attention to social security and welfare, while foreign countries pay more attention to social security and welfare.

(5)

From the perspective of the characteristics of research hotspots, the hotspots of the domestic policy environment research focus on intractable problems faced in terms of social and economic development, the problems of which closely represent social practice, but forward-looking and guiding research on social practice is insufficient. The international community attaches great importance to the discussion of theoretical models, which provide stronger theoretical guidance for the practice of policy environment optimization.

(6)

In the context of the evolution trend, the domestic policy environment optimization path is realized by the allocation of funds (“ShuXieShi”) to encourage innovation entrepreneurship (“ZaoXieShi”), and international policy environment research revolves around the main issues, including public health and environmental protection policy, governance theory, and forward-looking guidance.

(7)

In the context of research, domestic research on the policy environment frontier is not emergent, but will emerge with the high-speed development of China’s economic and social environments, and with domestic scholars being published in international journals, studies on the impact on China’s policy environment will become a future WOS database research hot topic.

(8)

The research topics of highly cited papers on the domestic policy environment focus on cultivating a good business environment to provide convenient conditions for enterprises’ technological transformation and industrial transformation, so as to reduce the risks of enterprise entrepreneurship and operation. The international focus is on the public, political and professional nature of the healthcare policy environment.

(9)

Domestic researchers have conducted emotional orientation studies on the policy environment. Overall, the rational emotional attitude toward the policy environment is dominant, indicating that the current policy environment optimization is in the transition period and has a large development space and prospect. The emotional orientation of the policy environment is not obvious in the international community, and the policy environment has not been systematically studied as a whole research object, so the research in this context should be strengthened in the future.

5. Suggestions for Optimizing China’s Policy Environment

The policy environment, from theory to practice, is developing. From the theoretical point of view, there is no systematic policy environment theory. In practice, the policy environment has not been used as a regular method or tool. This paper tries to construct a general analysis from the theory and practice of the policy environment, and makes a bold attempt at analyzing the policy environment from the concept to the method, hoping to play a role in attracting future contributions. Of course, the research on policy environment is not only a generalization of theory, but also a targeted study related to practice. In particular, we lack in-depth analysis on the specific details of China’s policy environment practice. This needs to be expanded in our future research.

To summarize, China’s policy environment optimization is in transition. There is a need to strengthen policy environment theory research and the policy environment optimization to provide theoretical guidance and public participation. Not only does the government need to step up promotion efforts, but there is also a need for people from all walks of life to raise awareness and to form a resultant force, promoting domestic and international theory reference and cooperation, and realizing the policy of ecological quality of ascension.

Based on the above research conclusions, we put forward the following optimization suggestions for China’s policy environment.

5.1. Improve the Adaptability of the Policy Environment

At present, the development of China’s policy environment is lagging behind, its adaptability to practice is not strong and its role in guiding practice is not obvious. Therefore, it is necessary to strengthen research on the policy environment and constantly improve the adaptability of the policy environment. On the one hand, based on the specific policy environment situation, it is necessary to carry out systematic and in-depth research on a specific policy environment (such as innovation and entrepreneurship policy, population and talent development policy, public health policy, etc.), grasp the characteristics of the specific policy environment, and put forward targeted optimization suggestions. On the other hand, we should pay attention to policy revision and improvement, and improve the adaptability of policy updating, revising and abolition.

5.2. Optimize and Improve the Policy Implementation Environment

In order for policies to be implemented, they need to be well implemented, and the policy environment cannot be optimized without optimization of the environment for policy implementation. At present, China’s policy implementation environment poses a prominent problem, the solving of which is the key to policy environment optimization. On the one hand, we need to improve the operability of policies and address the difficulty in their implementation. On the other hand, we need to promote a lower level of consistency between power and responsibility in terms of policy implementation, so that policies are implemented at different levels and no lower level fails to implement them.

5.3. Establish an Interactive Mechanism for Optimizing the Policy Environment

The policy environment involves a large number of policy subjects, so a multi-party interactive mechanism for policy environment optimization should be established. On the one hand, we need to broaden the channels of participation. We should not only open channels for obtaining multi-agent policies, but also open channels for collecting policy demands. On the other hand, we want to create a useful think-tank, as well as experts, scholars and other third parties, and actively involve them in the optimization of the policy environment in China. This will guide the country’s policy of environmental science research and enable it to critically absorb the domestic and foreign policy environment research results, in order to strengthen social security and welfare policy environment research, while promoting the benign development of economy and society. When promoting economic and social development, we should give full consideration to China’s social security and welfare undertakings.

5.4. Promote Theoretical Research on Policy Environment

Theory is the basis of practice, and the scientific nature of theory is closely related to the effectiveness of practice. Therefore, strengthening the theoretical research of policy environment is very important for the optimization and promotion of policy environment. From the above analysis, there are still many gaps in the theoretical research on the policy environment, and many issues such as the definition of basic concepts, the systematization of theories, and the disciplinarity of theories need to be further explored and consensus reached. From the perspective of discipline and science, policy environment research has not formed a systematic theory. To address the demand for policy environment development, the study of the policy environment not only requires the efforts of Chinese scholars, but also the cooperation between scholars at home and abroad, because the policy environment is in need of theoretical research, and promoting the social and economic development of the policy environment in all countries and areas is the crucial factor. For example, the business environment in the policy environment has become a competitive index to measure a country or region.