Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

https://doi.org/10.24294/jipd.v8i1.2569

Article

The role of industry-academia collaboration in enhancing educational

opportunities and outcomes under the digital driven Industry 4.0
Caroline Olufunke Esangbedo1, Jingxiao Zhang1,*, Moses Olabhele Esangbedo2, Seydou Dramane Kone1,
Lin Xu3

1
Chang’an University, School of Economics and Management, Xi’an 710000, China
2
Xuzhou University of Technology, School of Management Engineering, Xuzhou 221018, China
3
School of Foreign Languages, Northwest University, Xi’an 710069, China
* Corresponding author: Jingxiao Zhang, zhangjingxiao964@126.com

CITATION Abstract: We studied the role of industry-academic collaboration (IAC) in the enhancement
Esangbedo CO, Zhang J, Esangbedo
of educational opportunities and outcomes under the digital driven Industry 4.0 using
MO, et al. (2024). The role of research and development, the patenting of products/knowledge, curriculum development,
industry-academia collaboration in and artificial intelligence as proxies for IAC. Relevant conceptual, theoretical, and empirical
enhancing educational opportunities literature were reviewed to provide a background for this research. The investigator used
and outcomes under the digital driven
Industry 4.0. Journal of
mainly principal (primary) data from a sample of 230 respondents. The primary statistics
Infrastructure, Policy and were acquired through a questionnaire. The statistics were evaluated using the structural
Development. 8(1): 2569. equation model (SEM) and Stata version 13.0 as the statistical software. The findings indicate
https://doi.org/10.24294/jipd.v8i1.25 that the direct total effect of Artificial intelligence (Aint) on educational opportunities
69
(EduOp) is substantial (Coef. 0.2519916) and statistically significant (p < 0.05), implying
ARTICLE INFO that changes in Aint have a pronounced influence on EduOp. Additionally, considering the
indirect effects through intermediate variables, Research and Development (Res_dev) and
Received: 10 August 2023
Accepted: 25 September 2023 Product Patenting (Patenting) play crucial roles, exhibiting significant indirect effects on
Available online: 11 December 2023 EduOp. Res_dev exhibits a negative indirect effect (Coef = −0.009969, p = 0.000) suggesting
that increased research and development may dampen the impact of Aint on EduOp against a
COPYRIGHT priori expectation while Patenting has a positive indirect effect (Coef = 0.146621, p = 0.000),
indicating that innovation, as reflected by patenting, amplifies the effect of Aint on EduOp.
Notably, Curriculum development (Curr_dev) demonstrates a remarkable positive indirect
Copyright © 2023 by author(s).
Journal of Infrastructure, Policy and effect (Coef = 0.8079605, p = 0.000) underscoring the strong role of current development
Development is published by EnPress activities in enhancing the influence of Aint on EduOp. The study contributes to knowledge
Publisher, LLC. This work is licensed on the effective deployment of artificial intelligence, which has been shown to enhance
under the Creative Commons
educational opportunities and outcomes under the digital driven Industry 4.0 in the study area.
Attribution (CC BY) license.
https://creativecommons.org/licenses/ Keywords: Industry 4.0; industry-academia collaboration; artificial intelligence; patent
by/4.0/
development; curriculum; research and development; SEM

1. Introduction
Collaboration frameworks (Perkmann, Tartari et al., 2012) have been developed
in response to modifications of the working environments of different corporate
institutions (Cunningham and Link, 2014). Industries have always looked to
institutions of higher education as possible sources of new ideas and knowledge in
an effort to expand their knowledge base and to improve their ability to provide fresh
solutions to the complex problems confronting society (Perkmann, Neely et al.,
2011). According to Perkmann, Neely et al. (2011), in order to achieve their goals of
knowledge and innovation, a rising number of organizations are collaborating with

1
Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

academic institutions. Academics and policymakers are now more interested than
ever in the effect of university study on corporate innovation.
Universities have the potential to collaborate with businesses in a mutually
beneficial manner, thereby enhancing educational possibilities and value. This can be
achieved through the creation of improved resources for higher and advanced
education, facilitating the exchange of innovative ideas and technology transfer, and
cultivating an environment conducive to continuous learning. There exists potential
for enhanced collaboration between higher education institutions and industries, with
the aim of collectively attaining shared objectives and ensuring that graduates
possess the necessary skills and knowledge to successfully integrate into the labor
market and make meaningful contributions to global economies (Weagle et al.,
2019). This is one of the most effective strategies for technological development in
developed or industrialized countries and is a useful instrument for the effective and
efficient application of science and technology to solve social issues. There are
several ways in which these collaborations can take place, including through
cooperative research initiatives, joint curriculum creation, and joint product
patenting (product collaborations) (Lucietto et al., 2021). Schools and businesses can
work together to increase the eminence of education and design training programs by
means of collaborating. In an increasingly competitive global market, success
requires a workforce that can keep up with and learn from new markets and
technologies. Quality educational programs are clearly essential to the sustainability
of industrialization. Accordingly, there has been a shift in thinking regarding
education and industry partnership,which is now gathering steam (Weagle et al.,
2019).
Mukherji and Silberman (2021) posit that the phenomenon of industry-
academic collaboration, commonly denoted as industry-academia collaboration (IAC)
or university-industry collaboration (UIC), entails a mutually beneficial alliance
between academic establishments (such as universities and research institutions) and
industrial entities (including companies, businesses, and industries). The
establishment of partnerships between universities and industries, frequently
facilitated by government action, is widely recognized as a crucial factor in
enhancing regional and national innovation systems (O’Dwyer et al., 2023). The
purpose of this partnership is to leverage the respective experience, resources, and
capacities of the involved parties in order to promote the advancement of research,
innovation, and societal improvement in a mutually beneficial manner. Fischer et al.
(2019) argue that the establishment of partnerships between industry and academics
serves as a means to foster the flow of knowledge and expertise. Academic
institutions play a vital role in generating theoretical knowledge and conducting
research, whereas industry offer valuable practical insights and present real-world
issues. Collaborative endeavors encompass cooperative research initiatives aimed at
resolving industry-specific challenges, propelling technological advancements, and
nurturing the cultivation of innovative ideas (Alexander et al., 2020). Academic
research frequently generates novel technologies that are subsequently
commercialized through collaborative efforts, effectively bridging the gap between
academia and practical market applications. The partnership aims to synchronize
educational curriculum with the requirements of the industry, so augmenting the

2
Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

competencies of graduates and facilitating their seamless integration into the labor
market (O’Dwyer et al., 2023). Moreover, the institution provides opportunities for
internships, training programs, and workshops, hereby facilitating the exposure of
students and staff members to real-world industrial experiences.
The percentage of industry-academia collaboration in China has been increasing
in recent years. In 2019, the Chinese government issued a policy document that
called for increased collaboration between industry and academia, and set a goal of
doubling the number of industry-academia collaboration projects by 2025 (Ministry
of Science and Technology China, 2020). According to a report by the Chinese
Ministry of Science and Technology, the number of industry-academia collaboration
projects in China increased from 180,000 in 2015 to 320,000 in 2020, representing a
quantum leap. The total funding for industry-academia collaboration projects in
China also increased from 100 billion Yuan in 2015 to 300 billion Yuan in 2020 as
shown in Figure 1. The Chinese government is expected to continue to promote
industry-academia collaboration in the coming years. The government has identified
a number of key areas for industry-academia collaboration, including artificial
intelligence, big data, and semiconductors.

Figure 1. Percentage Increase in industry-academia collaboration.

Industry-academia collaboration in China has been steadily increasing in recent

years, fueled by government policies, initiatives, and a growing emphasis on
innovation-driven development. China’s “Double First Class” initiative, which aims
to cultivate world-class universities and disciplines, has encouraged closer ties
between academia and industry. The government has encouraged research
institutions and universities to work closely with industries to drive innovation and
technology transfer (Ministry of Science and Technology China, 2020). Particularly,
special economic zones and tech hubs in cities like Beijing, Shanghai, Shenzhen, and
Hangzhou have witnessed a surge in collaboration between universities and tech
companies. The establishment of innovation centers, research parks, and incubators
has created conducive environments for collaboration, allowing academia to
contribute theoretical knowledge and research expertise while industries provide
practical application and market-driven insights. Moreover, programs like the
“Industry-Academia Cooperation Project” by the Ministry of Education and various
research grants have further incentivized collaboration between academia and
industry (Ministry of Science and Technology China, 2020).

3
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

At the heart of this collaboration is the infusion of artificial intelligence in IAC.

Artificial intelligence (AI) is the ability of a machine to simulate human intelligence.
AI systems can learn from data, identify patterns, and make decisions in order to
solve problems. AI is used in a wide range of applications. Artificial intelligence (AI)
has emerged as a transformative force within the realm of education Murphy (2020),
finding widespread applications primarily within educational institutions in a
multitude of forms (L Chen et al., 2020). The inception of AI in education can be
traced back to its initial manifestations through computer and computer-related
technologies. Over time, its presence expanded into web-based and online intelligent
education systems (Murphy, 2020). Subsequently, AI further evolved to encompass
embedded computer systems, humanoid robots, and web-based chatbots, enabling
diverse modalities for executing the responsibilities and functions of instructors. The
integration of AI platforms has significantly streamlined administrative tasks for
educators, leading to enhanced effectiveness and efficiency in their roles. Moreover,
AI has proven instrumental in curriculum development and research activities
(Estevez et al., 2019). By leveraging machine learning and adaptability, educational
systems can tailor curriculum and content to suit the unique needs of individual
students. A distinctive feature of AI-enabled education is its capacity for
customization and personalization. AI algorithms analyze data and patterns, allowing
for the creation of personalized learning paths for students. This tailored approach
contributes to heightened engagement and retention rates among students, thereby
enriching the overall learning experience and augmenting the quality of education
provided (L Chen et al., 2020). In essence, AI has evolved to become a vital enabler
within the educational landscape, profoundly impacting teaching methodologies,
administrative processes, and student outcomes. As technology continues to advance
and AI capabilities become increasingly sophisticated, the educational sector stands
to benefit even more, with the potential to revolutionize the way knowledge is
imparted, acquired, and applied in the pursuit of academic excellence Global
Development of AI-Based Education.
A successful university–industry collaboration is led by academic management
(Rahm et al., 2000; Edmondson et al., 2012); an emphasis on long-term calculated
relationships (Calder, 2007); and a mutual vision and approach to the
accomplishment of a goal. All of these elements are crucial to the success of a
relationship with industry. In China, science and technology (S&T) now accounts for
56.4 percent of economic growth, up from 56.4 percent in 2016 and 59.5 percent in
2019; this represents an increase from 39.7 percent in 2003. Theratio of enterprise
R&D spending to national R&D spending increased from 62.37 percent to 77.46
percent over the same time period (Zheng and Wang, 2021). In China, company
innovation has become a driving force for national innovation. China’s global
competitiveness may be increased (J Hong et al., 2015), and the “average-earning
trap” can be avoided (Liu et al., 2017) by enhancing the innovative capacity of
Chinese businesses.
Collaboration is common in many different fields, as evidenced by several
authors. China is a massive economic powerhouse in Asia, and its influence on
global infrastructural development and technological innovation is immense. A
constructive industry-academic partnership is vital for this to be achievable (Agrawal,

4
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

2001). Industry–university collaboration is required for research and development

(Scandura, 2016), product patenting and licensing (Luan et al., 2010; Leydesdorff,
2004; Rosell and Agrawal, 2009; Dill, 1995; Balconi and Laboranti, 2006),
curriculum development, and delivery and evaluation (Guimón, 2013; Deborah,
2011). In line with Shewakena Tessema (2017), we believe that collaboration in
curriculum development and delivery would ensure that what is taught in universities
is pertinent to the goals of the industry. Numerous Chinese research institutes have
engaged in significant market-oriented operations since the late 1990s. Research
institutes have regularly worked with businesses on a wide range of innovative
projects due to their excellent R&D skills and increased market sensitivity (Liu et al.,
2017). In contrast, since 2005, as a result of the most recent wave of science and
technology (S&T) reforms, in addition to teaching and research as a “third mission”,
Chinese universities are starting to collaborate with businesses (W Hong, 2008).
To improve educational prospects for students, there has been a tremendous
push for collaboration between academia and industry over the past decade (Ziegler,
1983). The importance of providing students with suitable education, particularly at
the undergraduate level, by universities has long been recognized (Eli, 1986). The
“industry/academic gap” has been the subject of a slew of sessions at conferences,
frequently characterized by a lack of comprehension of the goals and approaches of
each side (Judy, 1986). Successful relationships are driven by individuals who have a
thorough understanding of both the academic and corporate worlds (Edmondson et
al., 2012). When industry and academics collaborate, qualified workers who are
prepared with the latest technologies to answer society’s concerns are offered to
industry. One of the numerous educational benefits of industry-academia
partnerships is the opportunity to learn about the process of innovation. It is
challenging to establish and use sets of indicators to evaluate industry-academia
collaborations, given the goal of these collaborations is to produce new results.
Traditional measures cannot capture the intricacies and complexities of innovation.
A wide range of indications must be used instead (Smith, 2006).
Through a computational process, artificial intelligence aims to explain all
characteristics associated with human intelligence. It is able to interact with its
surroundings through the use of sensors and can make decisions autonomously
(Aishath et al., 2019). Simply said, AI is a type of manufactured intelligence.
Intelligence is a distinguishable individual attribute or quality that can be
distinguished from all other individual properties. The behaviors of artificial
intelligence are also observable in the performance of specific tasks. The fourth
industrial revolution, or “Industry 4.0”, is propelled by innovative technology,
particularly improved information and communication technologies.
ICT stands for Information and Communication Technology. It represents a
broad term that encompasses all technologies used to manage, process, transmit, and
exchange information in various forms including data, text, sound, images, and video
(UNESCO, 2019). The core objective of ICT is to enhance communication and
facilitate efficient storage, retrieval, processing, and usage of information. Almost
every level of industrial operation is affected by advanced ICT. ICT is a significant
factor that enhances the responsiveness and effectiveness of production and supply
chains (Apiyo and Kiarie, 2018). The urge to delegate human duties and tasks in

5
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

various settings to machines has led to the rise of smart environments as the world
continues to enter the digital era. The educational context is one of these
environments. We hypothesize that the adoption of artificial intelligence will be a
prerequisite for achieving the desired level of educational opportunities.
China is one of the world’s leading centers of artificial intelligence (AI)
development, with its largest technology companies driving R&D. Its large populace
and varied industrial composition can generate enormous amounts of data and create
a vast market (Manyika et al., 2017). China’s future economic growth could be
contingent on the extensive adoption of artificial intelligence (AI) technologies in
R&D, product patenting, and curriculum development. The application of AI in
industry-academia collaborations has enormous economic potential for China due to
its rapidly evolving nature. The country will need to pay close attention to
developing its potential for innovation. For instance, despite the fact that Chinese
scholars have written and published more research articles on AI than American
researchers, their works have not had the same influence as those authored by
American or British authors. Furthermore, a comparable AI environment to that of
the United States, which has generated a lot more AI startup businesses than China,
is also lacking in China (Manyika et al., 2017). The American environment is vast,
creative, and diverse (including study institutions and further education colleges as
well as private corporations). It incorporates all of Silicon Valley’s well-known
strengths and has numerous advantages that are uncommon elsewhere (Biba, 2016).
In addition, current expansion of quantitative research is occurring (for example,
Kim et al., 2019; Belitski et al., 2019). It is still necessary to do quantitative research,
particularly in China, to determine the role of intermediaries (such as artificial
intelligence) in the connection between academic-industrial collaboration and
industrial innovation.
As an outcome of the rapid transformation of the technical growth of
industrialized nations, China, which is unquestionably the Asian superpower in every
field, must find a more efficient and expedient means of improving educational
prospects. To reach the current level of industry-academia collaboration in the USA,
a more effective method of enhancing educational outcomes is required in China.
The adoption of artificial intelligence enables the university and industry to make
great decisions, hence improving the speed and accuracy of strategic decision-
making processes, which will improve basic business operations and enhance the
product and services quality.
The research problem addressed in this study revolves around investigating the
transformative potential of industry–academia collaboration in the context of
enhancing educational opportunities and outcomes amidst the pervasive digital
landscape of Industry 4.0. The study recognizes the fundamental shift brought about
by Industry 4.0, driven by advanced digital technologies and artificial intelligence
(AI), and aims to comprehend how the synergy between academia and industry can
be harnessed to optimize educational prospects. It seeks to elucidate the mediating
role of AI, acting as a catalyst, in mediating the relationship between industry–
academia collaboration and educational outcomes. This research problem
encapsulates the pressing need to explore innovative strategies that align educational
systems with the rapidly evolving demands of the digital era, ultimately fostering a

6
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

symbiotic relationship between academia and industry to nurture a skilled and

adaptable workforce for the future. The investigation is grounded in the urgency to
bridge the existing knowledge gap and identify effective pathways to leverage this
collaboration, shaping a dynamic educational landscape within the ambit of Industry
4.0. The broad objective of the study is to examine the role of industry–academia
collaboration in enhancing educational opportunities and outcomes under the
digitally driven Industry 4.0. The study was guided by the following research
questions:
1) What is the effect of research and development on educational opportunities
and outcomes under the digital driven Industry 4.0 in China.
2) What is the effect of patenting of products/knowledge on educational
opportunities and outcomes under the digital driven Industry 4.0 in China.
3) What is the effect of curriculum development on educational opportunities and
outcomes under the digital driven Industry 4.0 in China.
4) What is the mediating role played of artificial intelligence on the relationship
between industry–academia collaboration and educational opportunities and
outcomes under the digital driven Industry 4.0 in China.

2. Literature review and theory

2.1. Industry academia collaboration (IAC)

Many researchers have examined the role of academic–industry collaboration in
various contexts, for example, in the United Kingdom (D’Este and Patel, 2007), the
United States (Ponomariov, 2013), Japan (Motohashi and Muramatsu, 2012), and
others (D’Este and Patel, 2007). IAC articles, in spite of the apparent diversity of the
study contexts, can be divided into three groups: IAC drivers, IAC patterns, and IAC
outputs. This study focused on the output dimension of academic–industry
collaboration. Alliances, joint ventures, networks, and consortia are the most
common types of industry-academia cooperation (Ankrah and Al-Tabbaa, 2015).
There are a variety of ways in which the participating organizations are linked.
Organizations can often jointly develop initiatives that focus on certain scientific or
technological topics through various kinds of partnership. It is also possible that,
under some circumstances, long-term collaborations are developed rather than being
a requirement to solve a technical challenge or produce commercial items rapidly.
Companies are looking for ways to increase their social capital as well as their
capacity for the generation of new ideas through long-term collaborations.
University-industry partnerships and collaborations may be widespread; however
there may be variations between industries, according to Perkmann and Walsh
(2007). Open and networked innovation activities have demonstrated that real
partnerships and collaborations—rather than merely abstract connections—play a
stronger role in fostering innovative activities and the potential of participating
organizations. It has been pointed out that corporations involved in these
relationships are looking for enhanced innovative capabilities, rather than rapid,
commercialized, tangible results. Large corporations are no longer the only ones
interested in collaborating with universities. Collaboration networks are being
formed by large and small organizations in order to further their creative efforts

7
Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

aimed at benefiting both markets and customers. The Triple Helix model is a
common way for public institutions to support university and industry collaborations.
It has been found to be a critical component in the development of regions and
countries around the world (Cai and Etzkowitz, 2020; Guimón, 2013; Dooley and
Kirk, 2007).

2.2. Collaboration in the Chinese context

The China growth model is based on the switch from massive industrial
investment to development encouraged by a creative society (Naughton and Tsai,
2015). One important strategy for generating this kind of inventive growth has been
suggested: science and technology. The enthusiasm of prominent Chinese
universities for university-industry collaboration (UIC) appears to be much lower
than that of their European and American counterparts, according to a similar
analysis conducted by Li et al. (2020). On average, each of the top US colleges
publishes twice as many UIC publications as China’s equivalent top-ranked
institutions. Tsinghua University, a Chinese university, produced 1581 papers
between 2013 and 2016, which is comparable to the University of Minnesota Twin
Cities, which was ranked 15th among 175 American universities in the 2018 Leiden
Ranking. Tsinghua University, ranked number one in China, is about on par with
Freie Universität Berlin, ranked 10th in the European Union. According to
Argyropoulou et al. (2019), there is no evidence to support this conclusion. They
claim that there a paradox of innovation is occurring in China. These authors argue
that while China’s scientific outputs and inputs are of the highest quality,
productivity gains are not as strong. As a result, China ranks low in terms of its
industry–academic collaboration when compared to other countries due to the lack of
collaboration and publication in scholarly publications.
Collaboration between industry and academics is capital-intensive. When
government funding is insufficient, participation in industry-academia partnerships is
poor. For instance, the UK government’s Higher Education Innovation Fund will
invest two hundred and thirteen million Euros to facilitate contact between
universities and industries. It has been observed in China that collaborations with
institutions of higher education are predominantly created with big corporations,
notably state-owned industries, leaving SMEs, which constitute a significant share of
the Chinese industrial base, with less assistance for their R&D operations (Liu et al.,
2017). In addition, the customary technological vision of creativity and the
importance placed on formal collaboration channels have restricted Uniform
Industrial Corporations (UICs) to influential colleges and the fields of science,
technology, engineering, and mathematics (STEM). First-tier and regional
universities are the two categories used to categorize universities in China. Regional
institutions are less research-focused and are run by their individual provincial
governments, but first-tier universities are generally thought to produce higher-
quality research and are directly governed by the Ministry of Education. In contrast
to the eight programs operated by regional universities, each top-tier university
engages in an average of 61 industry collaboration programs, according to a national
research project conducted by the Chinese Ministry of Education in 2019 (Ministry

8
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

of Education, 2019). According to a survey conducted by Hughes and Kitson (2012),

in the UK, nearly half of the 4452 collaborative projects recorded involved STEM
fields and the health sciences, with around 30% involving the arts and humanities,
and about 20% involving the social sciences. Although the importance of top-tier
academic institutions and the natural sciences has been acknowledged, more research
should be done to ascertain how regional institutions with lower rankings and the
social sciences may aid in industrial innovation.

2.3. The Helix models

The framework focuses on academia, universities, and industry as innovation
agents. In this study, academia and industry form the unit of analysis as an agent of
the incubation of innovations (universities) and the utilization of innovation
(industry). Their fruitful interactions and collaborative efforts promote economic
growth and innovation in the region. The Triple Helix hypothesis suggests that the
purpose of universities should go beyond social instruction and investigation to also
include the contribution to provincial advancement through the creation,
dissemination, and utilization of industry-driven knowledge. The Triple Helix model
is an “innovation-push” approach, in which innovation is pushed from the academic
world into industry, where it is then refined and put to use. According to this model,
the government is able to fulfill its responsibility to the public by funding research at
universities and outlining the path forward for regional or national innovation
systems through public policy. The conventional concept of a “triple helix” has been
developed further (Miller et al., 2016) in order to encourage interactions between all
of the various social sectors to jointly produce new knowledge and innovations.
Figure 2 is a path diagram that describes the pathways between the variables
included in the study and the covariance between the exogenous variables together
with the recursive nexus between the endogenous variable and the control variable
(artificial intelligence).

Figure 2. Path diagram (source: research model).

H1: Industry-academia collaboration through joint R&D does not significantly

enhance educational opportunities.
Firms’ level of R&D has been noted as a key determinant (Laursen and Salter,
2004) in industry–academia collaboration (Fernández López et al., 2014; Aiello et al.,
2019). Investment in internal R&D increases a company’s capacity for learning and

9
Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

absorbing external knowledge, because university knowledge may be difficult to

transfer to businesses (Cohen and Levinthal, 1989). It is via the use of artificial
intelligence (e.g., artificial intelligence, automation technologies, and others) that
corporations and academia are able to work together. The Chinese government has
made a huge investment in R&D, reaching up to 301.32 billion Yuan, which
represents a significant increase. According to Albahari (2017), academic institutions
and private enterprises enter into R&D partnership agreements to work together on
research projects, regardless of who is providing the funding. Such projects can
include anything from industry-funded cooperative research programs to research
pacts, experimentation, compliance and accreditation testing for organizations, and
joint publications with experts from the company. It also includes co-funding for
PhD students as well as industrial PhDs (Khachoo et al., 2018). When the worlds of
academia and industry come together, there is an assumption that both will benefit
(Payne, 2007; Breese, 2012).
H2: The development and patenting of products does not improve educational
opportunities.
Mazzocchi (2004) measured the industry–academia partnership by submitted
and granted patents, the patent utilization ratio, the percentage of supported ideas,
and the revenue from external licenses. Patenting uses these metrics. These metrics
show industry-academia collaboration. Relevant literature examined the roles of
intermediaries in university-industry collaboration (Kim, 2019; Belitski, 2019). We
hypothesize that using AI to patent manufactured goods will improve Chinese
education. Research promotes industrial innovation through product creation,
patenting, and human capital transfer (Perkmann, Tartari et al., 2012; Motohashi,
2006). Universities produce patentable goods when they support local and regional
economic development, commercialize research, and improve faculty contact with
industry scientists. Increased Chinese university patenting is driven by the “Chinese
Bayh-Dole Act” and a research assessment system that supports patenting by
researchers. Due to flaws in the evaluation system and the sequential structure of
university-industry ties, rapid growth in academic patents results in low patent
quality and low university patent utilization.
H3: Curriculum development and delivery does not significantly enhance
educational opportunities.
Courses, modules, programs, majors or minors, planned experiments, and
course delivery by outside organizations may be jointly produced and supplied. This
includes joint PhD supervision, the creation and organization of new study programs,
guest lectures by business representatives, curriculum evaluation, and student
business experiences. University–industry collaboration is important, especially for
curriculum development and implementation. Few curriculum development models
are specific enough to be used in practice or generalized outside their area of origin.
In recent years, many examples of curriculum design collaboration have been
published to adapt higher education to technical and social realities. Literature
examples include design in Portugal (Motohashi, 2006), renewable energy in Latin
America and Europe (Comodi et al., 2019), tourism (Dopson and Tas, 2004), the
automotive industry in the USA (Mears, 2009), industrial engineering in Thailand

10
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

(Koomsap et al., 2019), and the development of Industry 4.0 competences in Estonia
and Tanzania (Kusmin et al., 2018; Mgaiwa, 2021).
H4: There is no recursive nexus between artificial intelligence and educational
opportunities.
The recursive nexus between artificial intelligence and educational
opportunities that exist through artificial intelligence offers limitless opportunities to
students in school and industry experts (Jing, 2018; S Chen, 2018; He and Bowser,
2017). AI can assist school students in obtaining solutions to their most frequently
asked questions within seconds. This not only saves teachers a significant amount of
time but also reduces the time spent by students looking for answers or awaiting a
response to their inquiries. Artificial intelligence is a flourishing field of technology
that has the potential to change peoples’ experiences. In education, the role of AI has
been attributed to its ability to provide answers to the problems of society. This
development is currently being evaluated as a variety of scenarios (S Chen, 2018).
AI requires sophisticated infrastructure and a strong community of innovators. The
goal of the experimental artificial intelligence course conducted under the aegis of
the industry/academic partnership was to strengthen the partnership, as there is a
Granger causality effect of artificial intelligence on educational opportunities. We
believe that artificial intelligence (AI) has the power to moderate, reshape, and
transform the global technological landscape by improving educational opportunities.
AI is being broadly used across all major fields of human endeavor as well as for
national defense and security (Zhang et al., 2018). Consequently, its usage as a
moderating variable to expedite educational outcomes in industry–academia
collaboration is a positive factor in the continuing discussion over industry–academia
collaboration.

2.4. Gaps in literature

Despite the work and progress in the study of university-industry collaboration,
there are at least two knowledge gaps. First, most of the prior research on the
outcomes of collaboration has been primarily concentrated on technological
advancement without taking into account the opportunities that this has presented to
the educational sector as a spillover effect. This suggests that debates on the effects
of R&D collaboration on educational opportunities have been conspicuously absent
from most industry–academia collaboration studies, especially in the study area.
There is still a lack of consensus regarding the manner in which the industry-
academia collaboration enhances educational outcomes in the study area. Second,
this research contributes to the literature as well as practices that are currently in use
by building on past academic conversations and published works of literature.
Specific to China, the investigation of the function that a recursive nexus in artificial
intelligence plays in facilitating industry–academia collaboration to expand
educational opportunities is novel. It adds to the diminutive body of research on how
the influence of artificial intelligence is able to mediate this nexus. Even though the
Chinese use of artificial intelligence is not new, its application in research and
development, product patenting, and curriculum development has been low
compared to other developed countries (He and Bowser, 2017; Ryan, 2021). The

11
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

study equally highlights the directions of the movement of the exogenous variables
as they interact to affect educational opportunities.

3. Research methodology

3.1. Research design

A survey research methodology was used for this study in which structured
questionnaires were distributed to the target population by researcher assistants. The
study protocols were approved by the Chang’an University research committee. For
research of this nature, data can be collected from a variety of secondary databases.
Due to a paucity of crucial information required to answer our research questions,
secondary databases of industry–university collaboration in China do not contain
information on specific activities associated with industry–university collaboration in
the study area. To address study objectives 1–3, a dataset at the firm level that
combines industry and academic collaboration data is necessary which necessitate
the use of survey in obtaining the relevant data.

3.2. Study area

China has a vast geographic area with a wide range in terms of the degree of
regional development. These seven regions were chosen because they have similar
institutional contexts, which made it easier to control the sample selection bias. The
National Bureau of Statistics ranks them as the top seven provinces for GDP
(National Bureau of statistics of China, 2010). Comparatively to businesses in other
locations, these businesses tend to be more contemporary, aggressive, and innovative.
The general managers, CEOs, and R&D managers of these firms were targeted
because they understand their companies’ progress. This indicates that the
information gathered for scholarly research is accurate and trustworthy.

3.3. Sample and sampling technique

To gather data for the academic part of the project, seven (7) universities from
the seven regions were sampled evenly. A total of 327 participants were included in
the sample for this investigation. Knowledge of the topic area, experience, position,
and specialization were used as the criteria for selecting respondents. Also, the
proportion of samples selected by each partner was contingent on the availability of
employee-related R&D projects in any of these businesses. A total of 223 firms and
7 universities agreed to participate in the survey, making a total of 230 research units.
The study sample was not large due to limits on time and availability of willing
respondents to provide information on the research questions of the study. The
questionnaire was divided into two sections: the first collected demographic
information about the respondents, and the second referred to the study objectives.
Out of the 327 respondents included in the study, only 230 completed the survey,
leading to a response rate of 70.34%. Purposive sampling was used to identify the
suitable respondents for the study as earlier stated.

12
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

3.4. Data collection techniques

All participants were adequately informed about the purpose and procedure of
the study and their informed consent were obtained before participation. Random
individuals from Guangdong, Jiangsu, Zhejiang, Shandong, Henan, Beijing, and
Shanghai received questionnaires. IT, communication, business and economics,
electronics, construction, service firms, and manufacturing firms were the major
respondent industries. Data were collected from 13 August to 24 October 2022. The
study’s objectives were captured in a structured questionnaire that was developed by
the area’s firms and universities. Two professors and 5 PhD holders trained to
conduct online and face-to-face surveys collected the data. The questionnaire was
designed using google forms (Esangbedo, Zhang, Esangbedo, Kone and Xu, (2023)),
which is a popular online survey tool for gathering data from respondents, especially
from those located far away. Top managers of the firms and the selected universities
were contacted by the researchers and research assistants by phone and email to
inform them about the purpose of the survey and to solicit for their participation.

3.5. Validity and reliability of the instrument

A pilot study was carried out with one third of the total sample to ensure
representativeness. The pilot study was necessary as it provided the means to obtain
a first information by the researcher. The questionnaire was validated using content
validity which take into account the expert contributions from my supervisor and
statistician (Hair, 2011). Since construct validity statistics require a large
dataset(Ong, 2014), the content validity was assessed. This required the supervisors
to validate the wording of the instrument to ensure it was consistent with the
objectives of the study. Cronbach Alpha statistics were used to determine the
instrument’s internal consistency. The result of the reliability test is as shown in
Table 1:

Table 1. Cronbach alpha reliability coefficients.

Code Constructs Cronbach-α
A Research and Development (R&D)
A1 We are engaged in inventive and methodical efforts to expand the body of human knowledge. 0.9794
A2 The results of research and development are freely transferred or traded on the market. 0.98
B Development and patenting of products
B1 Most patent investment strategies are contingent on the investment’s potential market returns. 0.9791
B2 In terms of product development, there has been a rise in patenting due to increased interaction between the university and 0.9799
industry.
C Curriculum development and delivery
C1 The course description accurately describes the types of responsibilities a graduate can anticipate performing in the workplace. 0.9774
C2 The length of the program is sufficient to provide graduates with the knowledge and/or skills required to enter the field. 0.9781
D Educational opportunities
D1 When industry and academia collaborate, educational opportunities are plentiful. 0.9789
D2 Collaboration increases the industry’s access to qualified workers. 0.9776
E Artificial intelligence utilization
Aint There is substantial investment in artificial intelligence research. 0.9803

13
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Table 1 shows the Cronbach reliability of the instrument. From Table 1, the
statistics for the study’s individual variables show index values of 0.9794 for A1,
0.9800 for A2, 0.9791 for B1, 0.9799 for B2, 0.9774 for C1, 0.9781 for C2, 0.9789
for D1, and 0.9776 for D2. The Artificial Intelligence construct (Aint) index is
0.9803. The overall index of reliability is 0.9813, as shown in Table 1. According to
George and Mallery (2021), a Cronbach Alpha value of 0.70 is reliable for social
science research. In early research, a reliability of 0.70 or higher was sufficient
(Thorndike, 1995). Even with a reliability of 0.90, the standard error of measurement
is almost one-third as large as the standard deviation of the test scores, so 0.90 is the
minimum acceptable reliability, and 0.95 is the desired standard. Thus, this study’s
data collection instrument is reliable.
Table 2 shows the results of an item correlation test that was conducted to
determine whether any test item was inconsistent with the averaged behavior of the
others and could therefore be discarded. The 9-item study value scale reliability was
analyzed. The questionnaire’s Cronbach’s Alpha was 0.9813, which is acceptable.
Most items were shown to be worth keeping; deleting them would lower the alpha
value. No item’s Cronbach value was higher than the overall Cronbach Alpha value.
Deleted items would not improve the overall Cronbach Alpha statistics for any of the
study’s variables.

Table 2. Detailed cronbach alpha statistics.

Item Obs Sign Item-test Item-rest Average inter- alpha
correlation correlation item covariance
A1 230 + 0.926 0.9078 0.32478 0.9794
A2 230 + 0.9158 0.8914 0.31477 0.98
B1 230 + 0.9302 0.9119 0.32061 0.9791
B2 230 + 0.9139 0.8902 0.31835 0.9799
C1 230 + 0.9674 0.9569 0.30475 0.9774
C2 230 + 0.9523 0.938 0.31022 0.9781
D1 230 + 0.9373 0.922 0.32448 0.9789
D2 230 + 0.9595 0.9483 0.31478 0.9776
Aint 230 + 0.9038 0.8792 0.32369 0.9803
Test scale 0.31738 0.9813
Test scale = mean (unstandardized items).

3.6. Variable specification

Definition of constructs: This study used hypothetical constructs. Hypothetical
constructs are not directly observed but are assumed to explain observable
phenomena, according to Colman (2015).
Research and development: Research and development is the process by
which a company generates new knowledge to create new technologies, products,
services, or systems to solve societal challenges.
Product patenting and development: A patent grants investors and others
derivation rights and the right to exclude others from manufacturing, using, or selling

14
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

patented products or methods or processes for a limited period of time (Vedaraman,

1971).
Curriculum development and delivery: Curriculum development is a planned,
thoughtful, and deliberate process that improves student learning. It involves the
development and organization of learning activities to meet learning outcomes using
the best methods.
Educational opportunities: Educational opportunity is defined as anything that
adds value to the educational experience and better prepares you for meeting
academic challenges and challenges posed by the larger society.
Artificial intelligence utilization: Artificial intelligence was used as the
control variable in this study. It was included to establish how artificial intelligence
mediates between IAC and educational opportunities. The simulation of human
intelligence processes by machines, primarily computer systems, is known as
artificial intelligence. In this study, it was measured as the level of spending by firms
on the use of artificial intelligence for business operations.

3.7. Measurement of variables

The measurement items were adapted from Hou, Hong, Chen et al. (2019), who
studied whether academia-industry R&D collaboration promotes industrial
innovation in China with a focus on technology transfer institutions. The constructs
were modified for this study. The questionnaire was based on prior research and
interviews. The study have three independent variables of research and development,
patenting and curriculum development. It also has one moderating variable of
artificial intelligence and one dependent variable of educational opportunities. The
independent variables have six (6) measurement variables. Artificial intelligence and
educational opportunities have two measurement variables respectively. Twelve
error terms are associated with the structural equation model arising from all the
variables of the study.

3.8. Model specification

STATA version 13.0 for windows was used to create a structural equation
model (SEM) to determine the study’s variable relationships.
As shown in Figure 2, there are three independent variables of research and
development, patenting and curriculum development being moderated by artificial
intelligence leading to their impact on the dependent variable of educational
opportunities. The relationships and hypothesized pathways is indicated using arrows
between latent variables and observed variable. These relationships represent the
theoretical connections between the constructs in the model.
Two measurement variables items are associated with each of the construct,
leading to ten measurement items for all the variables used in the study. Twelve error
terms are associated with each variable, accounting for unexplained variance in the
model. Artificial intelligence mediates the relationship between the independent
variables and the dependent variable. The model specifies a and tests the hypothesis
that there is no recursive nexus between artificial intelligence and educational
opportunities.

15
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

3.9. Method of data analysis

Data obtained from this study was analyzed using descriptive statistics such as
frequencies and percentages. The relationship between the variables of the study
were modeled using structural equation model.
Fit indices and model evaluation criteria:
Several SEM model fit indices were used to evaluate the goodness of fit and
appropriateness of the model in modeling the relationship between the study
variables. These indices are:
Comparative Fit Index (CFI): The Comparative Fit Index (CFI) is a statistical
measure in Structural Equation Modeling (SEM) to assess the goodness of fit of a
hypothesized model. It is a comparative index, comparing the fit of the specified
model to a baseline or null model, often a model where the variables are assumed to
be uncorrelated. The CFI ranges from 0 to 1, with higher values indicating a better fit
of the specified model to the observed data. A CFI close to 1 (typically above 0.90 or
0.95) suggests that the hypothesized model fits the observed data well, indicating a
good fit.
Root Mean Square Error of Approximation (RMSEA): The Root Mean
Square Error of Approximation (RMSEA) was used in the study to assess the
goodness of fit of a hypothesized model by evaluating the discrepancy between the
observed data and the model's implied covariance matrix. RMSEA is important
because it accounts for the model’s complexity by adjusting for the degrees of
freedom. It is especially valuable for evaluating how well the model reproduces the
observed covariance matrix, providing insights into the discrepancies between the
model and the data.
Tucker–Lewisindex (TLI): The Tucker-Lewis Index (TLI), also known as the
Non-Normed Fit Index (NNFI), is a goodness-of-fit index used in Structural
Equation Modeling (SEM) to evaluate the fit of a statistical model to the observed
data. It is one of the several fit indices used to assess the adequacy of the model in
explaining the relationships between observed and latent variables.
LR test of model fitness (LR): The Likelihood Ratio (LR) test, also known as
the chi-square difference test, is a statistical test used to assess the goodness of fit of
a structural equation model (SEM) by comparing the fit of a specified model with a
more restricted or nested model. It helps in evaluating whether adding or removing
parameters significantly improves or worsens the model fit.

4. Result and discussion

4.1. Demographic characteristics of the respondents

Figures 3–5 on the sectoral distribution of respondents, show that 144 (62.61%)
of respondents were from the academic sector, while 86 (37.39%) were from
industry, thus giving academia a greater representation in the results of the survey.
Figure 3 shows that the majority of the sampled respondents (209; 90.87%) were
undergraduates, while 21 (9.13%) of the respondents were postgraduate students.
This shows a high level of participation of university students in collaborations
between the university and industry. This proves that forging a partnership at the

16
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

early stages of university education provides a key advantage for enhancing the
educational opportunities of Chinese citizens. More males participated in the survey
than their female counterparts. Males constituted 56.52% of the sample (130
respondents), while female respondents represented 43.48% of the sample (100
respondents).

Figure 3. Sectoral distribution.

Figure 4. Educational distribution.

Figure 5. Gender distribution.

4.2. Summary statistics

Table 3 summarizes all study variables. The study variables have a standard
deviation of less than 1, indicating that their values are close to the population mean.
Our study results are reliable because a high standard deviation indicates that values
are spread out and less reliable (Kollo et al., 2005). The maximum response value
was 4, representing strongly agree, and the minimum value was 1, representing
strongly disagree.

Table 3. Summary of the statistics.

Variable Obs Mean Std. Dev. Min Max
A1 230 3.4 0.5575993 2 4
A2 230 3.15217 0.6463668 1 4

17
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Table 3. (Continued).
Variable Obs Mean Std. Dev. Min Max
B1 230 3.3913 0.5865397 1 4
B2 230 3.17826 0.6188146 1 4
C1 230 3.24783 0.6767309 1 4
C2 230 3.22609 0.6482879 1 4
D1 230 3.33913 0.5510919 1 4
D2 230 3.3 0.6067308 2 4
Aint 230 3.42609 0.5845294 2 4

4.3. Test for multicollinearity

In Table 4, all items show positive covariance between one construct and
another. This implies that these variables move in the same direction in their
relationship (Kollo et al., 2005). For the covariance results, the values are low, less
than 0.5, which implies that they are not highly correlated. A high correlation, for
example, 0.9, could indicate the presence of multicollinearity in the dataset. The
consequence of multicollinearity is that it makes the standard errors much larger than
they would otherwise be, thereby decreasing the t-statistics and increasing the
probability. The consequence of this is that the study variables will not be significant.
The values are low, less than 0.5, which is suggestive of moderate correlation and
low multicollinearity.

Table 4. Correlation matrix.

A1 A2 B1 B2 C1 C2 D1 D2 Aint
A1 0.3109
A2 0.2707 0.4178
B1 0.3188 0.2895 0.344
B2 0.2603 0.3789 0.2749 0.3829
C1 0.3197 0.3944 0.3393 0.3705 0.458
C2 0.2934 0.3847 0.3129 0.3613 0.4153 0.4203
D1 0.2742 0.2844 0.2816 0.2755 0.3392 0.3116 0.3037
D2 0.2943 0.3297 0.31 0.3218 0.3882 0.3598 0.3083 0.3681
Aint 0.3004 0.2799 0.3216 0.2687 0.3263 0.3006 0.2654 0.2996 0.3417

4.4. Model fitness

The criteria for determining whether or not a model is a good fit are listed in
Table 5. The model’s accuracy can be assessed with four different metrics: the root-
mean-squared error of approximation (RMSEA), the comparative fit index (CFI), the
Tucker–Lewis index (TLI), and the LR test of model fitness (TLI). Browne and
Cudeck (1993) suggested that the LR should be greater than 0.05. A small RMSEA
(0.05) indicates a good fit between the hypothesized model and the observed data,

18
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

while a moderate RMSEA (0.05–0.10) indicates a fair fit, and a large RMSEA (>
0.10) indicates a poor fit. However, an RMSEA of 0.06 may indicate a good fit, as
suggested by Hu and Bentler (1999) in the root-mean-squared error of approximation
(RMSEA) formula. Higher values of CFI, a fit index with a range from 0 to 1,
indicate a better fit. Thus, CFI is the most popular measure of compatibility at 0.95
(Hu and Bentler, 1999; West et al., 2012). The TLI (Tucker and Lewis, 1973)
assesses the degree to which there has been a reduction in misfitting. As shown by
the data, the model fitness hypothesis cannot be supported (RMSEA = 0.048). This
means the model was fitted according to this measure of appropriateness. The CFI
and TLI values should be close to 1 to be considered acceptable; these values can be
used as criteria for model fitness (West et al., 2012). The study results show values
of CFI = 0.756 and TLI = 0.582. Evidence from these statistics suggests that the
study’s model is appropriate, and that the study’s estimates can be used to
confidently advise policymakers. Standardized root-mean-squared residual
coefficient of determination.

Table 5. Model fitness.

Fit statistic Value Description
Likelihood ratio chi2_ms (121) p > chi2 chi2_bs 1020.060 model vs. saturated baseline vs. saturated
(36) 4136.97
p > chi2 0.357
Population error RMSEA 0.048 Root mean squared error of approximation
90% CI, lower bound upper 0.037
bound
p-close −0.000 Probability RMSEA ≤ 0.05
Information criteria AIC 767.822 Akaike’s information criterion
BIC 881.279 Bayesian information criterion
Baseline comparison CFI 0.756 Comparative fit index
TLI 0.582 Tucker-Lewis index
Size of residuals SRMR 0.54 Standardized root mean squared residual
CD 0.998 Coefficient of determination

4.5. Model stability test

As shown in Table 6, all eigenvalues lie inside the unit circle (SI = 0.593987),
thus satisfying the structural equation model stability condition. Model stability is a
condition for accepting the results of the model for policy recommendations.

Table 6. Stability test.

Eigenvalue Modulus
0+ 0.5939587i 0.59396
0− 0.5939587i 0.59396
0 0
0 0

19
Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Table 6. (Continued).
Eigenvalue Modulus
0 0
0 0
0 0
0 0
0 0
0 0

All eigenvalues lie inside the unit circle. SEM satisfies the stability conditions.

4.6. Path analysis

Figure 6 depicts a pathway analysis, a method for determining the effects of
multiple variables on a specified outcome. The model shows that research and
development, patenting, and curriculum development are exogenous variables, while
educational opportunities and artificial intelligence usage are endogenous variables.

Figure 6. Path analysis.

The path for the covariance between the exogenous variables in the model is
represented by res-dev, patent, and curr. The recursive nexus between educational
opportunities and artificial intelligence is established. The path shows that all the
exogenous variables are positive and significant predictors of the dependent
variables.
Table 7 shows the conceptual model’s SEM results. SEM models the
relationships between latent variables, not their means, to control for measurement
errors(Preacher and Hayes, 2004). In the first system of the equation, the recursive
nexus between artificial intelligence and educational opportunities [aint & edu]
indicates that educational opportunities is a positive and significant predictor of
artificial intelligence (β = 1.099941, p < 0.05). This result supports the hypothesis
that the deployment of artificial intelligence positively and significantly improves
educational opportunities. The second system of the equation presents the nexus
between the endogenous variable of educational opportunities (eduop) and a set of
three other predictor variables: research and development (res_dev), product

20
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

patenting (patent), and curriculum development (curr). Res_dev positively and

significantly impacts educational opportunities (β = 1.099941, p < 0.05).

Table 7. Structural equation model result.

Coef. OIM z p > |z| [95% Conf. Interval]
Std. Err.
Structural eduop_cons 1.099941 0.0462845 23.76 0.000 1.009224 1.190657
Aint<− 4.628884 0.2671965 17.32 0.000 4.105188 5.152579
eduop<− Aint −0.3207327 0.066768 −4.80 0.000 −0.4515956 −0.1898699
Res_dev 0.4691786 0.0705133 6.65 0.000 0.3309751 0.6073822
Patent 0.3128488 0.0661369 4.73 0.000 0.1832229 0.4424746
Cur 0.4053915 0.0445172 9.11 0.000 0.3181393 0.4926437
var (e.A1) 0.0345934 0.007506 0.02261 0.0529279
var (e.A2) 0.1517086 0.0155744 0.1240584 0.1855216
var (e.B1) 0.044637 0.006473 0.0335938 0.0593103
var (e.B2) 0.1261177 0.0124299 0.1039641 0.1529919
var (e.C1) 0.0064627 0.0060898 0.0010194 0.0409731
var (e.C2) 0.0380115 0.0062344 0.0275617 0.0524233
var (e.D1) 0.0260292 0.0032946 0.0203105 0.033358
var (e.D2) 0.0197908 0.0036084 0.0138441 0.0282918
var (e. Aint) 0.0876614 0.0094441 0.0709749 0.1082708
var (e. eduop) 0.0055586 0.0031074 0.0018583 0.0166269
var (Res_dev) 0.2749718 0.0294761 0.2228652 0.3392612
var (patent) 0.2314285 0.0314248 0.1773518 0.3019939
var (cur) 0.4495108 0.0429452 0.3727507 0.5420782
cov (Res_dev, patent) 0.2720847 0.0290304 9.37 0.000 0.2151863 0.3289832
cov (patent, cur) 0.0630601 0.0190935 3.30 0.000 0.0256374 0.1004827

LR test of model vs. saturated: chi2(21) = 1020.06, Prob > chi2 = 0.0846.
In accordance with the hypothesis, artificial intelligence is a significant
predictor of educational opportunities (β= −0.3207327, p < 0.05). This result
demonstrates a negative correlation between artificial intelligence and educational
opportunities. This may be the case due to a number of factors, including low
investment in science and technology and the absence of national strategies
(Abrahams et al., 2010). The patent was signed in accordance with a priori
expectations (β= 0.3128488, p < 0.05). Curriculum development (cur) was shown to
have a positive impact on educational opportunities in the study area (β = 0.405391,
p < 0.05), and this effect was statistically significant and consistent with a priori
expectations. This indicates that an increase of one unit in [cur] will result in a
0.4053915.
The research and development construct is significantly predicted by item A2 (β
= 0.9803375, p < 0.05), which indicates that the outcomes of R&D are freely
transferred or traded in the market. B2 is a significant predictor of product patenting
(β = 0.9009662, p < 0.05). This indicates that there is an increase in patenting due to
greater interaction between the university and industry in product development.
Curriculum development is positively and significantly affected by C2 (β =
0.9299666, p < 0.05). The D2 measure of how collaboration increases the supply of
quality workers to the industry is a strong indicator of future academic success.

21
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Cov (res dev, patent) and Cov (patent, curr) are positively and significantly
correlated (cov1 = 0.2720847, p = 0.000 and cov2 = 0.0630601, p = 0.001). Positive
covariance indicates that the two variables are positively related and have similar
trends over time. However, because the variables all trend in the same direction,
there is not much of a correlation between them. The sign of the covariance between
two variables specifies the direction of their linear dependence. Coefficients are
positive if there is a general trend toward an increase or decrease in both variables
(Xie and Bentler, 2003).
Specific mechanisms through which AI contributes to educational opportunities
(direct and indirect effect).
This structural equation model (SEM) Table 8 focuses on the direct effects in
the context of how artificial intelligence (Aint) mediates the relationship between the
role of industry-academia collaboration and educational opportunities (EduOp)
within the digital-driven Industry 4.0 landscape. The coefficient (Coef.) of
0.2262638 indicates a statistically significant positive direct effect of artificial
intelligence (Aint) on educational opportunities (EduOp). Aint plays a crucial role in
enhancing educational opportunities, particularly in the context of Industry 4.0. The
statistical significance (p-value < 0.001) and a 95% confidence interval ([0.1764826,
0.2760449]) suggest a robust and reliable effect. The coefficient (Coef.) of
−0.009969 suggests a statistically significant negative direct effect of educational
opportunities (EduOp) on itself. This negative effect may imply a regulatory
mechanism where an increase in educational opportunities could lead to a slight
decrease in further opportunities, possibly due to saturation or other contextual
factors. The statistical significance (p-value < 0.001) and a narrow 95% confidence
interval ([−0.010293, −0.0096451]) indicate a highly reliable effect.

Table 8. Direct effects.

Coef. OIM z p > |z| [95% Conf. Interval]
Std. Err.
Structural
EduOp<-
Aint 0.2262638 0.025399 8.91 0.000 0.1764826 0.2760449
EduOp −0.009969 0.0001653 −60.32 0.000 −0.010293 −0.0096451
Res_dev −0.0089512 0.0116673 −9.75 0.000 −0.0908396 −0.1365746
Patenting 0.1316513 0.0024307 −6.16 0.000 −0.0102057 −0.0197337
Curr_dev 0.7254695 0.026716 27.12 0.000 0.6730372 0.7779017

The coefficient (Coef.) of −0.0089512 suggests a statistically significant

negative direct effect of research and development (Res_dev) on educational
opportunities (EduOp). Increased research and development may have a dampening
effect on educational opportunities, potentially due to resource allocation or other
factors. The coefficient (Coef.) of 0.1316513 suggests a statistically significant
positive direct effect of product patenting (Patenting) on educational opportunities
(EduOp). This implies that product patenting positively influences educational
opportunities, likely through innovation and knowledge dissemination. The

22
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

coefficient (Coef.) of 0.7254695 indicates a statistically significant strong positive

direct effect of curriculum development (Curr_dev) on educational opportunities
(EduOp). Enhancements in the curriculum development significantly boost
educational opportunities, underscoring the critical role of educational structures in
the context of Industry 4.0.
The direct effect of Aint on EduOp in the SEM model suggests that artificial
intelligence (AI) has a positive and statistically significant effect on educational
opportunities, even after controlling for other factors such as research and
development (Res_dev), product patenting (Patenting), and curriculum development
(Curr_dev). Overall, these direct effects illuminate how artificial intelligence and
other key variables directly influence educational opportunities, shedding light on
the dynamics of the relationship within the context of Industry 4.0.
This SEM Table 9 presents the result of the indirect effects in the context of
how artificial intelligence (Aint) mediates the relationship between the role of
industry-academia collaboration and educational opportunities (EduOp) within the
digital-driven Industry 4.0 landscape. The coefficient (Coef.) of 0.0257278 indicates
a statistically significant positive indirect effect of artificial intelligence (Aint) on
educational opportunities (EduOp). This indirect effect suggests that Aint has a
positive influence on EduOp through other pathways or mediating variables. The
statistical significance (p-value < 0.001) and a 95% confidence interval ([0.0200673,
0.0313883]) suggest a robust and reliable indirect effect. The coefficient (Coef.) of
0.1137071 suggests a statistically significant positive indirect effect of educational
opportunities (EduOp) on itself. This indirect effect implies that the level of
educational opportunities positively affects itself through various pathways or
mediated by other variables.

Table 9. Indirect effects.

Coef. OIM z p > |z| [95% Conf. Interval]
Std. Err.
Structural
EduOp<-
Aint 0.0257278 0.002888 8.91 0.000 0.0200673 0.0313883
EduOp 0.1137071 0.0116673 9.75 0.000 0.0908396 0.1365746
Res_dev −0.0010178 0.0001653 −6.16 0.000 −0.0013417 −0.0006939
Patenting 0.0149697 0.0024307 6.16 0.000 0.0102057 0.0197337
Curr_dev 0.082491 0.0115652 7.13 0.000 0.0598236 0.1051584

As shown in Table 10, the statistical significance (p-value < 0.001) and a 95%
confidence interval ([0.0908396, 0.1365746]) indicate a highly reliable indirect
effect. For Res_dev, the coefficient (Coef.) of −0.0010178 suggests a statistically
significant negative indirect effect of research and development (Res_dev) on
educational opportunities (EduOp). This implies that the impact of Res_dev on
EduOp is partially mediated through other variables. However, for the variable
Patenting, the coefficient (Coef.) of 0.0149697 suggests a statistically significant

23
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

positive indirect effect of product patenting (Patenting) on educational opportunities

(EduOp). This implies that the impact of Patenting on EduOp is partially mediated
through other variables. For, Curr_dev, the coefficient (Coef.) of 0.082491 suggests
a statistically significant positive indirect effect of curriculum development
(Curr_dev) on educational opportunities (EduOp). This indicates that the impact of
Curr_dev on EduOp is partially mediated through other variables.Overall, these
indirect effects shed light on the complex interplay of variables, particularly the
mediating role of artificial intelligence (Aint) and the impact of research and
development, product patenting, and curriculum development on educational
opportunities (EduOp) within the context of Industry 4.0. The significant indirect
effects underscore the importance of considering these mediators when exploring the
relationship between industry-academia collaboration and educational outcomes.

Table 10. Total effects.

Coef. OIM z p > |z| [95% Conf. Interval]

Std. Err.
Structural
EduOp<-
Aint 0.2519916 0.0282871 8.91 0.000 0.1965499 0.3074332
EduOp 0.1137071 0.0116673 9.75 0.000 0.0908396 0.1365746
Res_dev −0.009969 0.0001653 −60.32 0.000 −0.010293 −0.0096451
Patenting 0.146621 0.0024307 60.32 0.000 0.141857 0.151385
Curr_dev 0.8079605 0.023006 35.12 0.000 0.7628695 0.8530515

The structural equation model (SEM) total summary table provides crucial
insights into the relationship between the independent variables “Res_dev”,
“Patenting”, and “Curr_dev”, and the dependent variable “EduOp”, while
considering the mediating influence of “Aint”. The analysis reveals a multifaceted
picture of how “Aint” impacts “EduOp”. Firstly, the direct total effect of “Aint” on
“EduOp” is substantial (Coef. 0.2519916) and statistically significant, implying that
changes in “Aint” have a pronounced influence on “EduOp” without any mediation.
Additionally, considering the indirect effects through intermediate variables,
“Res_dev” and “Patenting” play crucial roles, exhibiting significant indirect effects
on “EduOp”. “Res_dev” exhibits a negative indirect effect, suggesting that increased
research and development may dampen the impact of “Aint” on “EduOp”, while
“Patenting” has a positive indirect effect, indicating that innovation, as reflected by
patenting, amplifies the effect of “Aint” on “EduOp”. Notably, “Curr_dev”
demonstrates a remarkable positive indirect effect, underscoring the strong role of
current development activities in enhancing the influence of “Aint” on “EduOp”.

4.7. Discussion
This study answers crucial research questions about how R&D, product
development, patenting, and curriculum development bring about educational
opportunities, contributing to previous research on industry–academia collaboration.

24
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Further, it shows the influences of causal variables and their respective directions.
The first set of equations revealed a positive and statistically significant relationship
(β = 1.099941, p < 0.05) between AI and educational possibilities. This indicates that
access to education can increase the application of AI in teamwork, and vice versa.
In the first set of equations, AI is a significant predictor of educational
opportunities, despite its negative sign (β = −0.3207327, p < 0.05). This is in contrast
to the views of Polt et al. (2001), who credited industry-academia collaboration with
increasing productivity in the workplace. The researchers found that higher levels of
innovation, productivity, competitiveness, and growth (educational opportunities) are
the desired outcomes when university–industry cooperation is used in innovation
processes (such as AI) (Argyropoulou et al., 2019). One of the biggest obstacles to
the implementation of AI in the classroom is the current state of funding and
investment in the field. Several things could be at play here. While artificial
intelligence has the potential to greatly enhance educational opportunities in China,
the country’s low investment in science and technology (such as AI) and lack of
national strategies in these areas make this goal extremely challenging to achieve, as
stated by Abrahams et al. (2010). The “innovation paradox” is not unique to Europe;
it is also present in China. This paradox describes a scenario in which high-quality
scientific outputs or inputs (like AI) have low conversion to productivity
(Argyropoulou et al., 2019; Dosi et al., 2006; Wang and Wang, 2016). The findings
indicate that while China has made some investments in AI, putting those resources
to use to improve educational opportunities has not been a top priority.
There is a connection between the exogenous variable “educational
opportunities” (eduop) and three explanatory variables: “research and development”,
“product patenting” and “curriculum development” (curr). [β > 1 p < 0.05)] . This
finding is consistent with the opinions of Hommaet et al. (2008) and Munyoki et al.
(2011), who argued that one common way that universities link up with industry to
carry out research and development collaborations is by providing opportunities for
student attachments and co-op placements in the productive sector. Student research
projects may also have input from industries if they address problems and issues that
are of immediate relevance to those sectors (Boersma et al., 2008). Similarly, Hou,
Hong, Wang et al. (2018) discovered that R&D collaboration between academic
institutions and commercial businesses boosts innovation productivity. Universities
are the key to science and technology innovation. Patents are an important indicator
of scientific and technological innovation. Positive patent signing is expected (β =
0.3128488, p < 0.05). When product patent and development increases by 1, the
endogenous variable increases by 0.3128488. In the study of Hou, Hong, Wang et al.
(2018), it was found that Chinese universities lack experience in successful industry
collaboration and market understanding, which may lead to inefficient collaborative
outcomes. Positive patenting figures indicate technological knowledge creation,
which benefits education (Griliches, 1990; Nagaoka et al., 2010). Incentives for
patenting in China have been successful (Prud’homme, 2017). This positive effect
could be due to political pressure in China via state-set patent targets tied to the
performance evaluations of managers at State Owned Enterprises and university and
government officials (Liefner et al., 2016; Cheng and Drahos, 2017).

25
Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

In China, the incentivization of patenting has shown considerable success, a

phenomenon attributed to various factors, including a unique blend of government
policies and institutional mechanisms that encourage and reward patenting activities.
This positive effect can be linked to the exertion of political pressure within China’s
governance structure, where state-set patent targets play a pivotal role. One crucial
factor driving this success is the alignment of patenting goals with the performance
evaluations of key stakeholders, such as managers at State Owned Enterprises
(SOEs), university faculties, and government officials. The Chinese government has
set specific patent targets that individuals and organizations are expected to achieve
within a defined timeframe. These targets are tied to performance assessments,
promotions, and other professional advancements. This effectively integrates the
pursuit of patents into the core objectives and career advancement pathways of
professionals within these sectors. As expected, curriculum development (cur) was
found to improve educational opportunities in the study area (β = 0.4053915, p <
0.05). A unit increase in [cur] will increase educational opportunities by 0.4053915.
This may be why instruction has shifted from a teacher-centered input model to one
that emphasizes student–teacher communication.
Cov (res dev, patent), Cov (patent, curr) are positive and significant (cov =
0.2720847, p = 0.000 & covβ = 0.0630601, p = 0.001). Positive covariance means
the two variables are related and move in the same direction. The study’s variables
move in the same direction, but the variance is low, so their relationship is weak.
Covariance determines the direction of the linear relationship. There is a positive
coefficient if both variables increase or decrease together (Xie and Bentler, 2003).
This shows that the study’s exogenous variables move in the same direction
regarding the explanation of educational opportunities.
For the direct and the indirect effect, the SEM results illuminate key dynamics
in educational opportunities (EduOp) within Industry 4.0. Artificial intelligence
(Aint) positively impacts EduOp, while EduOp’s negative direct effect on itself hints
at potential regulatory mechanisms. Research and development (Res_dev) negatively
affect EduOp, emphasizing the need for a balanced research approach. Conversely,
product patenting (Patenting) and curriculum development (Curr_dev) exhibit
positive direct effects on EduOp, highlighting the role of innovation and structured
curricula in enhancing educational prospects. These insights emphasize leveraging
AI, managing research effectively, and innovating product strategies and curricula to
boost educational outcomes in the evolving digital era. For the indirect effect result,
the SEM results underscore the significant indirect impact of artificial intelligence
(Aint) on educational opportunities (EduOp) within the context of Industry 4.0. This
indirect effect emphasizes the pivotal role of AI as a mediator, influencing EduOp
through various pathways. Additionally, the positive indirect effect of EduOp on
itself implies a self-enhancing relationship, suggesting that educational opportunities
contribute to their own augmentation. Conversely, research and development
(Res_dev) exhibit a negative indirect effect on EduOp, signifying a need for careful
management to ensure a positive impact. Furthermore, product patenting (Patenting)
and curriculum development (Curr_dev) demonstrate positive indirect effects on
EduOp, underlining the importance of innovation and structured curriculum in
shaping educational prospects within the Industry 4.0 landscape. These insights

26
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

emphasize the intricate interplay of factors and the critical mediating role of AI,
providing valuable guidance for optimizing educational opportunities amidst the
evolving digital landscape.

5. Conclusion and recommendations

The study examined the role of industry–academia collaboration in enhancing
educational opportunities and outcomes under the digital driven Industry 4.0 using
the proxies of research and development, patenting and development, curriculum
development mediated by artificial intelligence.The study was anchored on double
helix theory and relevant conceptual and empirical studies were reviewed to provide
the empirical background to the study. Purposive sampling was used while data was
collected using questionnaire. Validity and reliability of the instrument was done
using content validity and reliability of the instrument was accessed Cronbach Alpha
statistics. The study was estimated using descriptive and inferential statistics. The
inferential statistics deals with frequencies and percentages. Other results were
presented using graph. The results of the descriptive statistics showed the
distributions of the respondents in the study area. The specific objectives of the study
were estimated using structural equation model.
This study significantly contributes to the Triple Helix hypothesis by
investigating the pivotal role of industry–academia collaboration in augmenting
educational opportunities and outcomes within the digitally driven Industry 4.0
paradigm. Grounded in the Double Helix theory, the research employs proxies such
as research and development (R&D), patenting, and curriculum development,
mediated by artificial intelligence (AI), to delineate the intricate relationship between
academia and industry. Drawing on empirical evidence and conceptual frameworks,
the study underscores universities as not only conduits for disseminating knowledge
and nurturing innovative talent but also as crucibles for knowledge and technology
innovation. The findings illuminate that collaborative efforts between academia and
industry, fortified by AI, engender a virtuous cycle, positively influencing
educational opportunities. Notably, the results underscore the burgeoning
advancements in China, positioning the nation to stride alongside Western
economies in the digital economy of the 21st century, fueled by robust R&D,
patenting, and curriculum development. Furthermore, the research emphasizes the
imperative for increased investment and attention directed towards integrating AI
into educational pursuits within the studied domain. The study advocates for
resource allocation from governmental bodies and stakeholders to harness the
potential of AI in amplifying educational opportunities through industry–academic
collaboration. The insights gleaned from this study are paramount for businesses,
governments, and policymakers, urging a paradigm shift towards prioritizing
Industry–Academia Collaboration (IAC) to enrich educational prospects, thus
propelling socioeconomic development and sustainability, particularly in the era of
Industry 4.0.

Author contributions: Conceptualization of ideas and writing the first draft of the
manuscript with computational and data analysis, COE; Research supervision and

27
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

formulated this article’s objectives and general goals, certifying the correctness of
the manuscript with funding acquisition, JZ; Re-writing the manuscript research
design, statistical work, analysis of the results, and interpretation of the results, MOE;
Questionnaire design and data curation, SDK; Funding acquisition and confirmatory
data analysis, LX. All authors have read and agreed to the published version of the
manuscript.
Additional information: Authors report all methods were carried out in accordance
with relevant guidelines and regulations of Chang’an University.
Funding: This research is supported by the National Social Science Fund projects of
China (No.20BJY010); the National Social Science Fund Post-financing projects of
China (No.19FJYB017); the China Sichuan-Tibet Railway Major Fundamental
Science Problems Special Fund (No.71942006); the China Qinghai Natural Science
Foundation (No.2020-JY-736); the List of Key Science and Technology Projects in
China’s Transportation Industry in the 2018 International Science and Technology
Cooperation Project (No.2018-GH-006 and No.2019-MS5-100); the Emerging
Engineering Education Research and Practice Project of Ministry of Education of
China (No.E-GKRWJC20202914); the Higher Education Teaching Reform Project
in Shaanxi Province, China (No.19BZ016); the Humanities and Social Sciences
Research Project of the Ministry of Education of China (21XJA752003); the Going
Global Partnership: UK-China-ASEAN, Education Partnership Initiative funded by
British Council (“Integrated Built Environment Teaching & Learning in the Joint
Curriculum Development amid Digital-Driven Industry 4.0 among China, Vietnam,
and UK” ); the International Education Research Program of Chang’an University,
China, 2022 (No. 300108221113); and the National Natural Science Foundation of
China (No. 72074191).
Data availability statement: The datasets used and/or analyzed during the current
study available from the corresponding author on reasonable request. Please contact
the corresponding author (Jingxiao Zhang: zhangjingxiao@chd.edu.cn).
Conflicts of interest: The authors declare no conflict of interest.

References
Abrahams L, Burke M, Mouton J (2010). Research productivity-visibility-accessibility and scholarly communication in Southern
African universities. The African Journal of Information and Communication (AJIC) 10. doi: 10.23962/10539/19768
Agrawal AK (2001). University-to-industry knowledge transfer: Literature review and unanswered questions. International
Journal of Management Reviews 3(4): 285–302. doi: 10.1111/1468-2370.00069
Aiello F, Cardamone P, Pupo V (2019). New evidence on the firm-university linkages in Europe. The role of meritocratic
management practices. International Review of Applied Economics 33(6): 813–828. doi: 10.1080/02692171.2019.1608917
Aishath MA, Chaithra BK, Nishmitha B, et al. (2019). Survey on artificial intelligence. International Journal of Computer
Sciences and Engineering 7(5): 1778–1790. doi: 10.26438/ijcse/v7i5. 17781790
Albahari A, Pérez-Canto S, Barge-Gil A, et al. (2017). Technology Parks versus Science Parks: Does the university make the
difference? Technological Forecasting and Social Change 116: 13–28. doi: 10.1016/j.techfore.2016.11.012
Alexander A, Martin DP, Manolchev C, et al. (2020). University–industry collaboration: Using meta-rules to overcome barriers to
knowledge transfer. The Journal of Technology Transfer 45(2): 371–392. doi: 10.1007/s10961-018-9685-1
Ankrah SN, Al-Tabbaa O (2015). Universities-industry collaboration: A systematic review. Scandinavian Journal of Management
31: 387–408. doi: 10.2139/ssrn.2596018

28
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Apiyo R, Kiarie D (2018). Role of ICT tools in supply chain performance. International Journal of Supply Chain Management
3(1): 17–26 (2018).
Argyropoulou M, Soderquist KE, Ioannou G (2019). Getting out of the European Paradox trap: Making European research agile
and challenge driven. European Management Journal 37(1): 1–5. doi: 10.1016/j.emj.2018.10.005
Balconi M, Laboranti A (2006). University–industry interactions in applied research: The case of microelectronics. Research
Policy 35(10): 1616–1630. doi: 10.1016/j.respol.2006.09.018
Belitski M, Aginskaja A, Marozau R (2019). Commercializing university research in transition economies: Technology transfer
offices or direct industrial funding? Research Policy 48(3): 601–615. doi: 10.1016/j.respol.2018.10.011
Biba E (2016). Three Ways Artificial Intelligence is Helping to Save the World. Ensia.
Boersma FK, Reinecke CJ, Gibbons M (2008). Organizing the University–Industry Relationship: A case study of research policy
and curriculum restructuring at the North‐West University in South Africa. Tertiary Education and Management 14(3): 209–
226. doi: 10.1080/13583880802228216
Breese R (2012). Benefits realisation management: Panacea or false dawn? International Journal of Project Management 30(3):
341–351. doi: 10.1016/j.ijproman.2011.08.007
Cai Y, Etzkowitz H (2020). Theorizing the Triple Helix model: Past, present, and future. Triple Helix Journal 6(1): 1–38. doi:
10.1163/21971927-bja10003
Calder ES (2007). Best Practices for University-Industry Collaboration [PhD thesis]. Massachusetts Institute of Technology.
Chen L, Chen P, Lin Z (2020). Artificial intelligence in education: A review. IEEE Access 8: 75264–75278. doi:
10.1109/access.2020.2988510
Chen S (2018). China’s schools are quietly using AI to mark students’ essays… but do the robots make the grade? Society, South
China Morning Post, 27 May 2018.
Cheng W, Drahos P (2017). How China built the world’s biggest patent office: The pressure driving mechanism. IIC-International
Review of Intellectual Property and Competition Law 49(1): 5–40. doi: 10.1007/s40319-017-0655-1
Cohen WM, Levinthal DA (1989). Innovation and learning: The two faces of R & D. The Economic Journal 99(397): 569–596.
doi: 10.2307/2233763
Colman AM (2015). A Dictionary of Psychology. Oxford University Press.
Comodi G, Cioccolanti L, Mahkamov K, et al. (2019). Analysis of labour market needs for engineers with enhanced knowledge in
renewable energy in some European and Latin-American Countries. Energy Procedia 158: 1135–1140. doi:
10.1016/j.egypro.2019.01.279
Cunningham JA, Link AN (2014). Fostering university-industry R&D collaborations in European Union countries. International
Entrepreneurship and Management Journal 11(4): 849–860. doi: 10.1007/s11365-014-0317-4
D’Este P, Patel P (2007). University–industry linkages in the UK: What are the factors underlying the variety of interactions with
industry? Research Policy 36(9): 1295–1313. doi: 10.1016/j.respol.2007.05.002
Deborah R (2011). Partnering industry and education for curricular enhancement: A response for greater educational achievement.
Online Journal of Workforce Education and Development, 2, 1-15.
Dill DD (1995). University-industry entrepreneurship: The organization and management of American university technology
transfer units. Higher Education 29(4): 369–384. doi: 10.1007/bf01383958
Dooley L, Kirk D (2007). Universityindustry collaboration. European Journal of Innovation Management 10(3): 316–332. doi:
10.1108/14601060710776734
Dopson LR, Tas RF (2004). A practical approach to curriculum development: A case study. Journal of Hospitality & Tourism
Education 16(1): 39–46. doi: 10.1080/10963758.2004.10696783
Dosi G, Llerena P, Labini MS (2006). The relationships between science, technologies and their industrial exploitation: An
illustration through the myths and realities of the so-called ‘European Paradox.’ Research Policy 35(10): 1450–1464. doi:
10.1016/j.respol.2006.09.012
Edmondson G, Valigra L, Kenward M, et al. (2012). Making Industry-University Partnerships Work. Lessons from Successful
Collaborations. Business Innovati on Board.
Estevez J, Garate G, Grana M (2019). Gentle introduction to artificial intelligence for high-school students using scratch. IEEE
Access 7: 179027–179036. doi: 10.1109/access.2019.2956136
Esangbedo, C. O., Zhang, J., Esangbedo, M. O., Kone, S. D., & Xu, L.. Industry-academia collaboration survey. 6th March, 2023.
From https://bit.ly/3poGbVm

29
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Fernández López S, Pérez Astray B, Rodeiro Pazos D, et al. (2014). Are firms interested in collaborating with universities? An
open-innovation perspective in countries of the South West European Space. Service Business 9(4): 637–662. doi:
10.1007/s11628-014-0243-0
Fischer BB, Schaeffer PR, Vonortas NS (2019). Evolution of university-industry collaboration in Brazil from a technology
upgrading perspective. Technological Forecasting and Social Change 145: 330–340. doi: 10.1016/j.techfore.2018.05.001
George D, Mallery P (2021). IBM SPSS Statistics 27 Step by Step. Routledge. doi: 10.4324/9781003205333
Griliches Z (1990). Patent statistics as economic indicators: A survey. Journal of Economic Literature 28: 1661–1707. doi:
10.3386/w3301
Guimón J (2013). Promoting University-Industry Collaboration in Developing Countries. Word Bank.
Hair JF (2011). Multivariate data analysis: An overview. International Encyclopedia of Statistical Science 904–907. doi:
10.1007/978-3-642-04898-2_395
He Y, Bowser A (2017). How China is Preparing for an AI-powered Future. Wilson Briefs.
Homma H, Ikeda N, Attalage RA (2008). Strengthening university-industry linkage in developing countries through international
cooperation: Case of Sri Lanka through cooperation of Toyohashi University of Technology, Japan. In: Iskander M (editor).
Innovative Techniques in Instruction Technology, E-learning, E-assessment, and Education. Springer. pp. 432–436. doi:
10.1007/978-1-4020-8739-4_76
Hong J, Hong S, Wang L, et al. (2015). Government grants, private R&D funding and innovation efficiency in transition economy.
Technology Analysis & Strategic Management 27(9): 1068–1096. doi: 10.1080/09537325.2015.1060310
Hong W (2008). Decline of the center: The decentralizing process of knowledge transfer of Chinese universities from 1985 to
2004. Research Policy 37(4): 580–595. doi: 10.1016/j.respol.2007.12.008
Hou B, Hong J, Chen Q, et al. (2019). Do academia-industry R&D collaborations necessarily facilitate industrial innovation in
China? European Journal of Innovation Management 22(5): 717–746. doi: 10.1108/ejim-09-2018-0195
Hou B, Hong J, Wang H, et al. (2018). Academia-industry collaboration, government funding and innovation efficiency in
Chinese industrial enterprises. Technology Analysis & Strategic Management 31(6): 692–706. doi:
10.1080/09537325.2018.1543868
Hu L, Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new
alternatives. Structural Equation Modeling: A Multidisciplinary Journal 6(1): 1–55. doi: 10.1080/10705519909540118
Hughes A, Kitson M (2012). Pathways to impact and the strategic role of universities: New evidence on the breadth and depth of
university knowledge exchange in the UK and the factors constraining its development. Cambridge Journal of Economics
36(3): 723–750. doi: 10.1093/cje/bes017
Jing M (2018). China wants to bring artificial intelligence to its classrooms to boost its education system. Science & Research, 14
October 2017.
Khachoo Q, Sharma R, Dhanora M (2018). Does proximity to the frontier facilitate FDI-spawned spillovers on innovation and
productivity? Journal of Economics and Business 97: 39–49. doi: 10.1016/j.jeconbus.2018.03.002
Kim YC, Rhee M, Kotha R (2019). Many hands: The effect of the prior inventor-intermediaries relationship on academic
licensing. Research Policy 48(3): 813–829. doi: 10.1016/j.respol.2018.11.007
Kollo T, von Rosen D, Hazewinkel M (2005). Advanced Multivariate Statistics with Matrices. Springer Netherlands. doi:
10.1007/1-4020-3419-9
Koomsap P, Luong HT, Lima RM, et al. (2019). Roles of MSIE graduates to support Thailand sustainable smart industry. In:
Transdisciplinary Engineering for Complex Socio-technical Systems. IOS Press. doi: 10.3233/atde190110
Kusmin KL, Tammets K, Ley T (2018). University-industry interoperability framework for developing the future competences of
Industry 4.0. Interaction Design and Architecture(s) (38): 28–45. doi: 10.55612/s-5002-038-002
Laursen K, Salter A (2004). Searching high and low: What types of firms use universities as a source of innovation? Research
Policy 33(8): 1201–1215. doi: 10.1016/j.respol.2004.07.004
Leydesdorff L (2004). The university–industry knowledge relationship: Analyzing patents and the science base of technologies.
Journal of the American Society for Information Science and Technology 55(11): 991–1001. doi: 10.1002/asi.20045
Li T, Zhao Z, Wang Y, et al. (2020). The status of university-industry collaboration in China, EU and USA—A comparative
research on co-authored publications (2020). In: Proceedings of 2020 ASEE Virtual Annual Conference Content Access; 22–
26 June 2020.

30
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Liefner I, Kroll H, Peighambari A (2016). Research-driven or party-promoted? Factors affecting patent applications of private
small and medium-sized enterprises in China’s Pearl River Delta. Science and Public Policy 43(6): 849–858. doi:
10.1093/scipol/scw002
Liu X, Schwaag Serger S, Tagscherer U, et al. (2017). Beyond catch-up—Can a new innovation policy help China overcome the
middle income trap? Science and Public Policy 44(5): 656–669. doi: 10.1093/scipol/scw092
Luan C, Zhou C, Liu A (2010). Patent strategy in Chinese universities: A comparative perspective.Scientometrics 84(1): 53–63.
doi: 10.1007/s11192-010-0194-8
Lucietto A, Peters D, Taleyarkhan M, et al. (2021). Academic and industry collaboration: A literature review. In: School of
Engineering Education Faculty Publications. Purdue University.
Manyika J, Lund S, Chui M, et al. (2017). Jobs Lost, Jobs Gained: What the Future of Work Will Mean for Jobs, Skills, and
Wages. McKinsey Global Institute.
Mazzocchi S (2004). Open innovation: The new imperative for creating and profiting from technology. Innovation 6(3): 474. doi:
10.5172/impp.2004.6.3.474
Mears L, Omar M, Kurfess TR (2009). Automotive engineering curriculum development: Case study for Clemson University.
Journal of Intelligent Manufacturing 22(5): 693–708. doi: 10.1007/s10845-009-0329-z
Mgaiwa SJ (2021). Fostering graduate employability: Rethinking Tanzania’s university practices. SAGE Open 11(2):
215824402110067. doi: 10.1177/21582440211006709
Miller K, McAdam R, McAdam M (2016). A systematic literature review of university technology transfer from a quadruple helix
perspective: Toward a research agenda. R&D Management 48(1): 7–24. doi: 10.1111/radm.12228
Ministry of Education (2019). Annual Report on the National Higher Education. Ministry of Education.
Ministry of Science and Technology China (2020). Report of the Ministry of Science and Technology of the People’s Republic of
China on the Development of Industry-Academia Collaboration in 2020 (Chinese). Ministry of Science and Technology.
Motohashi K (2006). China’s national innovation system reform and growing science industry linkage. Asian Journal of
Technology Innovation 14(2): 49–65. doi: 10.1080/19761597.2006.9668618
Motohashi K, Muramatsu S (2012). Examining the university industry collaboration policy in Japan: Patent analysis. Technology
in Society 34(2): 149–162. doi: 10.1016/j.techsoc.2012.02.006
Mukherji N, Silberman J (2021). Knowledge flows between universities and industry: The impact of distance, technological
compatibility, and the ability to diffuse knowledge. The Journal of Technology Transfer 46(1): 223–257. doi:
10.1007/s10961-019-09770-9
Munyoki J, Kibera F, Ogutu M (2011). Extent to which university- industry linkage exists in Kenya: A study of medium and large
manufacturing firms in selected industries in Kenya. Business Administration and Management 1(4): 163–169.
Murphy R (2020). Artificial intelligence applications to support K-12 teachers and teaching: A review of promising applications,
challenges, and risks. RAND Corporation. doi: 10.7249/pe315
Nagaoka S, Motohashi K, Goto A (2010). Patent statistics as an innovation indicator. Handbook of the Economics of Innovation 2:
1083–1127. doi: 10.1016/s0169-7218(10)02009-5
National Bureau of statistics of China (2010). China Statistical Year Book. Tech. Rep.
Naughton B, Tsai KS (2015). State Capitalism, Institutional Adaptation, and the Chinese Miracle. Cambridge University Press.
doi: 10.1017/cbo9781139962858
O’Dwyer M, Filieri R, O’Malley L (2023). Establishing successful university–industry collaborations: Barriers and enablers
deconstructed. The Journal of Technology Transfer 48(3): 900–931. doi: 10.1007/s10961-022-09932-2
Payne M (2007). Benefits Management: Releasing Project Value into the Business. Project Manager Today.
Perkmann M, Neely A, Walsh K (2011). How should firms evaluate success in university–industry alliances? A performance
measurement system. R&D Management 41(2): 202–216. doi: 10.1111/j.1467-9310.2011.00637.x
Perkmann M, Tartari V, McKelvey M, et al. (2012). Academic engagement and commercialization: A review of the literature on
university-industry relations. Research Policy 42(2): 423–442. doi: 10.2139/ssrn.2088253
Perkmann M, Walsh K (2007). University–industry relationships and open innovation: Towards a research agenda. International
Journal of Management Reviews 9(4): 259–280. doi: 10.1111/j.1468-2370.2007.00225.x
Polt W, Rammer C, Gassler H, et al. (2001). Benchmarking industry-science relations: The role of framework conditions. Science
and Public Policy 28(4): 247–258. doi: 10.3152/147154301781781453

31
 Journal of Infrastructure, Policy and Development 2024, 8(1), 2569.

Ponomariov B (2013). Government-sponsored university-industry collaboration and the production of nanotechnology patents in
US universities. The Journal of Technology Transfer 38(6): 749–767. doi: 10.1007/s10961-013-9301-3
Preacher KJ, Hayes AF (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior
Research Methods, Instruments, & Computers 36(4): 717–731. doi: 10.3758/bf03206553
Prud’homme D (2017). Utility model patent regimes and innovation in China and Beyond*. In: Patents and Innovation in China
and Hong Kong. Cambridge University Press. pp. 29–78. doi: 10.1017/9781108163583.004.
Rahm D, Kirkland J, Bozeman B (2000). University-Industry R&D Collaboration in the United States, the United Kingdom, and
Japan. Springer Netherlands. doi: 10.1007/978-94-015-9574-2
Rosell C, Agrawal A (2009). Have university knowledge flows narrowed? Research Policy 38(1): 1–13. doi:
10.1016/j.respol.2008.07.014
Ryan S (2021). The U.S., China, and Artificial Intelligence Competition Factors. China Aerospace Studies Institute.
Scandura A (2016). University–industry collaboration and firms’ R&D effort. Research Policy 45(9): 1907–1922. doi:
10.1016/j.respol.2016.06.009
Shewakena Tessema B (2017). University-industry collaboration in curriculum development: Analysis of banking and finance
graduates’ attributes from educators and industries perspective. Education Journal 6(2): 87. doi: 10.11648/j.edu.20170602.13
Smith K (2006). Measuring Innovation. Oxford Handbooks. doi: 10.1093/oxfordhb/9780199286805.003.0006
Thorndike RM (1995). Book review: Psychometric theory (3rd ed.) by Jum Nunnally and Ira Bernstein New York: McGraw-Hill,
1994, xxiv + 752 pp. Applied Psychological Measurement 19(3): 303–305. doi: 10.1177/014662169501900308
Tucker LR, Lewis C (1973). A reliability coefficient for maximum likelihood factor analysis. Psychometrika 38(1): 1–10. doi:
10.1007/bf02291170
United Nations Educational, Scientific and Cultural Organization (UNESCO) (2019). ICT in Education: A Global Outlook.
United Nations Educational, Scientific and Cultural Organization (UNESCO).
Vedaraman S (1971). Patents: Recent Developments and Future Prospects on the National Level in India. WIPO Lectures,
Montreaux
Wang C, Wang L (2016). Unfolding policies for innovation intermediaries in China: A discourse network analysis. Science and
Public Policy 44(3): 354–368. doi: 10.1093/scipol/scw068
Weagle D, Ortendahl D, Ahern AP (2019). Universities and industries: A proactive countries. Innov. Solutions for Energy
Transitions 158: 1135–1140. Doi: 10.1016/j.egypro.2019.01.279.l
West SG, Taylor AB, Wu W (2012). Model fit and model selection in structural equation modeling. In: Hoyle RH (editor).
Handbook of Structural Equation Modeling. The Guilford Press. pp. 209–231.
Xie J, Bentler PM (2003). Covariance structure models for gene expression microarray data. Structural Equation Modeling: A
Multidisciplinary Journal 10(4): 566–582. doi: 10.1207/s15328007sem1004_5
Zhang L, Wang H, Li Q, et al. (2018). Big data and medical research in China. BMJ 360: j5910. doi: 10.1136/bmj.j5910
Zheng Y, Wang Y (2021). Research on the contribution rate of scientific and technological progress to Chongqing’s economic
growth based on the Solow Growth model. Proceedings of E3S Web of Conferences 235: 01012. doi:
10.1051/e3sconf/202123501012
Ziegler WL (1983). Computer science education and industry: Preventing educational misalignment. In: The Proceedings of the
Twentieth Annual Computer Personnel on Research Conference; 17–18 November 1983; New York, NY, United States. doi:
10.1145/800030.800626

32