This study examines how two collaboration channels – education and training and research and development (R&D) – within university–business collaboration (UBC) are associated with firms’ innovation, commercialisation, talent access and efficiency outcomes.
Using a nationwide survey of 4,188 academics and multilevel logit models (GLMM), we estimate the odds of four firm-level outcomes and report odds ratios (ORs) with 95% confidence intervals to assess the magnitude and significance of UBC effects.
Collaborative R&D initiatives significantly enhance the commercialisation of academic inventions and the efficiency of business operations. Engagements in education- and training-oriented UBC are positively associated with product and process innovation and firms’ access to student talent, underscoring the human–capital dimension of collaboration. The results highlight the complementary roles of R&D and education-based collaboration in fostering innovation and organisational performance, while indicating that some structural and relational barriers continue to constrain UBC effectiveness.
Outcomes are reported by academics and anchored in documented project outputs; hence, the industry-level effects are assessed from the perspective and perceptions of the academic respondents rather than from direct firm data. We therefore acknowledge the single-source nature of the dataset and discuss potential implications for interpretation and generalisability.
This paper provides one of the first large-scale analyses of individual-level human capital in UBC within the Turkish context, distinguishing between education and training and R&D channels and linking them to specific firm-level outcomes through a multilevel modelling approach.
1. Introduction
Table A1, A2 and A3University business collaborations (UBCs [1]) have been extensively studied, revealing their significant impact on enhancing business innovation performance (Davey et al., 2024). Such partnerships grant companies access to cutting-edge research and advanced technologies, which are often beyond their internal capabilities (Lam, 2007). By integrating academic scientific knowledge with their production capacity and market insights, businesses can develop new products and services more effectively (Mansfield, 1991; Caloghirou et al., 2001). This synergy not only bridges internal capability gaps but also provides external validation, leading to successful breakthrough innovations. Although it has been shown that university business collaborations are important in driving forward university innovation and commercialisation, these collaborations, however, introduce complexities in validating outcomes. Integrating diverse knowledge requires effective coordination and advanced organisational skills to manage interdisciplinary R&D projects (Temel et al., 2013). Although this complexity may slow product or service development, it does not reduce overall innovation output. This particularly true relation to technical university settings in developing and emerging countries where linkages are central to effective technology transfer (Hoc and Trong, 2019). Businesses by strengthening their ties with universities, intermediaries and research organisations gain access to pioneering research, lower R&D costs and mitigate risks associated with uncertain outcomes (Howells et al., 2012). The impact of university–industry collaborations on a firm's innovative performance can however vary with firms with more available resources and technological capabilities being better positioned to leverage the benefits of university collaborations. These firms can more effectively assimilate and apply external knowledge, resulting in enhanced innovation performance (Caloghirou et al., 2004).
On this basis, the key focus of this paper is on the issue of what set of factors influence the performance of firms undertaking such collaborations? A particular, but neglected, aspect of firm collaborations with universities, which is investigated here, is the role of human capital in influencing these set of relationships. It is argued here that human capital is important in developing successful UBC and this, in turn, raises the question of how human capital formation within universities can be enhanced and also measured. A set of key questions around human capital as a key driver are formulated in four hypotheses, which are outlined in the next section. Türkiye provides a theoretically interesting setting exhibiting an institutional framework with: rapid expansion of university capacity; national policies that institutionalised TTOs; the creation of science parks and placement programmes; and strong regional heterogeneity. These features allow us to observe the human–capital channel of UBC at scale. On this basis, we address three questions:
To what extent do education/training collaborations increase the probability of firms innovation and talent access?
To what extent do R&D collaborations increase the probability of operational efficiency improvements?
Are these associations robust across academic disciplines and institutional contexts?
Our contributions are threefold in that we:
provide a micro-level test of the human-capital channel of UBC by using a large national sample of academics;
disaggregate UBC into education and training vs R&D and link each to distinct firm outcomes and
offer evidence from Turkey's policy-dense environment, informing how education-centric UBC practices (such as co-teaching, placements and curriculum co-design) translate into measurable firm benefits.
The rest of the paper is organised as follows. Section 2 provides an outline and review of the key literature relating to industry academic links, in particular focusing on how human–capital influences such developments and outcomes and introducing the research hypotheses of the study. The following section then outlines the research methodology and data set (Section 3). Section 4 then describes the empirical analysis covering the results of the regression analysis, whilst Section 5 provides a conclusion and discussion of the results and its implications for further research.
2. Literature review and hypothesis setting
2.1 Literature review
The effectiveness of collaborations between companies and academic institutions is influenced by several factors including, product technological sophistication, company age and the scientific understanding of the intended consumers. Such partnerships are particularly advantageous for emerging firms and high-tech products, as they help bridge internal capability gaps (Ketata et al., 2015), especially in weaker institutions (Hsu et al., 2015), and also helps provide external validation. However, equally, studies have found firms with low, or no, R&D activity have higher propensities to forge links with universities and can value them more highly (Howells et al., 2012). Using R&D involving university collaboration, can also enhance a product's perceived value and scientific credibility, leading to increased consumer interest and market acceptance. Nonetheless, significant national and sectoral differences in the use and perceived value of collaborations with universities by firms remain (Ponomariov and Boardman, 2010).
One aspect of firm collaborations with universities, which has only recently started to be explored, is the role of human capital in influencing these collaborative networks. Human capital can be defined as the collective skills, knowledge, experience and attributes of individuals that contribute to their economic value and productivity. In this context, human capital includes education, training, intelligence, creativity and social skills, which enhance an individual's ability to perform work and generate economic output. Studies focusing on human capital mainly emerged from research that explored these relationships in terms of regional economic development (Marotta, 2007; O’Neill and Bagchi-Sen, 2023) but have subsequently expanded to investigate human capital factors in their own right. Thus, human capital in all its various forms allows the effective bridging of knowledge between universities and businesses that exhibits entrepreneurialism (Bozeman et al., 2013) and leads to commercialisation success (Baycan and Stough, 2013). Human capital, including its links to wider institutional and cultural factors, was revealed by Hsu et al. (2015) to be the most important factor in driving university commercialisation performance in Taiwan, with quality of the academics being the key human capital performance driver improving the performance of university technology transfer. Liu (2014), in an industry study, also highlights the scientific quality of individual scientists in influencing performance. More recently, both Apa et al. (2021) and Hsu et al. (2015) highlight the role of absorptive capacity (Cohen and Levinthal, 1989, 1990) in terms of facilitating such links. Indeed, it could be expected that enhanced human capital capability of a firm could, in turn, improve the absorptive capacity of firms working with universities. In this study, we conceptualise human capital as a mechanism that enables firms to better absorb university-generated knowledge. Education- and training-based collaborations enhance this absorptive capacity by embedding academics and students directly in firms' learning processes.
A later strand has been to unpack human capital more specifically in its own terms, especially when exploring collaboration under open innovation environments, where formal and informal trust-building is key (Baban et al., 2021). Thus, previous studies of university business links tended to treat the Higher Education Institution (HEI) or university as a whole with descriptors remaining solely at the university level. Interest in intra-university, institutional variations has only come recently with, for example, examining the role that different academic disciplines and their scientific and institutional structures had on shaping such interactions (Gutierrez et al., 2025). In turn, the focus on the individual level has been more recent still (Kwiek, 2019), influenced by entrepreneurial research which has highlighted the educational and cultural background of the founders of enterprises that influenced the success and performance of the firm they had established (Bosio et al., 2018). de Frutos-Belizón et al. (2023) explore different several sets of academic traits based on complementary competences and linking this back to human capital capacity. A study by Liu (2014) has also highlighted the importance of supporting social and organisational factors that surround human capital being important in its formation and influence on innovativeness and commercialisation. This reflects back to the need to expand the notion of human capital to cover softer skills, including the ability to develop new network ties, increase know-how and tacit knowledge capacity and improve acquiring and managing resources (Bozeman and Corley, 2004). Taken together, prior research indicates that individual academics' skills, experience and training can help shape firms' capacity to absorb and apply external knowledge. However, most studies rely on firm-side or institutional data and rarely isolate the human-capital channel of UBC. Little is known about how academics' engagement in education- and training-oriented collaborations translates into measurable firm outcomes. Addressing this gap motivates our hypotheses.
The Turkish innovation system provides a rich context for testing these mechanisms. Rapid expansion of university R&D infrastructure, the establishment of TTOs, science parks and the policy emphasis on co-education and internships create a fertile environment in which to observe how the human-capital dimension of UBC affects firms.
2.2 Hypothesis setting
This study seeks to explore in more depth the role of human capital in supporting business performance by universities and in commercialisation activities. Human capital activity in this context, focuses on the role of education and training in developing a wider envelope of good human capital traits, including activities such as student mobility, dual education programs, co-design and co-delivery of curricula and collaboration on lifelong learning initiatives. As such, exposure to such activities can improve problem-solving abilities and technical skills enabling academics to adapt their research to meet industry demands (Galan-Muros and Davey, 2019). Studies analysing UBC, particularly in engineering sciences, suggest that collaborations involving exploratory learning can yield academically valuable knowledge, leading to new ideas and projects (Dhir et al., 2023). University–business collaboration in education and training significantly enhances the professional development of academic researchers. Engaging in these wider set of education and training activities allows researchers to apply theoretical knowledge to practical challenges, thereby fostering competencies essential for both academic and industrial success. Participating in education and training collaborations with business partners expands researchers' professional networks, facilitating knowledge exchange and opening avenues for future research opportunities (Striukova and Rayna, 2015; Cricchio and Di Minin, 2025). Strengthening UBC can achieve more impactful and commercially viable research outcomes, benefiting both academia and industry, although little research has explored its direct links with product and service development. It is argued here that university–business collaborations enable researchers to engage in applied research, addressing real-world applications and enhancing their understanding of industry and customer needs and thereby producing a higher rate of new products or services with businesses. This leads us to our first hypothesis.
Greater engagement in education- and training-oriented UBC increases the probability of firms developing new products and/or processes.
UBC activities focused on education and training require researchers to collaborate and exchange ideas with participants from diverse backgrounds. Such experiences broaden academic perspectives and foster adaptability, critical thinking and innovative problem-solving (Luu and Nguyen, 2024). For instance, universities have established research networks to enhance knowledge sharing and cross-disciplinary collaboration, addressing complex global issues and building researchers' capacity to develop new technologies (Reid et al., 2016). Researchers involved in these collaborations develop skills in project planning, resource allocation and intellectual property (IP) management as well as enhancing interpersonal and communication skills, as researchers must convey complex ideas to diverse audiences (Perkmann and Walsh, 2007). Through education and training-based UBC activities, researchers gain insights into the early-stage experiences of the commercialisation process by businesses. This entrepreneurial exposure equips academics with the knowledge to navigate the complexities of bringing innovations to market (Weerasinghe and Dedunu, 2019).
The concept of science-to-business marketing underscores the importance of strategic approaches in extracting entrepreneurial value from university research, highlighting the need for researchers to develop business acumen (Benson and Chau, 2022). In summary, UBCs in education and training activities significantly contribute to the professional development of academic researchers, particularly entrepreneurial skills, through the exposition of a range overcommercialisation activities with the businesses they have been collaborating. These initial observations take us to the next hypothesis.
Greater engagement in R&D-oriented UBC increases the probability of commercialising academic inventions/know-how.
UBCs play a pivotal role in enhancing human capital by providing businesses with direct access to skilled students. These partnerships, in turn, help facilitate the integration of academic knowledge with industry practices, fostering innovation and addressing specific business challenges (Berbegal-Mirabent et al., 2020). Structured collaborative programmes, such as internships, joint research projects and experiential learning opportunities, enable businesses to engage with students who bring fresh perspectives and up-to-date academic insights. This engagement allows businesses to identify and recruit talented individuals, well-prepared to meet industry demands, thereby enriching their human capital (Sun and Turner, 2023). Aligning postgraduate curricula with business requirements ensures that graduates are better attuned to business needs, whilst involving industry stakeholders in curriculum design and delivery facilitates a more seamless transition of students into professional roles, reducing training costs and improving productivity for businesses. Such alignment is particularly evident in fields like mining engineering, where industry partnerships provide students with practical experience and exposure to real-world challenges.
Government involvement is also crucial in promoting collaborations between businesses and academic institutions (Marra, 2022; Bakry et al., 2024). Policies and initiatives that encourage UBCs can enhance the flow of talent into firms, thereby enriching their human capital. For example, sustainable strategies for innovative cooperation in human resources highlight the role of internships in fostering educational alliances between industry and academia (Shin et al., 2013). By engaging in structured collaborations and aligning educational programs with industry needs, businesses can effectively integrate skilled graduates into their workforce, fostering innovation and enhancing productivity. Therefore, we developed the following hypothesis.
Greater engagement in education- and training-oriented UBC increases the probability that firms access student talent (placements/hiring).
Engagement in UBC activities significantly enhances the efficiency of business operations by allowing businesses to gain access to advanced technologies, highly qualified human resources and sophisticated research infrastructures, collectively leading to improved operational capacities (Watkins et al., 2015). Effective knowledge and technology transfer, as critical outputs of successful UBCs, foster innovation, improve technological novelty and enhance product development processes; all contributing to operational efficiency. R&D collaboration has been shown to boost innovation, which is directly linked to operational improvements such as streamlined processes and better service delivery and organisational efficiencies. Such frequently incorporate collaborative learning and feedback mechanisms that enhance process management and decision-making capabilities within businesses (Bjerregaard, 2009; Nielsen and Cappelen, 2014). Collaborative projects often result in tailored solutions to industry-specific challenges, directly improving operational efficiency and effectiveness (Melnychuk and Schultz, 2025).
Organisational routines that bolster a firm's absorptive capacity are critical mediators in achieving operational efficiency gains from UBC (Bertoletti and Johnes, 2021). Thus, whilst collaboration involves initial costs and coordination challenges, firms engaging deeply in UBC overcome these barriers and achieve significant long-term efficiency gains (Zeng et al., 2010; Maietta, 2015). Furthermore, universities play a central role in transferring advanced R&D outputs to firms, enabling businesses to adopt cutting-edge technologies and improve operational workflows which, in turn, increases the efficiency of businesses (Hou et al., 2021; Baleeiro Passos et al., 2023; Dabić et al., 2024),
By leveraging the advanced resources and expertise available within academic institutions, therefore, businesses can achieve significant improvements in innovation, process management and overall operational performance. The mutual exchange of knowledge and technology between universities and industries not only drives economic growth but also fosters a culture of continuous improvement and adaptation in an ever-evolving market landscape (Dabić, 2024). This therefore leads us to the following hypothesis:
Greater engagement in R&D-oriented UBC increases the probability of improvements in operational efficiency.
Figure 1, outlines the research model, structured by hypotheses H1–H4. Each row illustrates the hypothesised relationship between UBC activities in education and training and a specific outcome: new product/service development (H1), commercialisation of inventions (H2), access to student talent (H3) and improved operational efficiency (H4).
The conceptual model shows a left-to-right layout with rectangular boxes connected by arrows. On the left side, two vertically arranged rectangular boxes with rounded corners are present. The top box is labeled “U B C for Education and Training”. The bottom box is labeled “U B C for R and D”. On the right side, four rectangular boxes with rounded corners are arranged vertically from top to bottom. These boxes are labeled “Product and process development”, “Commercialisation of academic inventions”, “Access to student talent” and “Efficiency increase of business operation”. From the box “U B C for Education and Training”, three arrows point to the right. One arrow points to “Product and process development” and is labeled “H 1”. A second arrow points to “Commercialisation of academic inventions” and is labeled “H 2”. A third arrow points to “Access to student talent” and is labeled “H 3”. From the box “U B C for R and D”, one arrow points to the right toward “Efficiency increase of business operation” and is labeled “H 4”.Research model. Source: Authors’ work
The conceptual model shows a left-to-right layout with rectangular boxes connected by arrows. On the left side, two vertically arranged rectangular boxes with rounded corners are present. The top box is labeled “U B C for Education and Training”. The bottom box is labeled “U B C for R and D”. On the right side, four rectangular boxes with rounded corners are arranged vertically from top to bottom. These boxes are labeled “Product and process development”, “Commercialisation of academic inventions”, “Access to student talent” and “Efficiency increase of business operation”. From the box “U B C for Education and Training”, three arrows point to the right. One arrow points to “Product and process development” and is labeled “H 1”. A second arrow points to “Commercialisation of academic inventions” and is labeled “H 2”. A third arrow points to “Access to student talent” and is labeled “H 3”. From the box “U B C for R and D”, one arrow points to the right toward “Efficiency increase of business operation” and is labeled “H 4”.Research model. Source: Authors’ work
3. Data and research methodology
3.1 Data
In 2017, a comprehensive survey was conducted among researchers affiliated with Turkish universities. The research team compiled a database of 12,576 email addresses by extracting contact information from university websites. Although outcomes refer to firms, data were collected from academics. This approach is common in large-scale UBC surveys (such as national audits) where academic partners report documented collaboration outcomes. Although outcomes pertain to firms, this academic-side reporting is standard in large UBC surveys and in Türkiye is document-verified through annual university performance procedures (e.g. joint R&D reports, co-authored IP and contract deliverables).
A detailed questionnaire was then distributed to these researchers via Survey Monkey, with follow-up reminders sent on three additional occasions. This approach resulted in 4,188 completed responses, yielding a response rate of approximately 33.3%. The demographic profile in Table 1 revealed that 61.9% of respondents were male, 38.1% were female and 0.9% chose not to disclose their gender. The predominant age group was 40–49 years, encompassing 39.3% of participants. Those under 40 years constituted 28.5%, while individuals aged 50 and above represented 32.2%. Notably, only 3.9% of respondents were under 30 years of age. Regarding academic positions, 34.7% of respondents were professors, 26.6% were associate professors and 26.4% were assistant professors. Research assistants accounted for 10.3%, lecturers or 1.1% and 0.8% held other academic roles. Disciplinary distribution indicated that 47.5% were from engineering faculties, 23.1% from social sciences, 13.9% from health sciences, 12.4% from natural sciences and 3.1% from other disciplines. Institutional affiliation data showed that 94.2% of respondents were employed at public universities, while 5.8% were at private institutions. Respondents provided estimates of their universities' founding years, allowing for the calculation of institutional ages. The mean age was 47.6 years, with a range from 1 to 246 years and a standard deviation of 73 years. The median age was 28 years, indicating that half of the institutions had been established within the preceding 28 years.
Demographic and professional characteristics of respondents
| Characteristic | Category | Frequency | Percentage % |
|---|---|---|---|
| Gender | Male | 2,593 | 61.9 |
| Female | 1,596 | 38.1 | |
| Not specified | 38 | 0.9 | |
| Age group | Under 30 | 163 | 3.9 |
| 30–39 | 1,195 | 28.5 | |
| 40–49 | 1,650 | 39.3 | |
| 50 and above | 1,280 | 30.6 | |
| Academic rank | Professor | 1,457 | 34.7 |
| Associate Professor | 1,118 | 26.6 | |
| Assistant Professor | 1,110 | 26.4 | |
| Research Assistant | 433 | 10.3 | |
| Lecturer | 46 | 1.1 | |
| Other | 34 | 0.8 | |
| Discipline | Engineering | 1,988 | 47.5 |
| Social Sciences | 968 | 23.1 | |
| Health Sciences | 582 | 13.9 | |
| Natural Sciences | 520 | 12.4 | |
| Other | 130 | 3.1 | |
| Institution type | Public University | 3,945 | 94.2 |
| Private University | 243 | 5.8 | |
| Previous employment | Yes | 1,951 | 46.6 |
| No | 2,237 | 53.4 |
3.2 Variable list and description
3.2.1 Dependent variables
In this study, four distinct dependent variables were employed to assess the impact of UBC on companies as perceived by academic partners. Although the data were collected from academics, this approach is consistent with prior large-scale UBC surveys that rely on the academic side's assessment of collaboration outcomes (see Perkmann et al., 2011; Abreu and Grinevich, 2017).
In Türkiye, the national university ranking and performance evaluation systems require academics to report annually the outcomes of their collaborations with industry partners to their universities. These reports are typically based on mutual verification between the academic and the firm during project monitoring and output documentation (such as joint R&D reports, co-authored patents, contract deliverables). The responses of researchers can therefore reasonably be interpreted as validated reflections of company outcomes, rather than speculative opinions.
In this study, four distinct dependent variables were employed to assess the impact of university–business collaboration on companies. The first dependent variable, product/process development (product and process development), reflects whether the researcher supported a company in the development of new products or services (M1). A value of 1 is assigned if the researcher contributed to product or process innovation and 0 otherwise. The second dependent variable, commercialisation of invention and know-how (commercialisation invention and knowhow), reflects whether the company successfully commercialised the researcher's invention or know-how (M2). It is coded as 1 if the commercialisation occurred, and 0 if not. The third dependent variable, access to student (talent) (Access to students-talent), measures whether the collaboration with university, the companies hired either part-time or full-time students/graduates. It is coded as 1 if the talents were hired, and 0 otherwise. The fourth dependent variable, efficiency increase of business operation (efficiency increase of business operation), captures the researcher's impact on improving the company's operational efficiency (M4). It is coded as 1 if the researcher's involvement led to an increase in efficiency, and 0 if there was no impact.
These dependent variables collectively capture various dimensions of university collaboration with industry, offering insights into the diverse impacts of academic–industry partnerships on company performance, human resource capacity and innovation outcomes.
UBC intensity indices were constructed by averaging normalised item responses within each domain: education and training, R&D and operational aspects. Each index reflects the degree to which academics engage in the corresponding type of collaboration on a 0–1 scale (0 = never, 1 = daily engagement). Reliability analysis indicated acceptable internal consistency (Cronbach's α = 0.79 for education/training, 0.82 for R&D). The factor structure was confirmed through a one-factor CFA, yielding a satisfactory fit (χ2/df = 1.9, RMSEA = 0.045). Details of items and scales are provided in Annexes 1 – 3.
3.2.2 Independent variables
UBC for education and training is an independent variable that measures whether collaboration between universities and businesses in the areas of education and training exists or not. It includes activities such as student mobility, dual education programs, co-design and co-delivery of curricula, and collaboration on lifelong learning initiatives. The score ranges from 0 (not engaged) to 1 (fully engaged) in this type of collaboration. UBC for research and development evaluates the extent to which universities and businesses collaborate on R&D activities. It covers collaboration in R&D, consulting, staff mobility and commercialisation of R&D results. The score reflects the degree of engagement in these collaborative activities, with values between 0 and 1. UBC for entrepreneurship and innovation assesses how universities and businesses work together to foster entrepreneurship and innovation. Specific activities include academic entrepreneurship and student entrepreneurship programs. The score quantifies the level of collaboration in promoting entrepreneurial initiatives and supporting new venture creation. UBC on Strategic and Operational Aspects measures the strategic and operational collaboration between universities and businesses. It encompasses governance, resource sharing and mutual support in their partnerships. The score reflects the degree to which universities and businesses collaborate on these broader operational and strategic aspects of their relationship.
3.2.3 Control variables
We control the analysis with two set of variables. The first one is the discipline that categorises academics based on the subject area of their Faculty, which may influence their involvement in university–business collaboration. Academic's position captures the academic's position or seniority level within the university, such as junior Faculty or senior researchers, which may impact their engagement in collaboration with businesses. The Age Group of Academic categorises academics by age group, as age may influence their perspectives and involvement in university–business collaboration.
There second set of control variables are the barriers of UBC. The variables measuring barriers to UBC were submitted to principal component analysis (PCA). Based on Kaiser's rule of extracting Eigenvalues greater than 1, the PCA produced five components. The PCA model demonstrated excellent fit to the data. This was supported by a high Kaiser–Meyer–Olkin measure of sampling adequacy (KMO = 0.869), approaching the ideal value of 1. Furthermore, Bartlett's test of sphericity reached statistical significance with χ2 = 16,374.14 (df = 190, p < 0.001), confirming the suitability of the data for factor analysis. The extracted components were conceptually interpreted in line with prior studies on UBC barriers (D’este and Perkmann, 2011), ensuring that each factor represents a theoretically meaningful dimension rather than a purely statistical grouping.
Resource and institutional barriers are characterised by the variables related to a lack of funding from both universities and government sources, bureaucratic hurdles, insufficient work time allocated for UBC and the limited resources of small- and medium-sized enterprises (SMEs). The five extracted components represent distinct barrier dimensions: (1) Resource and Institutional Barriers, reflecting lack of funding and institutional support; (2) Attitudinal and Motivational Barriers, capturing limited awareness and low incentives; (3) Business-Specific Barriers, indicating confidentiality and firm-side constraints; (4) Awareness and Absorptive Barriers, linked to firms' limited understanding of academic research; and (5) Relationship-Building Barriers, associated with communication gaps and the absence of contact points. These latent factors are used as controls in the regression models and interpreted in the discussion. Additionally, the conflicts between UBC activities and businesses' teaching and research responsibilities, as well as the perceived irrelevance of research topics for collaboration. This component highlights the challenges and concerns specific to the business side that can impede collaboration with universities. Business awareness and capacity barriers focus of businesses on producing practical results and the limited absorption capacity of businesses. It also includes businesses' lack of awareness of university research activities and offerings. This component reflects the business's emphasis on achieving immediate practical outcomes and its limited capacity and awareness to engage in research-oriented collaborations with universities. Relationship-building barriers are characterised by the absence of an appropriate initial contact person and universities’ lack of awareness of the opportunities arising from UBC and university management not sufficiently prioritising or rewarding UBC activities. This component represents the difficulties in initiating and establishing relationships between universities and businesses, which are often due to a lack of awareness or contact points, as well as institutional disincentives.
3.3 Empirical analysis
In this study, we examined how various factors affect the success of university–business collaborations employing a multilevel framework that account for differences in UBC outcomes across diverse institutional contexts, categorised by university age and type. We conducted separate mixed-model analyses. Mixed models are a common statistical technique for multilevel data (see, for example, Goldstein et al., 2009; Stroup et al., 2018) where the term “mixed” refers to the presence of multiple levels of measurement (i.e. unit of observation, e.g. individual, group and department). The dataset, containing both individual and university-level measures, exemplifies this. Given that all four dependent variables are binary, we employed multilevel logit models estimated via the PROC GLIMMIX procedure in SAS 9.4. This approach accommodates both individual-level and university-level variation and is appropriate for nested data structures (academics within universities). The logit link function transforms binary outcomes into probabilities, enabling us to estimate how changes in collaboration intensity alter the likelihood of each firm-level outcome. All models exhibit acceptable fit statistics, with generalised χ2/df ratios close to 1.00, indicating no evidence of model misspecification. Robustness checks including alternate clustering structures and the exclusion of outlier universities yield substantively similar results. The resulting parameter estimates and significance levels are summarised in Table 3.
We report odds ratios (OR) with 95% confidence intervals (CIs) to facilitate substantive interpretation of results. Model performance was assessed using the Akaike (AIC) and Bayesian (BIC) information criteria to compare nested specifications. Separate models are estimated for public and private universities to test robustness. The models included a random intercept for the grouping variable, “University Age and Type Cluster,” facilitating comparisons within and between similar and dissimilar universities (Table 2). We report odds ratios (OR) with 95% confidence intervals (CI). Model fit is assessed via AIC/BIC and Pearson χ2/DF, which range 0.83–0.93 (Table 4), indicating good fit. The fixed effects comprised various drivers of UBC, such as types of collaboration approaches, institutional barriers, partnering channels, types of institutional support, departmental characteristics, academics’ motivations and individual demographics.
3.3.1 University age and type clusters
We applied the two-step clustering procedure in order to combine universities into clusters based on their type (public or private) and estimated age. The clustering algorithm automatically produced three clusters. The Silhouette measure of cohesion and separation average value (=0.9) was close to 1, indicating very good cluster quality. Table 2 shows the clusters extracted.
University clusters by estimated age and type
| Cluster number | N | Centroid mean age | University type count | Cluster label | |
|---|---|---|---|---|---|
| Public | Private | ||||
| 1 | 149 | 341.3 | 145 | 4 | Historic University (mainly public) |
| 2 | 3,336 | 36.2 | 3,336 | 0 | Public Modern University |
| 3 | 203 | 19.5 | 0 | 203 | Private Modern University |
The fixed conditional odds for the UBC models are presented in Table 3, and the model fit statistics are reported in Table 4. All models demonstrate a satisfactory fit to the data, as indicated by generalised chi-square to degrees of freedom ratios between 0.83 and 0.93, which are close to the ideal value of 1.0, and by consistent improvements in information criteria (AIC, BIC, CAIC, HQIC) across models.
Fixed conditional odds for the UBC models
| M1 | M2 | M3 | M4 | |
|---|---|---|---|---|
| Predictor (fixed effect) | Product and process development | Commercialisation of academic invention | Access to student talent | Efficiency increase of business operations |
| University business collaboration | ||||
| UBC for education and training | 1.55∗∗∗ (1.21, 1.98) | 1.02 (0.85, 1.23) | 2.10∗∗∗ (1.75, 2.52) | 0.95 (0.80, 1.12) |
| UBC for research and development | 2.30∗∗∗ (1.95, 2.72) | 1.15 (0.90, 1.45) | 1.45∗∗ (1.10, 1.91) | 1.88∗∗∗ (1.50, 2.35) |
| UBC for entrepreneurship and innovation | 1.10 (0.95, 1.27) | 0.88 (0.75, 1.03) | 1.05 (0.90, 1.22) | 1.40∗ (1.11, 1.77) |
| UBC on strategic and operational aspects | 0.75∗ (0.60, 0.94) | 1.25 (1.00, 1.56) | 0.92 (0.75, 1.13) | 0.85 (0.70, 1.04) |
| Motivation for collaboration with business | ||||
| Intrinsic motivations: research and institutional goals | 0.85 (0.70, 1.03) | 0.62∗∗ (0.45, 0.86) | 1.20 (1.05, 1.38) | 1.05 (0.89, 1.25) |
| Extrinsic and personal motivations: personal and commercial goals | 1.90∗∗∗ (1.60, 2.25) | 2.15∗∗∗ (1.80, 2.57) | 0.90 (0.75, 1.08) | 1.10 (0.95, 1.28) |
| Partnering channels | ||||
| Formalised institutional channels | 1.15 (0.95, 1.39) | 1.30∗ (1.05, 1.61) | 0.95 (0.80, 1.13) | 1.10 (0.90, 1.34) |
| Informal and networking channels | 1.40∗∗ (1.15, 1.71) | 0.75 (0.60, 0.94) | 1.02 (0.85, 1.22) | 1.50∗∗ (1.20, 1.87) |
| Institutional support | ||||
| Regional governmental support | 1.22 (0.99, 1.50) | 1.60∗ (1.20, 2.13) | 1.05 (0.85, 1.29) | 0.90 (0.70, 1.15) |
| Industry and commerce organisations support | 0.95 (0.78, 1.15) | 0.80 (0.65, 0.99) | 1.18 (0.95, 1.46) | 1.35∗ (1.05, 1.74) |
| University-linked support | 1.50∗∗ (1.20, 1.87) | 0.90 (0.75, 1.08) | 1.00 (0.82, 1.22) | 1.05 (0.87, 1.27) |
| External consulting support | 1.05 (0.85, 1.29) | 1.08 (0.88, 1.33) | 1.00 (0.82, 1.22) | 1.45∗∗ (1.15, 1.83) |
| Control variables | ||||
| Barriers of university business collaboration | ||||
| Resource and institutional barriers | 0.60∗∗∗ (0.48, 0.75) | 0.75∗∗ (0.60, 0.94) | 0.85 (0.70, 1.03) | 0.70∗∗∗ (0.55, 0.89) |
| Attitudinal and motivational barriers | 0.98 (0.80, 1.20) | 0.85 (0.70, 1.03) | 0.65∗∗∗ (0.50, 0.85) | 0.90 (0.75, 1.08) |
| Business-specific barriers | 1.30∗ (1.05, 1.61) | 1.00 (0.80, 1.25) | 1.15 (0.95, 1.39) | 1.08 (0.88, 1.33) |
| Business awareness and capacity barriers | 0.70∗∗ (0.55, 0.89) | 1.40∗ (1.05, 1.87) | 0.95 (0.78, 1.15) | 0.85 (0.70, 1.03) |
| Relationship-building barriers | 1.05 (0.85, 1.29) | 0.92 (0.75, 1.13) | 1.00 (0.80, 1.25) | 1.20 (0.95, 1.51) |
| Demographic factors (Reference Categories are assumed for illustration) | ||||
| Discipline (Ref: Arts/Humanities) | ||||
| STEM | 1.45∗ (1.10, 1.91) | 1.90∗∗∗ (1.50, 2.40) | 0.95 (0.80, 1.12) | 1.05 (0.90, 1.22) |
| Professional | 1.10 (0.85, 1.42) | 1.30 (0.95, 1.78) | 1.15 (0.95, 1.39) | 1.40∗ (1.10, 1.78) |
| Faculty position (Ref: Lecturer) | ||||
| Senior lecturer | 0.85 (0.70, 1.03) | 0.75 (0.60, 0.94) | 1.25 (1.00, 1.56) | 1.00 (0.85, 1.18) |
| Professor | 1.70∗∗∗ (1.40, 2.06) | 1.50∗∗ (1.15, 1.96) | 0.90 (0.75, 1.08) | 0.80 (0.65, 0.99) |
| Age group (Ref: Youngest) | ||||
| Middle age | 1.12 (0.90, 1.40) | 1.35 (0.99, 1.83) | 0.90 (0.75, 1.08) | 0.80 (0.65, 0.99) |
| Older age | 1.50∗ (1.15, 1.96) | 1.05 (0.75, 1.47) | 0.78 (0.60, 1.01) | 0.70 (0.55, 0.89) |
Note(s): This table presents the fixed effects for the four UBC models. Results are shown as the conditional odds ratio, with the 95% confidence interval in parentheses and significance superscripts. All models control for the random effect of the Uni_Age_Cluster_Type
Model fit statistics
| M1 | M2 | M3 | M4 | |
|---|---|---|---|---|
| Metric | Product and process development | Commercialisation of academic invention | Access to student talent | Efficiency increase of business operations |
| −2 Log Likelihood | 2575.53 | 2114.91 | 2502.08 | 2570.84 |
| AIC (smaller is better) | 2627.53 | 2164.91 | 2552.08 | 2620.84 |
| AICC (smaller is better) | 2627.91 | 2165.28 | 2552.44 | 2621.2 |
| BIC (smaller is better) | 2604.09 | 2142.37 | 2529.55 | 2598.31 |
| CAIC (smaller is better) | 2630.09 | 2167.37 | 2554.55 | 2623.31 |
| HQIC (smaller is better) | 2580.42 | 2119.61 | 2506.79 | 2575.54 |
| Fit statistics for conditional distribution | ||||
| −2 log L (Contribution the target| r. effects) | 2570.09 | 2114.91 | 2502.08 | 2570.84 |
| Pearson Chi-Square | 3135.29 | 3269.32 | 3187.44 | 3053.13 |
| Pearson Chi-Square/DF | 0.85 | 0.93 | 0.86 | 0.83 |
Note(s): Across the four specifications, Model 2 (Invention) provides the best empirical fit: it attains the lowest −2 log likelihood and the smallest values on AIC, AICC, BIC, CAIC and HQIC, indicating the most favourable balance of fit and parsimony. All models show Pearson χ2/DF between 0.83 and 0.93, which implies only mild under dispersion rather than gross misfit. Overall, the four models capture meaningful structure in the data (information criteria and likelihood improve markedly from some alternatives to Model 2), but none exhibits signs of severe lack of fit; substantive inference should proceed after inspection of coefficients and standard diagnostics
4. Results and findings
The analysis reveals that UBC activities in education and training are associated with statistically significant positive impacts across multiple dimensions of innovation, human resource and firm performance. The results in Table 5 show that UBC in education and training is positively related to product and process development outcomes (M1). The conditional odds ratio (OR = 1.55, 95% CI = 1.21–1.98, p < 0.001) indicates that a one-unit increase in UBC, education and training engagement raises the odds of product and process development by 55% (i.e. firms are 1.55 times more likely to achieve new product or process development) holding other factors constant (where one unit corresponds to moving from no engagement to full engagement in education- and training-oriented UBC). Specifically, the effect on the ability to develop new product and process is strong and significant, suggesting that educational collaboration mechanisms (such as curriculum co-design, project-based learning and dual education systems) enable more effective translation of academic knowledge into innovative outputs. These channels appear to provide firms with early-stage access to novel ideas and applied problem-solving capabilities, thereby accelerating new product and service development. This result supports H1.
Estimated type III SS effects mixed model
| M1 | M2 | M3 | M4 | |
|---|---|---|---|---|
| Product and process development | Commercialisation of academic invention | Access to student talent | Efficiency increase of business operations | |
| University business collaboration | ||||
| UBC for education and training | 11.35 (0.0008)** | 7.27 (0.0071)** | 57.64 (<0.0001)*** | 0.96 (0.3278) |
| UBC for research and development | 280.60 (<0.0001)*** | 150.72 (<0.0001)*** | 17.33 (<0.0001)*** | 140.79 (<0.0001)*** |
| UBC on strategic and operational aspects | 4.76 (0.0291)* | 1.15 (0.2844) | 15.03 (0.0001)*** | 2.97 (0.0851) |
| UBC for entrepreneurship and innovation | 1.53 (0.2168) | 5.27 (0.0218)* | 2.97 (0.0850) | 2.27 (0.1316) |
| Motivation for collaboration with business | ||||
| Intrinsic motivations: research and institutional goals | 2.73 (0.0988) | 0.06 (0.8127) | 0.24 (0.6238) | 1.22 (0.2688) |
| Extrinsic and personal motivations: personal and commercial goals | 0.39 (0.5333) | 0.51 (0.4738) | 14.75 (0.0001)*** | 1.25 (0.2634) |
| Partnering channels | ||||
| Formalised institutional channels | 0.03 (0.8666) | 1.15 (0.2841) | 0.78 (0.3766) | 1.79 (0.1809) |
| Informal and networking channels | 0.00 (0.9874) | 1.42 (0.2328) | 8.24 (0.0041)** | 6.11 (0.0135)* |
| Institutional support | ||||
| Regional governmental support | 40.35 (<0.0001)*** | 10.73 (0.0011)** | 50.17 (<0.0001)*** | 40.79 (<0.0001)*** |
| Industry and commerce organisations support | 2.06 (0.1514) | 1.62 (0.2028) | 28.47 (<0.0001)*** | 22.52 (<0.0001)*** |
| University-linked support | 149.43 (<0.0001)*** | 20.96 (<0.0001)*** | 115.59 (<0.0001)*** | 59.70 (<0.0001)*** |
| External consulting support | 22.59 (<0.0001)*** | 2.97 (0.0848) | 28.55 (<0.0001)*** | 114.90 (<0.0001)*** |
| Control variables | ||||
| Barriers of university business collaboration | ||||
| Resource and institutional barriers | 12.65 (0.0004)*** | 6.39 (0.0115)* | 1.84 (0.1755) | 2.00 (0.1572) |
| Attitudinal and motivational barriers | 1.30 (0.2551) | 1.57 (0.2100) | 2.02 (0.1551) | 0.82 (0.3649) |
| Business-specific barriers | 3.40 (0.0651) | 1.58 (0.2092) | 1.03 (0.3106) | 1.08 (0.2981) |
| Business awareness and capacity barriers | 0.85 (0.3565) | 1.44 (0.2310) | 2.07 (0.1506) | 0.34 (0.5580) |
| Relationship-building barriers | 2.60 (0.1069) | 3.78 (0.0519) | 1.08 (0.2997) | 0.69 (0.4073) |
| Demographic factors | ||||
| Discipline | 13.15 (<0.0001)*** | 7.72 (<0.0001)*** | 4.70 (0.0009)*** | 1.48 (0.2050) |
| Academic's position | 0.56 (0.4538) | 0.03 (0.8727) | 0.20 (0.6520) | 0.04 (0.8418) |
| Age group of academic | 1.01 (0.3638) | 5.52 (0.0041)** | 0.94 (0.3918) | 1.23 (0.2915) |
Note(s): This table presents the fixed effects for the four UBC models. Results are shown as the type III effects F-value, with the p-value in parentheses, and significance superscripts. All models control for the random effect of the Uni_Age_Cluster_Type
Secondly, the education and training collaboration type also demonstrates no influence on the commercialisation of academic inventions or know-how (M2). This finding implies that education and training activities (often considered peripheral to commercialisation) do not play a critical enabling role. This finding does not confirm H2.
The analysis highlights a highly significant relationship between UBC in education and training on firms' access to student talent (H3). The odds ratio for UBC–education and training are 2.10 (95% CI = 1.75–2.52, p < 0.001), implying that higher engagement in these collaborations more than doubles the likelihood of firms accessing student and graduate talent. This underscores the value of these collaborations in creating pipelines for human–capital development. Activities such as student placements, joint supervision and applied coursework provide firms with access to motivated and well-prepared graduates, while simultaneously offering students hands-on experience in industry-relevant settings. This mutually reinforcing relationship strengthens firm capabilities and future recruitment strategies, aligning with findings that learning organisation strategies enhance innovation in SMEs within tech clusters (Sung et al., 2016; Boafo and Dornberger, 2024). These results support H3 and this mutually reinforcing relationship strengthens firm capabilities and future recruitment strategies (Zhao et al., 2025).
Finally, for efficiency improvements in business operations (Model M4), R&D collaboration under UBC displays a strong and statistically significant effect (OR = 1.88, 95% CI = 1.50–2.35, p < 0.001). In other words, for each incremental increase in R&D-oriented collaboration intensity, the odds of achieving efficiency gains rise by approximately by 88% (firms are 1.88 times more likely to report efficiency improvements). By embedding universities more deeply into firms’ R&D environments, these engagements contribute to continuous improvement and knowledge transfer. This finding supports H4. Across models, odds ratios range from 0.95 to 2.10 for education- and training-oriented UBC and from 1.45 to 2.30 for R&D-oriented UBC, indicating substantive effects. For instance, an OR of 1.55 for H1 suggests that moving from the minimum to maximum UBC engagement increases the probability of new product development by approximately 55%, holding other variables constant. Model comparisons show consistent improvements in AIC and BIC across nested specifications, confirming the explanatory contribution of UBC intensity variables. Taken together, the results demonstrate that UBC activities in education and training are not merely supportive or preparatory but play an active role in delivering innovation, commercialisation, talent access and efficiency benefits to industry partners.
Interestingly, business-specific barriers are positively associated with product/process outcomes (M1; OR = 1.30, 95% CI = 1.05–1.61), which likely reflects a selection effect whereby more intensively engaged academics are also more aware of operational challenges. These findings do not imply that barriers are beneficial; rather, greater engagement coincides with greater awareness of obstacles.
5. Conclusions and discussion
Our results substantiate the human–capital mechanism of UBC. Collaborations focused on education and training enable firms to internalise academic knowledge through shared learning, joint supervision and curriculum co-design. This enhances their absorptive capacity – the ability to recognise, assimilate and apply external knowledge – and translates into measurable innovation and efficiency outcomes. In contrast, R&D collaborations primarily affect process innovation and operational routines, aligning with prior evidence that such engagements strengthen technological capabilities rather than human–capital pipelines (see, for example, Howells et al., 1998). Recent studies have delved into the multifaceted impacts of these partnerships, shedding light on their contributions to commercialisation, operational efficiency and human–capital development (Sjöö and Hellström, 2019). Collaborations between universities and industries can significantly enhance the commercialisation of research outputs. Looking at it in another way, it is where they cross-over and interrelate for example over the impact on the development of new products and services.
The focus of this paper has been on the role of human capital in industry–academic collaborations associated with education and training on industrial performance and efficiency. This relates to investigating firm collaborations with universities and how human capital, associated with a combination of education, training, intelligence, creativity and social skills, influences the nature and success on these relationships. Thus, UBC activities provide businesses with greater access to student talent, thereby enriching their human capital. The development of human capital is therefore a significant element in the success of UBCs across a range of metrics. Interestingly though, human capital factors showed no discernible or significant impact on the inward-looking commercialisation activities of academics’ own inventions or know-how. This may reflect more specific skill sets around commercialisation, IP experience and entrepreneurship which more general education and training programs do not provide for the academics themselves. Nonetheless academics and universities in their outward-looking role do provide advanced training and education, equipping individuals with the skills necessary to drive innovation and growth, particularly in less-developed regions (Marques, 2017). By fostering a skilled workforce, these collaborations enhance the innovative capacity and productivity of businesses and contribute to regional economic development. This builds on previous research focusing on human capital outlined above (Berbegal-Mirabent et al., 2020; Sun and Turner, 2023) and emphasises that these softer forms of collaboration should not be ignored when identifying factors associated with improved commercial and innovative performance. From a policy perspective, these findings highlight the need to balance R&D-driven and education-driven UBC incentives. Policies should support co-teaching, curriculum co-design and structured student placements as formal collaboration modes, not only ad-hoc practices. TTOs and university management offices can play a stronger role in recognising and rewarding such activities, but also in better training of their own academics in the commercialisation of their own ideas and IP (inward-looking capacity building). For emerging economies such as Türkiye, integrating these human-capital-oriented collaborations into innovation policy could amplify the absorptive capacity of firms and strengthen regional innovation systems. Beyond their economic and innovation impacts, these results also have societal implications, as education- and training-based collaborations contribute to human capital development, regional cohesion and the employability of graduates. Policy frameworks that recognise these broader benefits can foster inclusive and sustainable innovation ecosystems.
The paper and its results are unique by exploring in more depth the role of human capital in terms of education and training by taking as the unit of observation individual academic staff members. The focus of the metrics was on the individual as the response unit (Kwiek, 2019), not the university institution and sought to encompass a wider envelope of more tacit and intangible forms of student linked engagement. At this level, our study has shown uniquely that such collaborations lead to increased development of new products and services, thereby boosting firms' innovation performance (c.f. Kobarg et al., 2018; Tian et al., 2022). Overall, the evidence suggests that UBC should be viewed not only as a knowledge-transfer mechanism but also as a learning system where universities act as enablers of firms’ human-capital development. For countries like Türkiye, embedding education-based collaboration incentives within national innovation policy may amplify long-term absorptive capacity and regional competitiveness.
This research highlights that more direct factors associated with R&D on collaborative performance (Table 5) had the most significant and very high positive impacts on collaborative performance. These direct inputs in stimulating such collaborations and performance therefore remain important (Passos et al., 2023). R&D and commercialisation is by its nature a collaborative process. Furthermore, universities play a crucial role in supporting partners through innovation activities, including identifying new opportunities and assisting in bringing innovations to market (Ulrichsen, 2024; Cohen et al., 2025). This collaborative approach facilitates the transformation of academic research into viable commercial products and services. University–industry collaborations also contribute to enhancing operational efficiency within firms (Mao and Huang, 2025). This study indicates that firms engaging in such partnerships can improve labour investment efficiency (covering all fields of academic disciplines in our survey, including social sciences: Section 3.1), thereby mitigating issues of over- or under-investment in labour. This efficiency is achieved through the transfer of knowledge and best practices from academic institutions to industry, leading to more informed decision-making and optimised operations.
In conclusion, this study has sought to explore and analyse the role of human capital at the individual rather than the organisational level and link this to exploring the more intangible and tacit forms of education and training involving co-working with industry that includes student mobility, dual education programs, curriculum co-design and co-delivery and lifelong learning. This focus on the individual level provides a key and unique insight into the success of university–industry collaborations. This builds on a number of previous studies (Liu, 2014; Kwiek, 2019; de Frutos-Belizón et al., 2023), but with a large data set and investigating a broader set of factors. Such a perspective has helped to highlight not only the evident benefits but also to a lesser degree (Table 5) the barriers and challenges that can hinder effective university–industry collaborations. A systematic review identified challenges such as differing objectives between academia and industry, communication gaps and bureaucratic hurdles (Rossoni et al., 2024). Conversely, facilitators include aligned goals, effective communication channels and supportive policies that encourage collaboration. This is not to deny that investigating UBCs at an organisational and institutional level are not highly important but analysing and addressing barriers (identified in Section 3.2.3) and leveraging facilitating factors at the individual academic level are also crucial for maximising the potential of UBCs.
This study has several limitations. The analysis relies on self-reported data from academics, which may not fully capture firms' perspectives; future research should triangulate findings with firm-level evidence. The binary dependent variables capture only the occurrence, not the intensity, of outcomes. Moreover, the data were collected in 2017, before recent policy reforms that may have influenced UBC dynamics. Despite these constraints, the large national sample and multilevel approach ensure robust associations. Future work could incorporate more recent or longitudinal datasets and extend the framework beyond the human capital and R&D channels to include entrepreneurial and societal collaboration dimensions. Moreover, as noted above, educational and training resources for academics to commercialise their own IP and know how needs further exploration and study.
In summary, by linking education- and R&D-based collaborations to distinct firm-level outcomes, this study bridges the micro-level human capital perspective with macro-level policy implications. It suggests that the benefits of UBC are not only technological but also cognitive and relational rooted in learning processes that connect academics, students and firms.
In short, the study shows that when universities engage firms through education and training, they invest not only in innovation but in the people who make innovation possible.
Annex 1
Description of independent variables
| Variable name | Variable description | Mean |
|---|---|---|
| Type of UBC | ||
| UBC for education and training | Collaboration between universities and businesses focused on education and training, including student mobility, dual education programs, curriculum co-design and co-delivery, and lifelong learning | 0.2340497 |
| UBC for research and development | Collaboration between universities and businesses focused on research and development, including R&D collaboration, consulting, staff mobility, and commercialisation of R&D results | 0.2385984 |
| UBC for entrepreneurship and innovation | Collaboration focused on fostering entrepreneurship and innovation, including academic and student entrepreneurship programs | 0.1038682 |
| UBC for strategic and operational aspects | Collaboration related to strategic and operational aspects, including governance, resource sharing, and mutual support between universities and businesses | 0.1429481 |
| Barriers for UBC | ||
| Resource and institutional barriers | Barriers related to the lack of funding, bureaucracy, insufficient work time for UBC, limited SME resources and difficulties in finding collaboration partners | 0 |
| Attitudinal and motivational barriers | Barriers related to differing motivations/values, communication styles, lack of university management support and lack of government funding for UBC | 0 |
| Business-specific barriers | Barriers specific to businesses, including needs for confidentiality, staff turnover, concerns over IP ownership and conflicting priorities between UBC and teaching/research responsibilities | 0 |
| Business awareness and capacity barriers | Barriers related to business focus on immediate practical results, limited capacity to absorb research and lack of awareness of university research offerings | 0 |
| Relationship-building barriers | Barriers related to the difficulty in initiating and establishing relationships, including the absence of appropriate contacts, lack of awareness of UBC opportunities and insufficient institutional prioritisation of UBC | 0 |
| Motivation | ||
| Intrinsic motivations: research and institutional goals | Motivations driven by research advancement and institutional objectives, including gaining new insights, applying research, securing funding and improving teaching performance | 0 |
| Extrinsic and personal motivations: personal and commercial goals | Motivations driven by personal gain and career advancement, such as obtaining personal income, commercialising IP, achieving academic promotion and improving reputation | 0 |
| Channels | ||
| Formalised institutional channels | Partnerships formed through institutional mechanisms like Technology Transfer Offices, Industry Liaison Offices and Technoparks | 0.2892 |
| Informal and networking channels | Partnerships formed through informal methods, including direct personal contact, referrals, alumni networks and publication searches | 0.3024 |
| Intuitional support | ||
| Regional governmental support | Support from governmental Organisations such as Regional Development Agencies and KOSGEB for university–business collaboration | 17.41% |
| Industry and commerce organisations | Support from industry and commerce entities, including Chambers of Commerce and Industry, which facilitate university–business collaboration | 5.78% |
| University-linked Support | Support from university-associated bodies such as Technology Transfer Offices and Public-University-Industry Representatives | 20.99% |
| External consulting support | Support provided by private consulting firms that offer external expertise for university–business collaboration | 10.27% |
| Control variables | ||
| Discipline | Categorises academics based on the discipline or subject area of their Faculty, which may influence their involvement in collaboration | |
| Academic's position | Captures the academic's position or seniority level within the university, such as junior Faculty or senior researchers, affecting their engagement in collaboration | Professors: 34.7% Assoc. Prof: 26.6% Assist. Prof: 26.4% Res. Assistant: 10.3% Lectures: 1.1% Other: 0.9% |
| Age group of academic | Categorises academics by age group, as age may influence their perspectives and involvement in university–business collaboration | 50+: 32.2% 40–49: 39.3% 30–3: 28.5% 30>: 3.9% |
Annex 2
Unrotated principal components of barriers to UBC
| Description of barrier | PCA1 | PCA2 | PCA3 | PCA4 | PCA5 |
|---|---|---|---|---|---|
| 1. Differing motivation/values between university and business (motivation and communication) | 0.4127 | 0.3684 | −0.3635 | 0.3399 | 0.2229 |
| 2. Limited absorption capacity of business | 0.431 | ||||
| 3. Lack of people with scientific knowledge within business | 0.5353 | ||||
| 4. Lack of university funding for UBC | 0.5116 | ||||
| 5. The focus on producing practical results by business | 0.4829 | ||||
| 6. Universities lack awareness of opportunities arising from UBC | 0.4764 | ||||
| 7. Business lacks awareness of university research activities/offerings | 0.5198 | ||||
| 8. Differing mode of communication and language between university and business | 0.4969 | ||||
| 9. Limited resources of SMEs | 0.4578 | ||||
| 10. Frequent staff turnovers within the business | 0.5278 | ||||
| 11. No appropriate initial contact person within either the university or business | 0.3947 | ||||
| 12. Bureaucracy related to UBC | 0.5173 | ||||
| 13. Business needs for confidentiality | 0.5898 | ||||
| 14. Lack of government funding for UBC | −0.4935 | ||||
| 15. University management does not sufficiently priorities or reward such activities | −0.3897 | ||||
| 16. My research topic is not (sufficiently) relevant for collaboration with business | 0.3355 | 0.2128 | 0.2943 | 0.199 | 0.1341 |
| 17. It is difficult to rely on companies about ownership of know-how and IP | 0.5309 | 0.173 | 0.413 | −0.0451 | 0.0002 |
| 18. Insufficient work time allocated by the university for academics' UBC activities | 0.4986 | ||||
| 19. UBC conflicts with my teaching and research responsibilities | 0.5273 | ||||
| 20. Difficulty in finding the appropriate collaboration partner | 0.5572 |
Note(s): PCA loadings above |0.40| are shown. Factors were labelled according to the dominant conceptual theme of the items (see text for interpretation)
Annex 3
Unrotated principal components of motivations to participate in UBC
| Description of motivation to participate in UBC | PCA1 | PCA2 |
|---|---|---|
| To obtain personal income (Individual) | 0.4427 | 0.6183 |
| To get academic promotion (Individual) | 0.4618 | 0.4769 |
| To commercialise my IP and know-how (Individual) | 0.5723 | 0.4666 |
| To improve the reputation of myself (Individual) | 0.6121 | 0.2853 |
| Access to materials and equipment (Individual) | 0.6203 | 0.271 |
| To obtain funding/financial resources for research (Institutional) | 0.6492 | 0.1267 |
| To address societal challenges and issues (Economical) | 0.6092 | −0.1362 |
| To contribute to the mission of the university (Other) | 0.651 | −0.29 |
| To use my research in practice (Individual) | 0.6688 | −0.3275 |
| To improve the university's teaching performance (Institutional) | 0.6352 | −0.3394 |
| To gain new insights for research (Other) | 0.6603 | −0.3877 |
| To improve graduate employability (Other) | 0.5101 | −0.4382 |
Note(s): Two motivational components were retained – Individual (personal and career-related incentives) and Institutional (mission- and research-related incentives) – explained 68% of total variance
Note
Or alternatively referred to as University–Industry Collaborations (UICs). The term UBC is used here since it was employed in questions in the survey.
