Research on the Optimization Path of Organizational Model and Governance Structure Adaptability for Industry-Education Integration in Vocational Colleges ()
1. Introduction
Against the backdrop of deepening industry-education integration in current vocational education, organizational models such as order-based classes, industry colleges, and vocational education groups have become important carriers for linking the education chain with the industrial chain. However, in practice, there are common issues: order-based classes often face unstable enterprise participation due to unclear profit distribution; industry colleges experience inefficient resource coordination due to a lack of decision-making mechanisms; and vocational education groups fall into governance dilemmas of being ‘linked but not unified’ due to loose responsibility structures. These problems directly affect enterprises’ evaluation of the quality of talent produced through integration (Hall et al., 2011). Although existing research has paid attention to the classification of organizational models and the factors influencing enterprise participation motivation, few studies have systematically explained how the inherent institutional logic differences in different models, through the adaptability of governance structure design, affect enterprise satisfaction with talent. This gap limits the targeting and effectiveness of governance optimization in industry-education integration (Michel, Meza, & Cejudo, 2022).
Therefore, based on institutional logic theory, this study constructs an analytical framework of ‘organizational model—governance structure adaptability—talent quality satisfaction’ to reveal how the market logic, cooperation logic, and coordination logic embodied in different models influence enterprise evaluation through governance dimensions such as clarity of decision-making and fairness of profit distribution. Through cross-provincial surveys and regression analysis across multiple institutions, the study not only provides empirical evidence for understanding the matching mechanism between model selection and governance design, but also offers practical guidance for vocational colleges to optimize cooperative governance and enhance the quality of talent development.
2. Construction of Theoretical Analysis Framework (1)
Theoretical Framework: Analysis of the Governance
Effectiveness of Industry-Education Integration
from the Perspective of Institutional Logic
2.1. Theoretical Framework: Analysis of the Governance Effectiveness of Industry-Education Integration
from the Perspective of Institutional Logic
This study aims to reveal how the institutional attributes of industry-education integration organizational models influence corporate satisfaction with talent quality through the adaptability of governance structures. To systematically analyze this causal chain of ‘organizational model-governance structure-performance outcomes,’ this study introduces Institutional Logic Theory as the core analytical framework. This theory, pioneered by Friedland and Alford and further developed by scholars such as Thornton and Ocasio, primarily explains how multiple institutional orders in social systems (such as market, state, profession, and community) shape organizational behavior and governance structures through their respective core principles and action templates, thereby impacting organizational performance (Cao, 2015). This theoretical perspective aligns closely with the complex context of industry-education integration, which involves multiple stakeholders and intertwined logics, providing a powerful tool for understanding the governance needs and effectiveness differences of various organizational models (Williamson, 2016).
The stakeholder collaboration theory was proposed by Freeman (1984), and its core assertions can be summarized into three points: First, the survival and development of an organization depend on the support of all ‘stakeholders’ (i.e., groups affected by or capable of affecting the organization’s decisions), rather than serving only a single entity such as shareholders; second, the premise of effective collaboration is the accurate identification of the core demands of each stakeholder and achieving a balance of interests through resource exchange; finally, the ultimate goal of collaboration is to build a ‘win-win’ ecosystem, in which all parties create value beyond individual capabilities through cooperation. In this theory, the ‘differences in interest demands,’ ‘efficiency of resource matching,’ and ‘fairness in responsibility sharing’ are considered key variables affecting the effectiveness of collaboration, which aligns closely with this study’s focus on ‘school-enterprise resource matching, responsibility-sharing, and goal coordination.’ (Govender & Taylor, 2015).
Furthermore, the different organizational models of industry-education integration can be regarded as concrete carriers of specific institutional logics. Institutional logic, as the “guiding principles that direct organizational behavior” (Thornton et al., 2012), is manifested in this study in three types: market logic, cooperation logic, and synergy logic. Market logic **emphasizes clearly defining transaction conditions through contracts, focusing on efficiency and striving for optimal cost-benefit outcomes. The order-based class is a typical manifestation of this logic, where enterprises invest in equipment, teaching staff, and other resources to obtain customized talents that meet job requirements. The key to governance lies in establishing clear and executable rules for the exchange of benefits (Williamson, 2016).
Cooperation logic focuses on deep embedding between parties based on trust and complementary resources, emphasizing joint governance and process coordination. Industry academies are closer to this logic, and their effectiveness depends on building a cooperative mechanism that leverages complementary resources and collaborative decision-making (Ostrom, 1990). Synergy logic applies to multi-stakeholder alliances, emphasizing balancing diverse demands through a networked governance structure to achieve system-level coordination. Vocational education groups, as multi-stakeholder alliance organizations, operate according to synergy logic and need to construct governance structures that can integrate resources from various parties and coordinate objectives (Gray, 1989). This correspondence between institutional logic and organizational models provides a theoretical basis for analyzing the design principles of governance structures and highlights the central position of the three logics—market, cooperation, and synergy—as the core of this analytical framework.
Order-based classes essentially reflect market logic, where enterprises invest resources (such as equipment and faculty) to acquire tailored talent that meets job requirements, with governance focused on establishing clear and actionable rules for interest exchange. Industry colleges align more closely with cooperative logic, emphasizing deep integration and joint governance by both schools and enterprises throughout the talent cultivation process, with effectiveness dependent on establishing complementary resources and co-decision-making mechanisms. Vocational education groups, as alliance organizations involving multiple actors, operate according to collaborative logic, requiring governance structures capable of balancing various demands and achieving networked coordination. The correspondence between institutional logics and organizational models provides a theoretical foundation for analyzing the design principles of governance structures.
2.2. Theoretical Framework: From Institutional Attributes to
Governance Effectiveness
Based on institutional logic theory, this study constructs an analytical framework of ‘institutional attributes—governance structure adaptability—talent quality satisfaction’ (as shown in Figure 1). This framework includes three key components:
First, define the institutional attributes of different organizational models. This is the starting point of the analysis. The institutional attributes of the order-based class are dominated by market logic, and its successful operation depends on establishing governance arrangements that ensure enterprise investments yield reasonable returns. The institutional attributes of industry colleges are driven by cooperation logic, and their effectiveness depends on the ability to establish governance structures that ensure deep involvement and shared responsibilities between schools and enterprises. The institutional attributes of vocational education groups are the most complex, shaped by market, cooperation, and collaboration logics, and their governance needs to meet the heterogeneous demands of multiple stakeholders, achieving integration and coordination at the system level. Second, operationalize the core mediating variable, ‘governance structure adaptability.’ Governance structure adaptability refers to the extent to which the governance mechanisms adopted by a specific organizational model (such as decision-making mechanisms, benefit distribution mechanisms, supervision and evaluation mechanisms, etc.) align with the intrinsic requirements of its dominant institutional logic. The higher the degree of alignment, the stronger the adaptability. This study focuses on two observable and measurable key dimensions: (1) clarity of the decision-making mechanism, i.e., whether the rules, procedures, and authorities for enterprises to participate in major talent training decisions (such as program development, curriculum design, and quality assessment) are clear and transparent, which directly relates to whether the shared governance required by cooperation and collaborative logics can be realized; (2) satisfaction with benefit distribution, i.e., enterprises’ subjective evaluation of the balance between their resource inputs (funding, technology, equipment) and the returns obtained (talent quality, technological outcomes, social reputation), which directly reflects whether the core concerns of market logic are met. Finally, establish ‘enterprise satisfaction with talent quality’ as the performance outcome variable. This variable represents enterprises’ comprehensive evaluation of the extent to which the talents cultivated through industry-education integration meet their needs in knowledge structure, skill level, and professional quality, and it is a key indicator for measuring the effectiveness of industry-education integration practices.
The core theoretical proposition of this study is that the fit between governance structure and the inherent institutional logic of its organizational model, in terms of the two dimensions of ‘clarity of decision-making mechanisms’ and ‘satisfaction with benefit distribution,’ serves as a key mediating mechanism influencing corporate satisfaction with talent quality. When the governance structure is highly aligned (for example, when an order-based training program has a clear and satisfactory benefit distribution mechanism that accurately responds to market logic; or when an industry academy establishes democratic and transparent decision-making mechanisms that fully reflect cooperative logic), the organization’s institutional demands (such as return on investment and voice) are effectively met, thereby significantly enhancing positive evaluations of talent development quality. Conversely, if the governance structure becomes misaligned with the institutional logic (for example, unclear benefit distribution in order-based classes, or decision-making in industry colleges monopolized by a single party), it can lead to cooperative friction and insufficient incentives, reducing enterprise satisfaction with talent quality. The performance of vocational education groups depends on whether their governance structures can effectively reconcile multiple logics and achieve a complex structural balance.
![]()
Figure 1. Analytical framework based on institutional logic theory.
This analytical framework organically connects the macro-level institutional environment, the meso-level organizational patterns, and the micro-level governance mechanisms and performance outcomes. It not only provides clear hypotheses about variable relationships for empirical analysis but also deepens the theoretical understanding of the root causes of differences in the effectiveness of industry-education integration governance, laying a solid foundation for exploring optimization pathways.
3. Research Content Design and Survey Implementation
3.1. Using Structured Questionnaires as Measurement Tools
This study takes as its research population the partner companies of higher vocational colleges in Sichuan Province that participate in three types of industry-education integration models: order classes, industry colleges, and vocational education groups, and constructs a sampling frame. To ensure the representative and structural validity of the sample, a stratified random sampling method was used: first, based on the regional economic development levels within Sichuan Province, the sampling range was divided into five strata: Chengdu Plain Economic Zone, Southern Sichuan Economic Zone, Northeastern Sichuan Economic Zone, Panxi Economic Zone, and Northwestern Sichuan Ecological Demonstration Zone; then, from each stratum, 2 higher vocational colleges listed as construction units under the “Double High Plan” or as provincial-level or above industry-education integration demonstration projects were selected, totaling 10 sample colleges; finally, from the school-enterprise cooperation database of each selected college, partner companies were randomly selected according to a ratio of approximately 1:2:1 for order classes, industry colleges, and vocational education groups (This proportion is mainly set with reference to the distribution of the number of industry-school cooperation models within the province as published in the ‘Annual Report on the Quality of Higher Vocational Education in Sichuan Province (Sichuan Provincial Education Department, 2022)’. Since industry colleges require a high level of cooperation depth and generally involve key institutions, their actual distribution density is relatively high, resulting in a slightly higher proportion). The sample size was determined following the principle of 10 times the number of independent variables in multiple regression analysis, with a 20% buffer for invalid questionnaires, resulting in a total of 300 questionnaires distributed. After excluding incomplete or logically inconsistent questionnaires, 242 valid responses were collected, with an effective recovery rate of 80.7%, meeting the basic requirements for statistical analysis. The study used a structured questionnaire as the measurement tool, which was designed strictly in accordance with psychometric standards. The content covered four parts: (1) Basic information of the enterprise (5 items), used to control for variables such as enterprise size, industry, and years of cooperation; (2) Organizational model identification (1 item), where companies confirmed the primary cooperation model they participated in; (3) Governance structure adaptability scale (8 items), including two dimensions: clarity of decision-making mechanisms and satisfaction with benefit distribution, scored on a 5-point Likert scale; (4) Talent quality satisfaction scale (6 items), used to measure companies’ overall evaluation of graduates’ job fit, professional skills, and innovation ability. The questionnaire underwent content validity assessment by three experts in vocational education, achieving an agreement rate of 85%. In the pre-survey phase (n = 30), two items with decision values below 3 were removed through item analysis. The final scales demonstrated good reliability (Cronbach’s α for the governance structure scale was 0.81, and for the talent quality satisfaction scale was 0.84). Data collection was conducted from June to September 2025.
3.2. Descriptive Statistics of Sample Basic Characteristics and Key
Variables
The descriptive analysis results of the basic characteristics of valid samples (N = 242) are shown in Table 1. The sample structure indicates that: in terms of enterprise scale, small and medium-sized enterprises account for the majority (81.4% in total), which is consistent with the common pattern of school-enterprise cooperation in vocational colleges; in terms of industry distribution, manufacturing (40.5%) and information technology (26.9%) dominate, reflecting the close alignment between the policies of industry-education integration and industry upgrading directions; in terms of cooperation duration, nearly half of the enterprises (47.5%) are in a stable cooperation period of 4 - 6 years, suggesting that most school-enterprise cooperation relationships have entered a mature stage; in terms of participation mode, the industry academy model accounts for the highest proportion (45.0%), indicating that the current practice of industry-education integration is evolving from shallow cooperation to deep collaboration. Overall, the sample is highly representative across different dimensions. Data quality tests show that the Cronbach’s α coefficients of key variables range from 0.80 to 0.84, the KMO value for exploratory factor analysis is 0.86, Bartlett’s test of spherical is significant (p < 0.001), and common method bias is not significant, indicating that the measurement instruments have good reliability and validity, and the data are reliable.
Table 1. Basic characteristics of the sample (N = 242).
Variable Type |
Frequency (n) |
Percentage (%) |
Enterprise scale |
≤100 people |
85 |
101 - 500 people |
112 |
≥501 people |
45 |
Industry |
Manufacturing industry |
98 |
Information technology |
65 |
Service industry |
52 |
other |
27 |
Years of cooperation |
≤3 years |
78 |
4 - 6 years |
115 |
≥7 years |
49 |
Participate in the integration of industry and education model |
Order Class |
85 |
Industry College |
109 |
Vocational Education Group |
48 |
The descriptive statistics of the core variables (see Table 2) reveal that satisfaction with talent quality has the highest mean (M = 3.91), followed by clarity of the decision-making mechanism (M = 3.85), while satisfaction with benefit distribution is relatively lower (M = 3.72). This result preliminary confirms that the core motivation for enterprises to participate in industry-education integration—acquiring high-quality talent—is largely being met. At the same time, it suggests that the benefit distribution mechanism remains a weak link in current collaborative governance. In terms of dispersion, satisfaction with benefit distribution has the highest standard deviation (SD = 0.68), indicating significant variation in evaluations across different enterprises, potentially influenced by factors such as enterprise size, industry attributes, and cooperation models. The standard deviations for decision-making mechanisms and talent quality evaluations are relatively smaller, suggesting that enterprises have a more consistent assessment of process standardization and output effectiveness. These findings provide an important basis for further in-depth analysis of the role of governance structures (Shen & Ouyang & Ouyang, 2021).
Table 2. Descriptive statistics of key variables (N = 242).
Variable |
Mean (M) |
Standard deviation (SD) |
Range of values |
Clarity of decision-making mechanism |
3.85 |
0.62 |
1 - 5 |
Satisfaction with profit distribution |
3.72 |
0.68 |
1 - 5 |
Talent Quality Satisfaction |
3.91 |
0.59 |
1 - 5 |
4. Hypothesis Test Results
4.1. Differences in Governance Structure under Different
Organizational Models
The results of the one-way ANOVA indicate that under different models of industry-education integration, there are significant differences in enterprises’ perceptions of governance structures (see Table 3). The inter group difference in decision-making mechanism clarity is significant (F = 12.34, p < 0.001): enterprises in the industry college model have the highest clarity of decision-making mechanisms (M = 4.12, SD = 0.51), significantly higher than that of the order-based classes (M = 3.56, SD = 0.65, p < 0.001) and vocational education groups (M = 3.78, SD = 0.60, p < 0.01); the clarity in vocational education groups is also significantly higher than in the order-based classes (p < 0.05). The inter group difference in satisfaction with benefit distribution is also significant (F = 8.76, p < 0.001): the industry college model has the highest satisfaction with benefit distribution (M = 3.90, SD = 0.58), significantly higher than the order-based classes (M = 3.45, SD = 0.72, p < 0.001); vocational education groups (M = 3.62, SD = 0.65) are significantly higher than the order-based classes (p < 0.05), but lower than the industry college model (p < 0.05).
These results indicate that the depth of collaboration in the organizational model directly affects perceptions of governance structures: the industry college, as a deeply integrated model of ‘jointly run by schools and enterprises with shared resources,’ establishes joint school-enterprise committees and clarifies decision-making processes (e.g., jointly formulating training programs), allowing enterprises to fully participate in talent training decisions, thus achieving the highest clarity in decision-making mechanisms. At the same time, industry colleges clarify benefit distribution through rules such as ‘retention incentives and technology achievement dividends,’ improving enterprise satisfaction. Order-based classes, focusing on ‘job matching,’ often have enterprises ‘passively receiving students,’ with fewer opportunities to participate in decision-making and unclear benefit distribution (e.g., only agreeing on ‘priority hiring’ without specifying retention rates), resulting in the poorest perception of governance. Vocational education groups involve multiple stakeholders including government, industry, enterprises, and schools, leading to more complex decision-making processes (e.g., coordinating multiple interests), and benefit distribution is less precise compared to industry colleges but better than order-based classes.
Table 3. Differences in governance structure dimensions under different organizational models (N = 242).
Variable |
Order Class
(n = 85) |
Industry College
(n = 109) |
Vocational Education Group
(n = 48) |
F-value |
p-value |
Clarity of decision-making mechanism |
3.56 ± 0.65 |
4.12 ± 0.51 |
3.78 ± 0.60 |
12.34 |
<0.001 |
Satisfaction with Profit Distribution |
3.45 ± 0.72 |
3.90 ± 0.58 |
3.62 ± 0.65 |
8.76 |
<0.001 |
4.2. The Impact of Governance Structure on Talent Quality
Satisfaction
The results of the multiple linear regression analysis are shown in Table 4. Model 1 includes control variables and has a significant explanatory power for talent quality satisfaction (adjusted R2 = 0.10, p < 0.05), with enterprise size and years of cooperation showing a significant positive impact. After adding the clarity of the decision-making mechanism and satisfaction with benefit distribution in Model 2, the model’s explanatory power significantly improved (adjusted R2 = 0.25, ΔR2 = 0.15, p < 0.001). Both key independent variables exhibited significant positive effects (decision-making mechanism clarity β = 0.32, satisfaction with benefit distribution β = 0.28, p < 0.01), strongly confirming the central role of governance structure adaptability in enhancing talent quality satisfaction. Model 3 further introduced the interaction terms between organizational model and governance structure, and the model’s explanatory power increased again (adjusted R2 = 0.30, ΔR2 = 0.05, p < 0.05). Specifically, the interaction effect of ‘Industry College × Decision-Making Mechanism Clarity’ was significant (β = 0.18, p < 0.05), indicating that the deep cooperation model of Industry Colleges can enhance the positive effect of decision-making participation on talent satisfaction. In contrast, the interaction effects of order-based classes and vocational education groups were not significant, which, from a governance perspective, confirms the boundary of the effects of different models: order-based classes focus on short-term benefit exchange, while the synergy of vocational education groups is constrained by the complexity of multi-stakeholder coordination.
In summary, the significant interaction occurs precisely between the core dimension of ‘cooperative logic’ and ‘co-governance’ (clarity of decision-making mechanisms), which is not accidental but theoretically inevitable. It indicates that the effectiveness of governance mechanisms is not universal; its efficacy heavily depends on whether it is ‘in resonance’ with the institutional logic embedded in the organizational model (Li, 2015). This finding of our study transforms ‘governance structure adaptability’ from an abstract concept into a verifiable mechanism, clearly pointing out: for industry-academia alliances aiming to deepen cooperation, establishing a joint decision-making mechanism that is clear in responsibilities and transparent in operation is the ‘key lever’ to unleash their collaborative potential and improve educational outcomes.
Table 4. Results of multiple linear regression analysis (Dependent variable: talent quality satisfaction, N = 242).
Variable |
Model 1 |
Model 2 |
Model 3 |
Constant term |
3.21** |
2.56*** |
2.34*** |
Company size |
0.15* |
0.12* |
0.10 |
Industry |
0.08 |
0.06 |
0.05 |
Years of collaboration |
0.12* |
0.10* |
0.09 |
Clarity of decision-making mechanism |
— |
0.32** |
0.28** |
Satisfaction with profit distribution |
— |
0.28** |
0.25** |
Industry-academia collaboration × clarity of
decision-making mechanism |
— |
— |
0.18* |
Adjusted R2 |
0.10 |
0.25 |
0.30 |
ΔR2 |
— |
0.15*** |
0.05* |
F-value |
3.56* |
10.23*** |
8.78*** |
Note: p < 0.05, p < 0.001; Model 1 is the control variables model, Model 2 is the core independent variables model, Model 3 is the interaction term model.
5. Research Findings
This study, through empirical analysis, arrives at three core findings. First, there are gradient differences in governance structures under different industry-education integration organizational models: industrial colleges significantly outperform vocational education groups and order-based classes in both decision-making clarity and satisfaction with benefit distribution, reflecting their governance advantages in deep collaboration. Second, regression analysis confirms that both dimensions of governance structure have a significant positive impact on satisfaction with talent quality, with the contribution of decision-making clarity slightly higher than that of benefit distribution satisfaction, highlighting the crucial role of procedural fairness in enterprise evaluation. Finally, there is a significant interaction effect between organizational model and governance structure; the industrial college model can enhance the positive impact of decision-making clarity on talent satisfaction, revealing the alignment between governance effectiveness and institutional environment.
These findings systematically address the core research questions and distill the key mechanism of ‘governance structure adaptability.’ The results indicate that improving the performance of industry-education integration depends not only on the perfection of governance structures but more crucially on governance design being aligned with the institutional logic embedded in the organizational model. For example, the ‘collaborative logic’ of industrial colleges requires democratized decision-making to integrate school-enterprise resources, whereas the ‘market logic’ of order-based classes relies more on clear benefit distribution to ensure exchange efficiency (Li, 2023). Based on the above conclusions, this study provides the following implications for university management and policy-making: For university administrators, the implication is that they should abandon a ‘one-size-fits-all’ approach to collaboration. When selecting or designing an industry-education integration model, they must first assess the goals and core logic of the cooperation: if the aim is to quickly respond to specific job skill demands, priority can be given to the order-class model, with a focus on improving benefit contracts; if the goal is deep strategic-level talent cultivation and collaborative innovation, planning industrial colleges should be prioritized, with efforts to establish a shared governance mechanism that is clear in responsibilities and transparent in decision-making. For policy-makers, the implication is that differentiated guidance and incentives need to be implemented. When evaluating and supporting industry-education integration projects, indicators should be set based on the intrinsic logic of different models (Rossi, 2010). For order-class programs, the focus should be on employment matching and contract compliance; for industrial colleges, attention should be paid to the completeness of governance structures, the stability of joint investment, and the innovativeness of talent cultivation, guiding all parties to shift from short-term transactions to long-term cooperation. Therefore, future practice should aim to build governance arrangements that correspond to the characteristics of each model. Through precise governance design, the motivation for all parties to participate can be stimulated, thereby substantially improving the quality and satisfaction of enterprise talent and promoting the in-depth development of industry-education integration.
This study has several limitations, which provide directions for future research. First, the cross-sectional data restrict the robustness of causal inferences (Williamson, 2016). Although theory supports that ‘governance structure adaptability’ affects ‘talent quality satisfaction,’ the possibility of reverse causality cannot be completely ruled out, and future studies could employ longitudinal tracking designs to verify this. Second, the sample is concentrated in a single province; while this helps control for regional variations, the generalizability of the conclusions needs to be tested across a broader area (Shen & Hu, 2024). Additionally, this study focuses on two core governance dimensions: decision-making mechanisms and benefit distribution. Future research could incorporate mechanisms such as risk sharing and conflict coordination to more comprehensively reveal the impact of governance systems. Highlighting these limitations aims to objectively advance related discussions rather than negate the theoretical and practical value of this study.
Funding Project
This paper is a phased research result of the Sichuan Post and Telecommunication College-level scientific research project “Research on the Current Situation and Optimization Strategies of Industry-Education Integration in Higher Vocational Colleges in the New Era” (Project No. YDXJKY202437).