Designing of an open innovation model in science and technology parks

The main purpose of this research is to design and test an open innovation model in science and technology parks through a mixed exploratory approach. The data collection tool in the qualitative section is a semi-structured interview with 15 experts in the field of science and technology parks. Purposive sampling was used until theoretical saturation is achieved. The grounded theory was used to coding the interviews. As a result of open, axial, and selective coding in MAXQDA, 1551 free codes, 202 concepts, 73 sub-codes, and 21 main codes extracted. These codes categorized in six themes of causal conditions, central phenomenon, strategies, contextual conditions, intervening conditions, and outcomes. Initial questionnaire was administrated among 516 experts and managers of science and technology parks in Tehran province and affiliated companies. Confirmatory factor analyses showed high model fit with CFI equal to .93 and high internal consistency for the total score equal to .98. The results of this study showed that the proposed model of open innovation in science and technology parks in Iran has a good fit. Supplementary Information The online version contains supplementary material available at 10.1186/s13731-022-00203-w.


management;
(3) technology monitoring; (4) market knowledge; (5) predicting market needs, and (6) product and technology planning. The goal of open innovation is to capitalize on the discoveries and innovations of others in the innovation process, as opposed to closed processes, in which companies operate only with their own professional ideas, capabilities and capacities (Schwab et al., 2013).
In the following years, various theories about open innovation emerged. Tushman and Anderson (1986) proposed the theory of discontinuities innovation versus fundamental innovation. According to this theory, knowledge is the main foundation of any company's ability, then any change in knowledge indicates a change in the organization's ability to provide a new service and it emphasized the organization's ability to innovate. Martins and Terblanche (2003) in their theories expressed the cultural factors influencing organizational innovation as risk acceptance, open communication between employees, ambiguity and conflict, encouraging new theories, job commitment, strong and supportive leadership, customer orientation, and increasing benefits. Baumard (2009) expressed the theory of innovation with an integrated and networked system; its most important components are introduced as integrating development strategies between different internal and external organizations, simulation in research and development and close communication with people in the society, company flexibility to change and speed in development and more focus on quality to price (Spithoven et al., 2013).
There are many benefits to using open innovation, including early participation in new technologies and job opportunities, delays in financial commitment, mitigation of losses on initial outflows, and delays in outflows if investment is possible (Vanhaverbeke et al., 2008). Open innovation also makes it possible to exploit the knowledge of smart people outside the company; it enables the simultaneous use of domestic and foreign research and development; there is no need for the researcher to invest in order to make a profit from the research; the key to success will be the best use of internal and external ideas (Docherty, 2006). Mansfield (1986) showed that innovative projects that rely heavily on external development have shorter development times and require less investment than similar projects that rely entirely on R&D.
In this study, Grounded Theory (Strauss & Corbin, 1990) was used to develop the model for the open innovation in science and technology parks in which using a set of data, a theory is generated. In this approach, the data are analyzed through open, axial, and selective coding. Growing connections between the developed concepts merge to form an integrated framework with a single central category in the axial coding also called theoretical coding (Glaser, 1987). Strauss and Corbin (1990) recommend examining the data and codes using a coding paradigm that concentrates on and relates causal conditions, context, intervening conditions, strategies, and outcomes to figure out the relationships between the categories.
In Iran, in the field of open innovation development, extensive research has been done and its role in university-industry cooperation has been pointed out and research cooperation has shown a positive and significant effect on university-industry cooperation (Madhooshi & Kiakajuri, 2018). An open innovation model in small and medium enterprises has been designed using grounded theory (Babaei Farsani et al., 2018). In another study, the pattern of establishing open innovation in education is designed (Rashki et al., 2020). Despite extensive research in the field of open innovation in Iran, only

Methods
In this research, the exploratory mixed method has been used. Research strategy, in the qualitative phase was grounded theory with the aim of theorizing. Grounded theory is a method for constructing a theory based on facts and data (Glaser & Strauss, 1967). The statistical population of the study includes all experts and activists in the field of science and technology parks and the sample size includes 15 of these people who were selected by purposive sampling. The inclusion criteria for participants were having at least 5 years experience of working at science and technology parks and a relevant university/college degree. Preliminary data were collected through semi-structured interviews with the target population. Data collection continued until theoretical saturation was reached. Data obtained from interviews were coded through grounded theory research approach and analyzed using MAXQDA software. The interviews were conducted on the phone due to the limitations regarding COVID-19 conditions that caused difficulties for in person interviews. Dimensions, main codes, and subcodes extracted from the interviews provided in Table 1.
In the quantitative phase, we used the developed questionnaire from the qualitative part to collect the quantitative data. The questionnaire is a self-reported 100-item with a five-point Likert scale and six factors of causal conditions, central phenomenon, strategies, contextual conditions, intervening conditions, and outcomes (see Additional file 1: Appendix S1). Data collection in quantitative phase was conducted virtually through digital format of the questionnaire sent to the participants' email addresses. We tested the findings acquired from the qualitative part using confirmatory factor analyses using SPSS v24 and Mplus v7.4 in a convenient sample consisting 516 participants. Sample size was calculated using GPower v3.1.9.2 with the effect size of 0.35 and power (1-β) of 0.95.

Development and content validity of the questionnaire
In the qualitative phase of the research, open, axial and selective coding were used. In open coding process, the data obtained from the interviews were carefully reviewed, the main and sub-themes were identified. A total of 1551 free code, 202 sub-codes, 73 main codes, and 21 dimensions were extracted. Table 3 shows the dimensions, sub-codes, and main codes extracted from the interviews.
In selective coding, the main variable or underlying process embedded in the data, how, its stages and consequences are plotted (Glaser & Strauss, 1967). Based on the obtained relationships, the concepts obtained from open and axial coding were linked to each other in the selective coding stage and were reflected as a model in science and technology parks according to Fig. 1. This figure displays the relationships between dimensions in the model and how the causal conditions was led to outcomes.  After developing the open innovation model, the initial items of the Open Innovation Questionnaire of Science and Technology Parks were written. The content validity of the questionnaire was examined by an expert panel. Content analysis was performed using content validity index (Waltz & Bausell, 1981) and 9 items were removed in this phase.

Confirmatory factor analysis
The initial questionnaire was piloted and members of Tehran Science and Technology Park answered the questions of the questionnaire and the data were studied using confirmatory factor analyses. Kaiser-Meyer-Olkin test was performed to evaluate the adequacy of the research sample. The value obtained is equal to 0.979 which indicates that the sample size is sufficient, Table 2 shows these results.
Values above 0.70 indicate the adequacy of the sample size. The KMO value showed that the variance of the sample variables was sufficient for the structural validity test. In addition, the Bartlett test was significant at the level of p = 0.05. The Bartlett sphericity test measures the correlation between variables and investigates whether the variables are sufficiently correlated to perform structural validity. Therefore, the assumptions for conducting factor analysis were approved.
The matrix of components and items of the questionnaire after rotation showed that all items had a factor load greater than 0.5, so none of the items were removed (Truong & McColl, 2011). According to the developed model in qualitative phase, we tested the fit of the first-and second-order six-factor model solution. The results  Innovation and Entrepreneurship (2022) 11:4 of confirmatory factor analysis in Mplus v7.4 with maximum likelihood estimation method and 20 iterations are reported in Table 3. The result of confirmatory factor analysis showed a comparative fit index equal to 0.934 which indicates a good fit (Kline, 2005) and the residual square root of the standardized root mean is 0.056. The RMSEA below 0.08 indicates the low error of the measurement model (Hair et al., 2010). The factor loadings of items in their respective factors for the first-order model are presented in supplementary files. Second-order model solution was performed to compare it with the first-order model. This analysis also was performed using Mplus v7.4 and maximum likelihood estimation method. Table 4 shows the fit indices for the second-order model using confirmatory factor analysis.
The comparative fit index of the model was equal to 0.875, hence, the fit indices reduced compared to the first-order model. Since, the first-order model has a better fit it was determined as the final model of open innovation questionnaire.

Reliability
Cronbach's alpha was estimated as an index of reliability, it measures internal consistency of the items in their pertinent factors and indicates the correlation between items Fig. 1 The model of the axial coding based on Grounded Theory (Corbin & Strauss, 2014) Innovation and Entrepreneurship (2022) 11:4 and the construct under study. Table 5 displays Cronbach's alpha values for open innovation factors and the total score. Cronbach's alpha of the factors varies between 0.96 and 0.98 and the Cronbach's alpha of the whole questionnaire was estimated 0.98, which indicates the high internal consistency of the factors and the total score. According to the results, the open innovation model questionnaire indicated to have sufficient validity and reliability for science and technology parks in Iran.

Discussion
The aim of the present study was to design and test an open innovation model in Iranian science and technology parks, the mixed exploratory research approach. As a result of the qualitative part, six themes of causal conditions, central phenomenon, strategies, contextual conditions, intervening conditions, and outcomes were extracted and the initial questionnaire was developed. Psychometric properties of the questionnaire were   Innovation and Entrepreneurship (2022) 11:4 studied; validity of the questionnaire was measured using confirmatory factor analysis and reliability was estimated through Cronbach's alpha. The results showed that the open innovation questionnaire for science and technology parks is a valid and reliable instrument and may be helpful with finding plausible existing deficiencies and limitations for utilizing open innovation in designing, developing and improving of science and technology parks.
In the interviews conducted in this study, the importance and usefulness of using open innovation in science and technology parks in Iran was emphasized and the interviewees mentioned many benefits for it that were reduced to sub-codes, main codes, and themes. Consistent with the findings of this study, Spithoven et al. (2013) examined the impact of open innovation on the innovative performance of small and medium enterprises compared to large firms and found that the impact of open innovation in small and medium enterprises is often different. Crema et al. (2014) also analyzed the relationship between company strategy, open innovation, and innovation performance with a focus on small and medium-sized firms and concluded that firms that follow an innovative strategy invest more in main technical skills and competencies and companies that choose diversity strategies are more likely to choose open innovation management practices exclusively. A meta-analysis of the literature on open innovation in small and medium-sized enterprises showed that they improve their overall innovation performance by choosing open innovation. They found that a large number of studies were conducted with a quantitative approach. Surprisingly, unlike many other disciplines, North American researchers have made a limited contribution but European scientists, along with some researchers from Korea, China and developing countries, have been working actively in this field (Hossain & Kauranen, 2016).
Similar study on open innovation in cyber security research institutes through Grounded Theory approach resulted in 10 sub-categories and three main categories (Ghouchani Khorasani et al., 2019). Babaee Farsani et al. (2018) also designed an open innovation model in active small-and medium-sized enterprises using Grounded Theory led to 34 sub-categories and 11 main categories in which some sub-categories were similar to sub-categories founded in the current study such as "increase absorption capacity", "improving cultural factors", and "teamwork and collective thinking". Hence, literature review showed applying Grounded Theory in model development for open innovation in industry has been functional and beneficial as it was for development of a model for open innovation in science and technology parks.

Conclusions
In the interviews conducted in this research, the importance and usefulness of using open innovation in science and technology parks in Iran was emphasized and the interviewees mentioned many benefits for it. However, participants noted that limitations, such as the lack of technology appropriate to the organizational conditions in the market, being costly, and relatively time-consuming process for implementation that need to be addressed by the authorities. In sum, the results of the present study, which was conducted with the aim of development and validation an open innovation model in Iranian science and technology parks, showed that this model is suitable for implementation in Iranian context and can measure the open innovation in science and technology parks to be used by managers, and researchers.

Implications for stakeholders
We recommend that each science and technology park assign to one specific industry since it facilitates for managers to design, develop and enhance the open innovation for that particular industry with regard to the characteristics of it. Establishing science and technology parks in universities could be helpful to quicker exchange of open innovation interactions between universities and parks. Last but not the least, the developed questionnaire in this study could be used as a tool to discover shortcomings and deficiencies in the path of promoting open innovation in the science and technology parks.
Additional file 1. Open innovation questionnaire for science and technology parks.