Knowledge spillover from Master of Science Theses in Engineering Education in Sweden

ABSTRACT

Educational collaboration activities are generally under-researched part of the interactive learning between universities and industry. The study based on 529 Master of Science (MSc) theses from a major Swedish engineering school, the faculty of engineering LTH at Lund University, shows that MSc theses are a frequently used collaboration mechanism between university and industry in engineering education. Collaborating firms benefit mainly by recruiting students and the transfer of technological knowledge. Moreover, mainly regional large firms benefit from the educational collaboration and firms collaborate mostly on knowledge related to product innovations and in the early stages of the product innovation process. Small and medium-sized enterprises (SMEs) tend to relate to newer engineering disciplines, have relatively more MSc thesis projects related to product innovations and later process stages, specifically in prototype development and testing, than large firms have. The paper suggests a number of university actions to increase knowledge spillover to SMEs.

KEYWORDS:

• University-industry collaboration
• educational collaboration
• MSc theses
• knowledge spillover
• product innovation
• small and medium-sized firms

Introduction

Master theses and project work in the students’ education are according to Bramwell and Wolfe (2008) an important but in university-industry collaboration research often underestimated part of the knowledge spillover and interactive learning between universities and industry, especially for small and medium-sized enterprises (SMEs). Knowledge spillover (e.g. Andersson, Quigley, and Wilhelmson 2004; Audretsch and Feldman 1996; Agarwal, Audretsch, and Sarkar 2010) means that firms and organisations can benefit from the knowledge gained from university research and development (R&D) and education e.g. by engaging students in a bachelor or master thesis, recruiting the educated students as well as come in contact with research and teachers/researchers (Kunttu 2017).

University-industry collaboration is typically described as involving collaborative R&D, contract R&D, personnel mobility and educational collaboration (e.g. D’Este and Patel 2007; Perkmann et al. 2013). While educational collaboration activities with universities often are mentioned as potential sources of innovation knowledge for the industry, it has with few exceptions being empirically investigated (e.g. Bramwell and Wolfe 2008; Kunttu 2017). The topic of university-industry collaboration in engineering education has received some research interest but limited to identifying societal drivers, such as globalisation and speed of Information and Communication Technology (ICT) development and related challenges and opportunities for the engineering education (Korhonen-Yrjänheikki, Tukiainen, and Takala 2007). Research focusing on the actual impact and knowledge spillover to the industry from educational collaboration activities, in the engineering field, seems to be limited to four studies by Bramwell and Wolfe (2008), Kunttu (2017), Lucia et al. (2012) and Thune (2011). Bramwell and Wolfe's (2008) case study of University of Waterloo, a technical and natural science university in Canada, shows two main advantages for firms involving themselves in internships and master theses: (1) the companies will be able to more thoroughly observe, test and evaluate job candidates before possibly recruiting them, and (2) students transfers new knowledge, both explicit and implicit, to the firms. Two other case-based studies by Kunttu (2017) in Finland, and by Lucia et al. (2012) in Spain, corroborate these findings and show that in a European educational system the master of science (MSc) theses are one of the most commonly used knowledge spillover mechanism to the industry. However, none of these studies systematically researches the knowledge spillover from MSc theses to the industry.

Bramwell and Wolfe (2008) mostly discuss the benefits for SMEs while the studies by Kunttu (2017), Lucia et al. (2012) and Thune (2011) concern benefits for large firms. This poses the question to what extent knowledge from MSc theses spills over to large firms or to SMEs. On the one hand, we could expect SMEs, because of their more limited resource base, to more frequently use external sources of knowledge than large firms as prior research has indicated (e.g. Edwards, Delbridge, and Munday 2005). On the other hand, we could also expect SMEs to be a less frequent user of external knowledge sources than large firms are as they have access to fewer resources, smaller organisations and fewer people to commit to external knowledge exchange (Lee et al. 2010). The smaller resource base of SMEs also indicates that SMEs would need more assistance, e.g. intermediaries (Lee et al. 2010), to be able to utilise external knowledge in a more comprehensive way. Bramwell and Wolfe (2008) indicate that university-industry educational collaboration is particularly valuable to SMEs. By investigating the differences of knowledge spillover from MSc theses to large firms and SMEs we could increase knowledge on the current situation of knowledge spillover and its distribution to large firms or SMEs. In case of SMEs being disadvantaged, we could propose supportive components in the engineering education to better serve the needs of innovation-related knowledge for SMEs.

The lack of systematic research regarding university-industry educational collaboration in general and knowledge spillover from engineering students’ master theses to the industry in particular has motivated us to do a survey study of MSc theses in engineering. We have studied the MSc theses at one of Sweden’s major engineering schools, the Faculty of Engineering LTH¹ (from now on LTH), at Lund University in Sweden, situated in the most southern part of Sweden. LTH has roughly 9000 students (mostly engineering students but also students in architecture, industrial design etc.), some 700 engineering graduates and 500 MSc theses each year. This study intends to contribute to the university-industry educational collaboration by investigating the extent and form of innovation-related knowledge spillover from master thesis projects in engineering education and possible differences between large firms and SMEs. While there is also knowledge spill in from industry to university, e.g. from more research-oriented MSc theses, this study focuses only the knowledge spillover from university to the industry through MSc thesis projects. The research questions are:

• To what extent are master theses done in collaboration with industry in engineering education?

• What type of firms are involved in the theses, i.e. large firms or SMEs?

• What type of innovation-related knowledge spills from the master thesis projects to the firms and are there differences between large firms and SMEs?

• How can engineering education increase the innovation-related knowledge spilled to SMEs?

The rest of the paper is structured as follows. Next section will briefly review the research on knowledge spillover from an educational collaboration between university and industry. Thereafter we shortly present the method used in the study. In the third section, we present the results and then finish with concluding discussions of the results and implications for engineering education.

University-industry educational collaboration and knowledge spillover

Theories of knowledge spillover from university R&D to surrounding companies (e.g. Audretsch and Feldman 1996) generally base their reasoning on theories of knowledge-based (endogenous) economic growth (Romer 1986). Andersson, Quigley, and Wilhelmson (1996) found that innovative activity linked to R&D knowledge is geographically concentrated and that this concentration can be explained by proximity to highly educated labour, university R&D and firm R&D. However, the proximity to university R&D is for most firms less interesting than proximity to a pool of skilled labour and a constant flow of graduate students (Bramwell and Wolfe 2008; Kunttu 2017). Even for companies that have extensive R&D activities, higher education and access to a pool of educated labour often play a more important role than proximity to relevant university R&D (Andersson, Quigley, and Wilhelmson 2004).

There are several reasons why university-industry educational collaboration, relative to R&D collaboration, are and will be even more attractive for industry. In most developed economies, the service sector dominates the economy and particularly the knowledge-intensive service sectors are growing (OECD 2012), a sector that relies heavily on recruiting higher educated engineers. Moreover, the increasing globalisation and ICT development throughout almost all industries put a premium on young engineers with generally better language, cultural and ICT skills than previous generations of engineers (Korhonen-Yrjänheikki, Tukiainen, and Takala 2007).

In Bramwell and Wolfe’s study (2008) point to two major advantages with student internships and thesis work; (1) Through internships and thesis work, companies are given the opportunity to more thoroughly observe, test and evaluate job candidates before hiring. (2) Students transfer new knowledge and skills, which means, among other things, that they educate employees in certain areas, such as qualified use of ICT. Kunttu (2017) corroborate these findings in her study of nine cases of educational collaboration activities, most of them with an engineering content, that master thesis projects are the most important forms of knowledge transfer for the industry. The firms involved in educational collaboration activities mostly appreciated the benefits of (1) recruiting student graduates with specific competences, (2) gaining skills and new information from the academic world, (3) facilitating industry to utilise new information from the university, and (4) deepened understanding of university research processes and collaborative practices.

The transfer of new skills and new information in educational collaboration is one source of external knowledge in firms’ open innovation strategies (Lucia et al. 2012). The case study by Lucia et al. (2012) provides an example of how a large firm (Bosch and Siemens Home Appliance Group – BSH) collaborates with the University of Zaragoza in both research and education in order to develop the induction heating technology since 1990. Over the years, the collaboration has helped BSH to develop induction heating from an exclusive and bulky appliance to a built-in, efficient and mass-produced device (Lucia et al. 2012). The case study provides examples of the product as well as process innovations and involvement at various stages of the new product development process.

Thune (2011) reviews the research on R&D-based university-industry partnerships and identifies three categories of success factors (contextual, organisational, process factors) for such partnerships. The contextual factors are choice of partners and geographical and cognitive proximity. Organisational factors concern formal organisation of and top leadership commitment for collaboration activities. Process factors are how the partnerships are actually managed, i.e. project management, and how communication is handled in the partnership. In her four case studies, she checks the relevance of these success factors for educational collaboration in engineering. She finds that the three success factors are relevant also for educational collaboration albeit with a few differences in emphases. She finds that geographical proximity seems more important in educational collaboration and the cumulative manner of successful partnerships (prior successful collaboration leads to more collaboration) requires the universities to allocate resources to initiate and coordinate new partnerships by initiating small projects with new industrial partners.

In order to determine what type of innovation-related knowledge that is spilled we have chosen to define innovation according to OECD’s definition (OECD 2005) based on Schumpeter’s (1942) innovation definition. As we expect most master theses to be related to product innovations we have used Cooper’s (2008) definition of the innovation process ‘as a series of stages’ and used six stages in this process, from screening to commercialisation, to position innovation-related knowledge spillover in the innovation process (for details see Box 1). As recommended by Cooper (2008) we have broadened the scope of the stage-gate process to include also other types of innovation projects, such as innovations in manufacturing, distribution, recycling and maintenance processes, by asking for development or adaptations in technologies, methods, concepts, and products (see Appendix A).

Box 1. Innovation definition and stages in the innovation process used in this study.

Innovation definition

‘An innovation is the implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organisational method in business practices, workplace organisation or external relations’. (OECD 2005, 46)

Stages in the innovation process

‘The innovation process can be visualised as a series of stages, with each stage comprised of a set of required or recommended best-practice needed to progress the project to the next gate or decision point’. (Cooper 2008, 214)

Stages used in this report are; screening of new ideas and technologies, scoping of ideas and technologies, development of a business case, development of a prototype, testing and validation, commercialisation.

Kunttu (2017), Lucia et al. (2012) and Thune (2011) all provide examples of large firm educational collaboration with the university. Laursen and Salter (2004) found that large firms dominate university-industry research collaboration, especially the large firms with access to R&D-capabilities and large firms with an open innovation strategy, i.e. firms that used many external knowledge sources in their innovation processes. Large firms have in general better access to R&D-capabilities than SMEs as well as a better capacity to search for new knowledge (Lee et al. 2010). Thus, we expect large firms to dominate also educational collaboration even though educational collaboration could be seen as relatively more valuable for SMEs (Bramwell and Wolfe 2008). Lee et al. (2010) propose that SMEs need of innovation-related knowledge is more oriented to the exploitation phase of innovation, e.g. development of prototypes, marketing, and commercialisation activities, rather than the exploration phase of innovation, e.g. technology development, ideation, the latter being more compatible with large firms’ R&D-oriented innovation process.

Empirical context

The Swedish higher education system has generally limited interaction with industry as most of it is performed by state government funded and organised universities and higher education institutes (Jacob, Lundqvist, and Hellsmark 2003). However, the level of interaction between university and industry varies greatly between different faculties, i.e. applied sciences such as engineering having a much more frequent collaboration with industry in R&D and education than less applied sciences (Bengtsson 2011, 2013). Since the mid-1990s there have been national policy efforts to support research knowledge coming into use with a law specifying ‘a third mission’, education and research being the first two missions, for Swedish universities coming into effect in 1997 (Jacob, Lundqvist, and Hellsmark 2003). As part of this national policy to increase interaction between university and industry the Swedish state funds a regional system of university Technology-transfer offices (TTOs) connected to all Swedish universities focusing on university spin-offs (Bengtsson 2017). The national innovation agency, Vinnova, finances various university-industry innovation programmes, which requires co-financing from industry (Bengtsson 2017). However, there has been very limited attempts from the Swedish government to support university-industry educational collaboration programmes and projects (Bengtsson 2013).

In Sweden, compulsory internships are unusual in the higher education of engineering degrees. Instead, there is a tradition to finish the degree programmes, both at bachelor and master level, with an independent major study reported in a bachelor or master thesis. Some theses are done in collaboration with firms, public organisations, and non-profit organisations while others are more research-oriented. The knowledge spillover effects of these theses are, on a more systematic level, at least to our knowledge, not known. This research focuses on MSc theses completed at the end of five years of study at the Faculty of Engineering LTH at Lund University (LU). The faculty of engineering LTH is, in terms of a number of students, the largest faculty at LU, and mainly educate engineers in 16 educational programmes leading to an MSc degree in engineering. At LTH, thesis projects are the only systematic opportunity in the undergraduate and graduate degree programmes to offer the students contact with working life and practice their knowledge in cooperation with a company or other organisation over a longer time (30 credit points equivalent to one semester or approx 4.5 months for the MSc thesis). Either individual students (approx 60% of the theses) or maximum two students (40%) perform the MSc projects. The objective of the MSc theses at LTH is: The aim of the degree project is for the student to develop and demonstrate the requisite knowledge and skills to work independently as an engineer (LTH Course Syllabus for Degree Projects 30 credit points 2016). Of course, many students have contact with working life and firms through summer employment or part-time employment but this is not part of their educational programme.

Method

The lack of prior systematic knowledge on the extent and type of knowledge spillover from MSc theses in engineering education motivated us to use a quantitative method surveying all MSc theses at a major engineering faculty in one year. The sample consisted of all registered MSc theses in engineering at the Lund University engineering school, LTH, completed during the year 2016. We identified 529 MSc theses supervised at 16 LTH departments. We have coded all master theses on authors, subject, the title of theses, supervisors, external collaboration partner (if any) and connections to departmental research. A thesis was coded as ‘collaborative’ if the thesis contained a clear statement of being done in collaboration, or on commission with one (or in a few cases several) specific firm(s) or organisation(s) or had an external supervisor from a firm or organisation. In a limited number of theses, the collaboration firm or organisation was anonymous due to secrecy concerns from the firm or organisation. They were nevertheless coded as ‘collaborative theses’ but could then not be included in the second survey to the collaborating firms or organisations.

The second survey was conducted as a telephone interview to the external collaboration partners (firms, organisations) and the specific external supervisor or contact person mentioned in the master thesis. The questionnaire to the firms and organisations contained two sections of questions. The first section contained general questions, e.g. regarding the firm or organisation’s general reasons to collaborate on MSc theses. The second section concerned questions specific to the master project the contact person had supervised, e.g. the thesis relatedness to the type of innovation and product development process stage (based on the definitions provided above in Box 1), implementation of results and recruitment of master students. The questions used in this study are listed in Appendix A.

339 of the 529 MSc theses (64%) were classified as collaborative theses. In 315 of the theses, the firm or organisation name could be identified and 280 of the theses specified at least one contact person. Thus, in 59 of the 339 theses of the company name and/or contact person were not disclosed leaving 280 theses to base the firm survey on. Of the 280 questionnaires sent to firms/organisations and their contact person, we received 139 responses giving a total response rate of 49.6%. We used only the large firm and SME responses as the public sector and non-profit organisations’ responses were too few to use in statistical analysis, i.e. we used 124 responses in our analysis. We performed the statistical analysis using the data analysis tools in Excel.

Validity and reliability

The MSc-thesis survey contained all MSc-theses during one year, i.e. the whole population of MSc-theses in one year. Over time, MSc-theses may change in terms of collaboration frequency and type of partners, but we have no indication that the year 2016 was significantly different in this regard from previous years’ MSc-theses. Two researchers coded independently from each other the collaborative and non-collaborative MSc-theses. In only four cases (of originally 343) they differed in coding, four of the anonymous MSc-theses, resulting in classifying these four theses as non-collaborative and ending up with 339 collaborative MSc-theses. Thus, interrater reliability is quite satisfactorily for the MSc-survey.

The overall response rate for the firm survey (49.6%) is satisfactorily and similar for both groups of firms, large firms (50.6%) and SMEs (45.7%), see Table 1 below. Concerning construct validity the questions regarding reasons to collaborate we based on the study by Bramwell and Wolfe (2008) and the questions regarding the phase of development and type of development are based on previously validated concepts and scales by Cooper (2008) and OECD (2005). Overall, we view external validity for LTH as good and, for reasons elaborated at the end of the paper, for the entire higher Swedish engineering education.

Table 1. Response rates firm and organisation survey. Note that the responses from public and non-profit organisations are not included in this study.

Survey to firms and organisations	Questionnaires sent	Responses
Large firms (250 or more employees)	172	87 (50.6%)
SMEs (1–249 employees)	81	37 (45.7%)
Public (state, regional, municipal), and non-profit organisations	27	15 (55.6%)
Total	280	139 (49.6%)

Results

In the following, we will report the results from the two surveys: the MSc thesis survey and the collaborating firm survey. Starting with descriptive results from the MSc thesis survey, we below list the distribution of the 529 MSc theses at LTH according to collaboration with different types of firms and organisations (Table 2). A clear majority of MSc theses at LTH are done in collaboration with firms/organisations.

Table 2. Distribution of collaborative MSc theses according to type of organisation and region.

Type of organisation	Number of master theses in collaboration (percent of the total in collaboration)	Whereof in regional organisations (percent of collaborative theses for the type of organisation)
Public sector and non-profit organisations	28 (8%)	19 (19/28 = 68%)
SMEs (1–249 employees)	94 (28%)	67 (67/94 = 71%)
Large firms (250-or more employees)	193 (57%)	141 (141/193 = 73%)
Total nr of master theses where firm/organisation is known	315 (93%)	227 (227/315 = 72%)
Organisation anonymous	24 (7%)	–
Total number of collaborating theses (in relation to total nr of collaborative theses)	339 (100%)	–

The firms mostly involved in the collaborative theses are predominantly the large firms and especially Swedish multinational enterprises (MNEs), i.e. Swedish-based firms with international activity in at least six foreign countries. Most of these large firms have a regional presence, i.e. a head office or larger unit located in Southern Sweden, i.e. the region of Skåne.

A clear majority (72%) of MSc thesis projects collaborated with firms and organisations in the region. Knowledge spillover from educational collaboration tend to be absorbed primarily by regional firms (Bramwell and Wolfe 2008; Thune 2011), thus our survey confirms this pattern.

The distribution of MSc theses according to an engineering discipline, here we use LTH departments as an approximation of engineering discipline, show some interesting patterns (Table 3).

Table 3. Distribution of MSc theses according to LTH departments.

LTH Departments	Nr of MSC theses	Nr of collaborative MSC theses	Collaborative MSc theses (%)	(%) Large firm	(%) SME
Immunotechnology	5	0	0	0	0
Physics	30	6	20	50	17
Mathematics	17	6	35	100	0
Building & environm techn	63	32	51	56	28
Chemistry	27	14	52	86	7
Electrical and IT	46	25	54	64	28
Technology and Society	49	30	61	50	23
Chemical engineering	25	16	64	56	19
LTH Nr of MSc theses and means	529	339	64	57	28
Construction Sciences	15	10	67	90	10
Computer science	44	32	73	56	41
Biomedical engineering	37	27	73	56	30
Food technology	12	9	75	44	33
Design sciences	62	49	79	41	41
Industrial managem & logistics	51	43	84	60	21
Mechanical engineering	22	19	86	58	26
Energy sciences	24	21	88	52	33

Looking at the distribution of collaborative MSc theses, engineering disciplines with a strong basic research profile, such as Physics, Immunotechnology, and Mathematics, have a limited share of collaborative MSc theses. At the other end of the distribution curve with high shares of collaborative theses, we have more applied engineering disciplines, such as energy sciences, mechanical engineering, and industrial management. Large firm oriented disciplines tend to be the older and more established disciplines such as Construction, Electrical, Chemistry, and Mechanical engineering. SME-oriented disciplines tend to be the ‘newer’ engineering disciplines, i.e. Computer, Biomedical, Food, and Design engineering.

Looking at the heavy users of MSc theses (Table 4), defined as firms with more than 4 master theses during a year, they are all Swedish MNEs (ABB, Axis, Ericsson, IKEA, Sandvik, Scania, Sweco, Tetra Pak) or foreign-owned MNEs with Swedish subsidiaries (E.ON, WSP), except two regional firms (Kraftringen, Modelon), whereof only one firm (Modelon) is an SME. The 15 firms, heavy users, account for 31% of all collaborative theses while the remaining 69% of collaborative theses were related to 145 firms or organisations with three or fewer MSc theses. Overall, the results indicate that large firms dominate not only university-industry research collaboration (Laursen and Salter 2004) but also university-industry educational collaboration.

Table 4. Heavy users of MSc theses.

Collaborating firms/organisations involved in four or more master theses	Firm/organisation located in the same region as Lund University (regional), in rest of Sweden (national) or international	Number of master theses
ABB	Out of region, LF, Swedish MNE	4
Alfa Laval	In region, LF, Swedish MNE	4
Axis Communications	In region, LF, Swedish MNE	17
E.ON	In region, LF, Foreign MNE	8
Ericsson	In region, LF, Swedish MNE	7
IKEA	In region, LF, Swedish MNE	4
Kraftringen	In region, LF	5
Modelon	In region, SME	5
Sandvik	In region, LF, Swedish MNE	5
Scania	Out of region, LF, Swedish MNE	5
Skanska	In region, LF, Swedish MNE	6
Sweco	In region, LF, Swedish MNE	6
Tetra Pak	In region, LF, Swedish MNE	14
WSP	In region, LF, Foreign MNE	4
ÅF	In region, LF, Swedish MNE	4
Nr of ‘heavy user’ firms: 15	Nr of master theses	98

Turning to the collaborating firm survey we asked why the firm involved themselves in MSc thesis projects.

The major reasons are listed in Figure 1 below. As prior research has indicated (Bramwell and Wolfe 2008; Kunttu 2017) recruiting new employees, solving problems in the firm, get new technological knowledge and ideas and maintaining contact with the education and the school are the most important reasons. When comparing the means for large firms and SMEs with a t-test there were no statistically significant differences in the reasons to involve the firm in master thesis projects (see Table 5 below).

Figure 1. Five major reasons to collaborate according to firm responses.

Table 5. Standard t-tests of means, two-tailed, between large firms and SMEs regarding the reason for collaboration, job offers, and implementation. Statistically significant differences require a p-value at or below 0.05.

Variable	Large firms (n = 87) Mean	SMEs (n = 37) Mean	t-value	p-value
Recruit new employees	0.64	0.59	0.51	0.614
Solve problems	0.52	0.49	0.31	0.758
Knowledge new technologies	0.43	0.46	−0.23	0.819
New ideas	0.33	0.35	−0.19	0.849
Maintain contacts with school	0.29	0.24	0.54	0.586
Job offer	0.55	0.41	1.49	0.139
Results implemented	2.72	2.78	−0.24	0.808

About half of the firms also gave a job offer to the students and most students accepted the job offer (see Figure 2). Giving a job offer and students accepting the offer was relatively more frequent for the large firms than SMEs. Some 40% of the firms reported that they had implemented the results of the master thesis project. SMEs tended overall to have relatively more use for the results, though differences were not statistically significant.

Figure 2. Job offers given to thesis students and implementation of thesis results according to firm responses

Thus, MSc thesis projects are a common mechanism for both large firms and SMEs to recruit engineering students and the majority of them, at least partly, implement the results. Again, the differences in means between the two groups, large firms and SMEs, were not significant (Table 5).

Innovation-related knowledge spill was defined as four different types of innovation according to the Oslo manual by OECD (2005) based on Schumpeter’s (1942) innovation definition and six stages of the innovation process (Cooper 2008). For both type of firms product innovation-related knowledge dominates the MSc thesis projects. However, product innovation-related knowledge is relatively more frequent for SMEs than large firms (statistically significant), while process innovation is relatively more frequent for large firms than SMEs (not statistically significant, see Table 6). About 15% of the MSc theses could not be classified as related to any of the four types of innovations (Figure 3) indicating that some MSc theses have a broader focus, e.g. how an environmental law change may affect a firm’s businesses.

Figure 3. Main type of innovation related to the MSc thesis

Table 6. Standard t-tests of means, two-tailed, between large firms and SMEs regarding type of innovation and innovation process stages.

Variable	Large firms (n = 87) Mean	SMEs (n = 37) Mean	t-value	probability
Product innovation	0.44	0.65	−2.21	0.030*
Process innovation	0.32	0.19	1.61	0.111
Organisational innovation	0.01	0.05	−1.08	0.286
Marketing innovation	0.0	0.0	–	–
Screening	0.60	0.38	2.27	0.026*
Scoping	0.23	0.24	−0.16	0.875
Business concept	0.05	0.03	0.53	0.592
Prototype development	0.01	0.14	−2.12	0.040*
Prototype test	0.01	0.19	−2.68	0.010**
Commercialisation	0.02	0.03	−0.13	0.898

*Indicates significant difference at p-value at or less than 0.05.

**Indicates significant difference at p-value at or less than 0.01.

Concerning the stages in the innovation process a majority of the theses were related to the initial phases of innovation, i.e. screening new technologies, especially for large firms, or scoping new technologies to firm-specific purposes (Figure 4). The theses collaborating with SMEs were more frequently related to development and testing of prototypes. Overall, few theses concerned building a business case or development of commercialisation plans or actions. The difference between large firms and SMEs is here statistically significant, meaning that large firms tend to collaborate more on the screening of new technologies while SMEs more on later stages of the innovation process, i.e. prototype development and testing (Table 6).

Overall, the typical MSc thesis deals with new to the firm technologies related to the development of new products or new processes in the firm in the early innovation stages. On the one hand, theses collaborating with SMEs tend to be relatively more oriented towards product innovations, prototype development, and testing than large firms. On the other hand, large firms’ theses tend to concern process innovations, (though not a statistically significant difference) and screening of new technologies more frequently than SMEs.

Figure 4. Main stage in the innovation process focused in the MSc theses.

Conclusions, implications for engineering education and further research

Our surveys of MSc thesis projects at LTH have empirically shown that: (1) MSc theses are a frequently used collaboration mechanism between university and industry in engineering education, (2) collaborating firms benefit mainly by recruiting students, (3) and by the transfer of technological knowledge. Moreover, (4) mainly large firms in the region benefit from the educational collaboration and (5) SMEs tend to be connected to newer engineering disciplines. Finally, we have shown that: (6) overall firms collaborate mostly on knowledge related to product innovations and in the early stages of the product innovation process, but (7) large firms tend to screen new technologies more than SMEs, and (8) SMEs tend to have significantly more MSc thesis projects related to product innovation and in the later product innovation process stages, specifically in prototype development and testing, than large firms have.

The empirical results confirm earlier research by Bramwell and Wolfe (2008) regarding the benefits (recruitment of students and transfer of knowledge) of master thesis projects, as well as the dominance of regional diffusion of knowledge. The dominance of large firms in educational collaboration is a new result, even though the pattern has been observed before in university-industry research collaboration (Laursen and Salter 2004). In addition, the results regarding knowledge spillover related to product innovations and early product innovation stages have not been reported before as well as the SMEs’ thesis projects being significantly more focused on product innovations, prototype development and testing, while large firms’ thesis projects focus significantly more on the screening of new technologies.

Clearly, both large firms and SMEs seek to collaborate with engineering education for similar reasons and to a large degree, they also receive the benefits from the collaboration. Yet, the overwhelming dominance of regional large firms means that many SMEs might be disadvantaged in engaging in educational collaboration with engineering students (cf. Lee et al. 2010). How could engineering schools and their educational programmes provide more support to SMEs to compensate for these disadvantages? The results from the collaborating firm survey, specifically the profile of theses related to SMEs compared to large firms, i.e. mainly product innovations and prototyping support in product development, combined with Thune’s (2011) success factors for educational collaboration, forms the basis for the following five recommendations to engineering schools and their managers in order to increase knowledge spillover to SMEs.

• Engineering school management should build and maintain long term networks with SMEs making better connections and interactive relations in education with regional SMEs a prioritised objective for the school signalling this priority to staff, students and the regional industry. For an example of a well-established university-SME network see Berglund, Birgersson, and Cederfeldt (2008). This may be especially important for the older established engineering disciplines.

• Engineering school management should instruct various internal units that work as intermediaries between university and industry, such as university-industry relational offices, career centres, and labs, to increase their efforts and services for SMEs in educational collaboration. This could include the establishment and promotion of an easy-to-use Internet-based service for SMEs to register their interest for a thesis project.

• Open up labs, equipment and facilities for no or low-cost use for SMEs in collaboration with students and teachers.

• Actively market the opportunities of thesis collaboration projects to regional SMEs, possibly sending them a voucher for an MSc thesis project to be ‘cashed in’ at the career centre or some other relevant university unit. Organising a fair specifically for SMEs could be another option (Granath 2012).

• Measure the educational collaboration activities and include key metrics for SME activity and follow-up progress over time.

Our results are based on the survey of 529 MSc theses in engineering at a major Swedish engineering faculty and a questionnaire to collaborating firms and organisations. Are these results generalisable to other Swedish and European engineering schools and faculties? The Swedish higher education system is uniform in the sense that it is state regulated and for the most part state organised and funded. Moreover, there are regular national quality assessments that focus mainly on the BSc and MSc theses, which have had a homogenising effect on thesis projects and reports throughout Swedish higher education institutes (UKÄ 2015). There might be differences between the larger Swedish engineering faculties and universities that are research oriented and smaller engineering universities and faculties where education dominates. The larger institutions have a significant proportion of MSc theses that are connected to their research. In our case regarding LTH, the proportion of research connected MSc theses was 20%. Still, the number of collaborative theses are much more frequent. At smaller engineering universities and faculties, where education dominates, the incidence of collaborative MSc theses might be even higher than for the larger institutions. Nevertheless, we believe our results are generalisable to the Swedish engineering educations. The external validity of the results of engineering education in other European countries is a question for further research.

We propose the following questions for further research:

1. What is the incidence and character of collaborative MSc theses in engineering education in other European countries and how do they differ? Is there a pattern of older engineering disciplines connected to large firms and newer engineering disciplines connected to SMEs?

2. How do engineering faculties and universities organise and facilitate MSc thesis projects for SMEs? Which measures are effective to increase SME thesis project activities?

3. Given a more frequent use of the open innovation strategy (Lucia et al. 2012) in firms, how do firms organise and facilitate MSc thesis projects? How can engineering schools best facilitate firms’ open innovation strategies?

This paper has focused on the knowledge spillover from MSc thesis projects to the firms and organisations. There are also knowledge flows in the other direction, so called ‘knowledge spillins’. A final suggested research question is thus:

1. What are the knowledge spillins to an engineering faculty/engineering department/educational programme/individual teachers from collaborative MSc theses?

Finally, as mentioned before, there are a number of reasons why educational collaboration may become relatively more important than research collaboration for engineering schools. The growth of the knowledge-intensive service sector and the ICT sector in particular and their need of higher educated engineers is one such important reason (Korhonen-Yrjänheikki, Tukiainen, and Takala 2007). These sectors show a lot of entrepreneurial and SME activity, i.e. new start-ups and many small firms (OECD 2012). In order for an engineering school to support this growth, to facilitate student transfer from study life to work life as well as the transfer of innovation-related knowledge, it may be prudent to organise educational collaboration activities accordingly.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Carl-Johan Asplund is a doctoral student at the Faculty of Engineering at Lund University studying educational collaboration and interaction between university and industry. As a lecturer he has more than 30 years teaching experience in marketing, strategy and entrepreneurship courses.

Lars Bengtsson is Professor of Industrial Engineering and Management with a focus on technology strategies and business models at Faculty of Engineering LTH, Lund University. He has published some 100 books, book chapters, conference papers and journal articles on strategy, entrepreneurship, innovation, university-industry collaboration and the case study method. He has 25 years of teaching experience on strategy, entrepreneurship and innovation.

ORCID

Lars Bengtsson http://orcid.org/0000-0001-6897-327X

Additional information

Funding

This work was supported by Vinnova under [grant number 2016-05266].

Notes

1 The acronym LTH stands for the Lund University engineering faculty’s name in Swedish: Lunds Tekniska Högskola. In English, the faculty’s formal name is Faculty of Engineering LTH.

Appendix A. Questionnaire to firms and organisations

Questions (response alternatives)

1. Which are the main reasons for your company or unit to engage students in master thesis projects (choose min 1 – max 3) Choose a maximum of three for the following reasons:

   • Solve a concrete problem (yes/no)

   • Get new knowledge about a new technology (yes/no)

   • Get an independent review of (parts of) our business (yes/no)

   • Recruit a new employee (yes/no)

   • Get new ideas (yes/no)

   • Need a market or customer survey (yes/no)

   • Maintain contact with the university and the education (yes/no)

   • Do not know (yes/no)

   • Other reason. Which? (open answer)

2. What stage of a developmental process did the thesis project mainly concern?

   • Investigation of new technologies / methods / concepts (or conditions for) that the company / organisation had not previously tested and not yet decided whether to use or not. (yes/no)

   • Development / adaptation of technologies / products / methods / concepts that the company / organisation decided to use. (yes/no)

   • Development of a business case for a new technology / product / method. (yes/no)

   • Development of prototype. (yes/no)

   • Testing/validation of prototype. (yes/no)

   • Measures to commercialise and implement the technology / product / method / process. (yes/no)

   • The above phases are not applicable in the thesis project. (yes/no)

   • Do not know. (yes/no)

3. What kind of development in your company / organisation was the thesis project related to?

   • Development of a new product / service or significant improvement of existing products / services. The product or service should be new or significantly improved with respect to its capacity, ease of use, components or subsystems. (yes/no)

   • Development of a new or significantly improved process in the company / organisation. New or significantly improved methods of manufacturing goods or services, logistics, delivery or distribution methods, support activities such as maintenance systems, procurement, accounting or computer technology. (yes/no)

   • Development of the organisation, i.e. new method of organising business activities, including knowledge management, new organisation of the workplace or new organisation of external relationships not previously used by the company. (yes/no)

   • Development of marketing, i.e. significant change of aesthetic design, new product / service packing, new media and marketing methods, new product placements, sales channels, or new pricing method. (yes/no)

   • The above types of development are not applicable in the thesis project (yes/no)

   • Do not know (yes/no)

4. It is my view that the results of the maser thesis project have been used in the company or unit by influencing decisions or being implemented in the business.

5. (Strongly agree, agree, partly agree, disagree, strongly disagree, do not know)

6. We gave the master student a job offer in our company

   • Yes, and he/she accepted the job offer (yes/no)

   • Yes, but he/she declined the job offer (yes/no)

   • No (yes/no)

   • Do not know (yes/no)