Ansprechpartner: Dipl.-Kfm. Sean Patrick Saßmannshausen, Research-Assistenten: Daniela Putsch & Tim Seidel

Introduction

In 1985, Sue Birley published one of the very first reviewed papers (if not the first at all) dedicated to the role of networks and social capital in the field of SME and entrepreneurship. The first reviewed paper was published in 1986. The number of following publications related to this subject matter increased almost year by year. Meanwhile, the number of reviewed papers has increased dramatically. Most of them combine theoretical assumptions and empirical tests of hypotheses. In addition, from the beginning of the early 1990^th more and more attention is drawn to the closely related field of “social capital”. We reviewed 66 papers on networks and SME respectively entrepreneurship. Papers added to our research were published from 1985 to 2006 and have undergone a double blind review process before publication. Our two main objectives have been to draw the current “summa theological” or “state of the art” in SME’s network research on the one hand and—on the other hand—to stress out, which advances have to be taken in order to help research any further. According to the conferences track, our paper focuses on this second field of interest.

Methodology

We conduct a literature review using narrative meta analyzes of 66 reviewed papers on network and SMEs. We enumerate, point out and compare (1) theoretical sounding, (2) hypotheses, (3) methodology, (4) data basis and (5) findings of each paper to give an overview on both, advances and lacks. We furthermore track for citation biases. Thereby, we can show and explain heterogeneity in findings according some common popular hypothesis. In case of homogeneous findings across papers, we can show that homogeneity of findings might in some cases not be due to a reflection of real world. Instead it seems to be caused by poor theoretical reasoning or a lack of appropriate method. In theoretical and methodological respect, we make some suggestions on how to solve such problems.

Results

In quite some papers, a repeated examination of common research patterns can be observed. Firstly, those patterns cover certain research questions like the following: Are strong or weak ties more important to the entrepreneurial process? Can the entrepreneurial network success hypothesis be tested positively? Is the social network of any influence on (1) the tendency to become an entrepreneur, (2) the opportunity recognition, or (3) the exploitation ability? What leverage effects can be measured regarding the use of social capital? Secondly, those patterns do also cover certain special research areas like the role of networks or of social capital in the gender or emigrant context, as well as networks and innovation, or networks and finance respectively venture capital. It can be shown that something like research redundancy has established concerning both, topics and methods, probably caused through scholarship, or inspiration of older papers, or expectations on how to be accepted by a journal’s editorial board.

Implications

In order to overcome redundancy, we suggest both: methodological and theoretical changes in future research on entrepreneurial networks. An analyzes of the many papers published shows the establishment of an research mainstream which is primarily repeating quantitative tests of the same hypothesis again and again, sometimes—so to speak—changed into a new suit. E.g. the differences in support from strong or weak ties have been tested over and over again using similar methods, without showing significant differences in result (what is a result itself). If research is supposed to develop, new hypotheses have to be developed and new categories need to be invented. The invention of new research categories could be achieved by multi-dimensional recombination of well known categories. For example the dimension “strength of ties” could be accomplished by the dimension of trust and dynamic aspects (time). The resulting matrix would offer at least eight fields in which ties could be classified, instead of the two fields “strong” and “weak” which are commonly in use. As far as hypotheses are concerned, in spite of deductive approach, we suggest first to revisit economical theory to derive hypotheses from. To give just one example, Hayek’s theory of group selection could be fruitful to explain the differences in network success across different networks. Current network research can explain why firms with bigger networks might be more successful, but it fails to explain, why the results in case of similar networks differ, even under similar outer conditions (ceteris paribus). Hayek’s theory could teach researchers to emphasize more on network qualities than just on structure, such like the often-measured network density, diversity and size. This implicates changes in empirical strategies and research design. Last but not least we recommend more creativeness and audacity when choosing research topic or setting up methodology. Popper’s definition of scientific evolution lives not only on examination of the same questions with similar methods, but from asking new questions.