The Board of the Global Entrepreneurship Research Association (GERA), which governs the Global Entrepreneurship Monitor (GEM) project, identified at its meeting on 13th - 14th June 2016 the need to address confusion between GEM and the Global Entrepreneurship Development Institute (GEDI), and their product the Global Entrepreneurship Index (GEI).
This position statement is intended to clarify to members and users of GEM that there is no formal relationship between GEM and GEDI. While the GEI was developed by members of a GEM National Team, there is not and has never been a formal relationship between GEM and GEDI. The GERA Board has concerns about the methodology behind the GEI and does not endorse the results derived from this methodology.
Background: the GEM project
The GEM project was initiated in 1997 by a team of distinguished researchers from Babson College and London Business School. Their initiative responded to a recognized deficit in research activity focusing on the entrepreneurial phenomenon and the lack of evidence-based policymaking.
To date, this pioneering combination of research with policy implications has been conducted in over 100 economies; nearly 500 researchers from around the world are currently involved in what is the biggest entrepreneurial venture in the social sciences.
On an annual basis, GEM members collect a wealth of primary data on entrepreneurship in accordance with a robust conceptual framework, making use of established survey tools and a rigorous research methodology to ensure the highest quality data possible. Analysis of the resulting data leads to the production of Global, National and Special Topic reports presenting a portfolio of indicators.
The project requires the commitment of hundreds of researchers and academic institutions, in addition to millions of dollars of funding support from a large number of governments, development agencies, banks, universities and business associations. This enormous intellectual and financial investment allows the production of the most comprehensive database on entrepreneurship in the world, providing highly valuable primary data on the entrepreneurial intentions, attributes and activities of individuals as well as the ecosystems in which entrepreneurship takes place.
GERA provides open access to this data, a vital public service which benefits the entire research community, especially junior researchers, educators who want to develop research-based educational programs and policy makers interested in designing evidence-based policies.
GEM is a not-for-profit joint effort for the public good. Its conceptual framework, survey tools and research methodology are constantly scrutinised, developed and improved by the GEM community, and remain the collective intellectual property of GERA and the GEM National Teams. No individual member, including the GEM founders, has ever asserted individual property rights over these intellectual contributions.
GEDI and the Global Entrepreneurship Index
The GEI was developed by members of a GEM National Team outside of their work for GEM. It was initially put forward by these members as a progression of the GEM conceptual framework, but subsequently registered as an independent organisation (GEDI). The GEI is calculated using data sourced from GEM, for example the 2016 GEI uses a total of 31 indicators, of which 16 are sourced from GEM.
GEM accepts the risks associated with providing open access to its database. It provides ample guidance on the appropriate use of its data on its website and in many academic papers, and accepts that its primary data will be used and interpreted by third parties in their own analysis.
However, GERA has become aware of confusion surrounding the relationship between GEM and GEDI. Given the fact that the GEI currently appears to rely heavily on GEM data, the GERA Board wishes to make known that although the GEI uses data from GEM, it is not endorsed by GERA.
GERA also has concerns about the assumptions made in the creation of GEI, including the following:
Assumptions by Geographical Proximity - using data available for selected neighbouring economies to generate data where none exists.
The 2016 GEI report features 132 economies. In 34 of these economies, GEM data has never been collected. The GEM indicators used in each of these economies were generated by assuming that a neighbouring economy or economies was similar. Following are some examples of this concern:
- GEM data for Malaysia and Singapore is averaged to generate data for Brunei Darussalam;
- GEM data for Qatar is assumed to be representative of Bahrain, Kuwait and Oman;
- GEM data for Burkina Faso, Cameroon and Uganda is averaged to generate data for 15 different sub-Saharan African countries.
GERA is concerned that this could cause issues if policy makers base decisions on GEI results in one of these economies, if academics publish research with data from these economies, or if an organization re-publishes this data as representative of these economies.
Time Series Assumptions - using GEM variables from a wide time window, combining it with non-GEM variables from different years, and presenting it as current. In this way:
- The 2016 GEI for Dominican Republic, Hong Kong, Jordan, Lebanon, Morocco and Serbia is based on GEM data collected in 2009 and non-GEM data collected in 2011, 2012, 2013 and 2014;
- The 2016 GEI for Iceland, Montenegro and Saudi Arabia is based on GEM data collected in 2010 and non-GEM data collected in 2011, 2012, 2013 and 2014;
- Most other 2016 GEI data is based on averages of GEM data collected in 2013 and 2014.
Many other examples of these practices can be found in table 5.2, pp 75-78 of the 2016 Global Entrepreneurship Index.
The GERA Board welcomes the use of GEM data to disseminate knowledge about entrepreneurship, to generate insights that extend our understanding of this phenomenon, and to inform decisions and actions that can enhance the quality of entrepreneurship and its impact on society.
The GERA Board cannot support or endorse the usage of GEM data by the GEDI. GEM is not, and has never been, affiliated with the GEDI.
If members or users have any questions they should contact Mike Herrington (firstname.lastname@example.org) who will be happy to assist.