Every OSE diagnostic combines big data, network theory, and 10+ years of longitudinal observation — across 5 theoretical layers, 14 quality indicators, and a multi-awarded scientific council. This page documents the science behind every score.
Each layer captures a distinct dimension of capacity. Together they form the OSE Quality Score.
Resources, infrastructure, hard support
The tangible substrate of an ecosystem: workspaces, equipment, technical mentorship, MVPs labs, prototyping facilities. The Resource-Based View (Barney 1991) frames this as the stock of physical and technological resources.
Theoretical roots: RBV (Barney 1991) · Penrose (1959)
Knowledge, training, mentoring
The diffusion and absorption capacity of know-how across an ecosystem. Includes formal training programs, executive education, technical bootcamps, and the network of mentors and advisors. Drives entrepreneurial competence accumulation.
Theoretical roots: Cohen & Levinthal (1990) · Davidsson & Honig (2003)
Networks, social capital, trust
The web of connections between actors — the “missing link” of public policy. Measured via network analysis: density, centrality, communities (Louvain), modularity. Granovetter's weak-tie thesis applies: bridges between communities matter more than internal density.
Theoretical roots: Granovetter (1973) · Burt (1992) · Stam (2015)
Institutions, legitimacy, culture
The institutional architecture: laws, norms, public discourse, recognition signals. Captures whether entrepreneurship is celebrated or marginalized. Includes regulatory friction, brand recognition of major actors, presence of awards/rankings/media coverage.
Theoretical roots: Scott (2014) · DiMaggio & Powell (1983) · Kibler et al. (2014)
Mindset, agency, entrepreneurial intent
The aggregate cognitive disposition of a population towards entrepreneurship. Risk tolerance, opportunity recognition, self-efficacy, fear of failure. Driven by exposure (role models, success stories) and education. The OSE’s most distinctive layer — most ecosystem indices ignore it entirely.
Theoretical roots: Krueger (2007) · Bandura (1997) · Kautonen et al. (2015)
Each actor receives a per-layer score and a global OSE Quality Score. Aggregation at the country level produces the values you see on Expert.
Full mathematical definitions in our methodology paper — available on /research.
Centrality (degree, betweenness, eigenvector), modularity, clustering coefficient, robustness scores.
Louvain method to identify sub-communities and cross-community bridges (weak ties).
Natural extension of RBV to relational resources — captures who orchestrates flow vs who hoards.
Treats the ecosystem as a meta-organization (Berkowitz & Bor 2018) — coherence, governance, mission alignment.
Time series of every metric across 10+ years — identifies disruptions, accelerations, regime changes.
Continent + income tier (LIC/LMC/UMC/HIC) baselines + Spearman correlations with WorldBank macro indicators.
| Traditional approach | OSE approach |
|---|---|
| Self-reported surveys | Big Data + qualification |
| N=200 sample, biased coverage | N=16,000+, exhaustive |
| Static snapshot, 1 year | 10+ years time series |
| Isolated actors | Network analysis |
| Macro-economics (GDP, FDI) | Meso-analysis (value chain) |
| Anglo-centric indicators | Africa/MENA-grounded indicators |
Curated insights from our 117 ecosystems — new country reports, ranking shifts, methodology notes. No fluff, no spam, unsubscribe in 1 click.
By subscribing you agree to our privacy policy. We never share your data.
