Posted by Ph.D. thesis “The Relationship of Proximate-Peer Characteristics of Seed Accelerator Participants to Member Firm Survival Outcomes“, Matthew Crowley
When it comes to startup success, is diversity a strength—or a silent risk? That’s the provocative question Matthew Crowley explored in his Ph.D. research on seed accelerators. And to unlock the hidden layers of gender and ethnic composition within founding teams, he turned to a hero in the tech stack: NamSor API.
🚀 The Research at a Glance
Crowley’s study examined over 1,000 startups that went through Y Combinator between 2013 and 2017. The goal? To understand how gender and ethnic diversity among founders—within firms, cohorts, and alumni networks—impacts the chances of a startup surviving five years after graduation.
But Crunchbase, the go-to startup database, had a limitation: it didn’t include detailed demographic information.
🧠 Enter NamSor: Making Invisible Data Visible
To fill this gap, the study used NamSor, a powerful name-to-demographic inference API. By analyzing the names of startup founders, NamSor estimated gender and ethnicity, enabling Crowley to calculate composite diversity scores for each startup and group.
This enriched data made it possible to:
- Quantify gender and ethnic heterogeneity within founding teams.
- Build composite diversity indexes using Blau’s index.
- Analyze diversity across three layers: firm, cohort, and alumni network.
Without NamSor, this critical dimension of analysis would have been inaccessible or prohibitively expensive to gather manually.
📊 What the Data Revealed
Surprisingly, the findings challenged simplistic notions of “diversity equals strength”:
- More diversity = more risk? Founding teams with high gender and ethnic diversity were twice as likely to shut down within five years.
- However, diverse alumni networks—though initially associated with higher closure rates—became positive influences over time, aiding survival 2.5 years after demo day.
This nuanced result underscores a deeper truth: diversity’s impact is context-dependent and evolves over time.
💡 Why This Matters for Accelerators & Startups
Crowley’s research offers practical implications:
- Accelerator design must account for team composition.
- Diverse teams might need tailored support or mentorship early on.
- Tools like NamSor can democratize access to demographic insights, enabling data-driven inclusion strategies.
🛠️ Tech for Inclusion: Beyond Just Names
NamSor’s role in this research is a great example of how AI-powered data enrichment tools can support inclusive innovation—not just through buzzwords, but with quantifiable impact.
As more startups and ecosystems strive to balance diversity with resilience, automated demographic analysis tools like NamSor could become foundational for responsible entrepreneurship research and practice.
Crowley’s study shows that diversity in startups isn’t a silver bullet—it’s a complex factor that needs thoughtful support. And thanks to NamSor, researchers can now explore these dynamics in depth, using only a list of names and some smart code.
Abstract: This study examined how composite gender and ethnic heterogeneity present at three levels of proximate-peer groups of accelerators—firms, cohorts, and alumni—relate to firm survival within five years after program graduation. This research addresses the problem that accelerator membership alone does not guarantee firm success by examining the factor of heterogeneity on survival using a fixed-design, quantitative, correlational approach via a Cox proportional hazards model. Using archival data from Crunchbase on Y Combinator cohorts running from 2013 to 2017 augmented with demographic output from NamSor, the study investigated whether diversity at various levels of proximate-peer interactions related to the hazard of firm survival. The results indicated a significant relationship between intrafirm and alumni-level composite heterogeneity and the hazard of survival. Higher levels of gender and ethnic heterogeneity within founding teams doubled the hazard of closure within five years after accelerator demo day. Similarly, increased diversity among alumni increased the hazard of closure immediately after graduation, ultimately becoming a positive force on firm survival two-and-a-half years after demo day. These findings contribute to social capital, social network, and cognitive diversity theory, highlighting the complex impact of early-firm diversity on post-accelerator outcomes and demonstrating the importance of network ties and time variance in capital and resource exchange. Finally, this study provides practical guidance to accelerator programs related to how programmatic interventions can better support diverse teams to promote long-term success.
Credits : summarization by ChatGPT, illustration by Gemini.
About NamSor
NamSor™ Applied Onomastics is a European vendor of sociolinguistics software (NamSor sorts names). NamSor mission is to help understand international flows of money, ideas and people. We proudly support Gender Gap Grader.