Big Data Directions in Entrepreneurship Research: Researcher Viewpoints

Extracting Meaning from Data

Travis Howe, Ewing Marion Kauffman Foundation

Our world is becoming increasing Funesian in that we are perceiving and storing more and more information in the form of data. But, as with Funes, access to information is not the same as understanding. Are we also better at extracting meaning from all of this data? What does understanding rely on – is it only possible through sophisticated data-processing techniques or is something else required? This paper will briefly discuss three common pitfalls related to the challenge of extracting meaning from data.

Visual Analytics for Entrepreneurship Research

Rahul C. Basole, Accenture AI

Visual analytics, which fuses data visualization with analytical models, is a promising emerging methodological approach in the enterprise sciences that aims to alleviate these issues. Broadly considered, visual analytics enables scholars and practitioners to more rapidly digest data, see patterns, spot trends, and identify outliers, thereby improving comprehension, memory, and decision making. It also facilitates the proposition and hypothesis generating process and ultimately accelerates time-to-insight. In addition to augmenting human intelligence, visualizations also aid in communication and explanation of complex phenomena, the critical last mile in data-driven research. The impact of visual analytics can thus be quite substantial for both entrepreneurship research and practice.

Computational Modeling Approach to Understanding Entrepreneurial Ecosystems

Yushim Kim, Arizona State University

The notion of an “ecosystem for entrepreneurship” has emerged as a promising conceptual framework because the decision to engage in entrepreneurship is made as entrepreneurs identify, interpret, and act upon opportunities that are embedded in a system. The entrepreneurial ecosystem (hereafter, EE) involves entrepreneurs as well as other critical actors, such as financial firms, universities, and public organizations that support new and growing firms. The concept also includes the entrepreneurial processes and institutional constraints that are interlaced together. The EE has been described as a system of “dynamic local, social, institutional, and cultural processes and actors that encourage and enhance new firm formation and growth”.

The literature does not give clear answers to critical questions, such as how institutional and individual actors are interlaced together and which conditions are necessary to ensure vibrant and sustainable innovation systems. The understanding of the delicate and complex relations among various ingredients that enable successful EEs may not easily develop without the aid of a suitable framework and tools. This thought piece introduces an approach that can be fruitful in exploring this challenge and such questions.

Measuring Digital Entrepreneurship at the Grassroots: What Role Will It Play in Community Resilience?

Karen Mossberger, Arizona State University; Caroline Tolbert, University of Iowa; Scott LaCombe, Smith College

The landscape of entrepreneurship has been changing because of digital technologies, and we have lacked the tools to adequately capture these trends and their impacts on communities. This is especially true for measuring the development and effects of microenterprises and start-ups that are too small or too new to be counted in traditional sources such as government data on small businesses.

Yet the internet has also led to experimentation with large, new datasets – often referred to as “big data” – to generate new policy-relevant insights. The analysis of such data has proliferated with increased computational power and the use of machine learning and algorithms.

We present new data on the density of active domain name websites in communities as a measure of local economic activity. GoDaddy, which is the world’s largest registrar of domain names, has collaborated with researchers from University of Iowa and Arizona State University, sharing de-identified data on the 20 million active U.S. domain name websites that have traffic and services attached to them. At least three-quarters of these websites are commercial.