Advancing Entrepreneurship Theory Through Replication: A Case Study on Contemporary Methodological Challenges, Future Best Practices, and an Entreaty for Communality

Abstract

Given that replication studies are important for theory building, theory testing, knowledge accumulation, and domain legitimacy, we attempted to replicate 19 seminal studies of new venture emergence that used PSED-type data; only six attempts were successful. Our humbling experience highlights how changes at the author, journal, and institutional levels—indeed, a communal effort—can encourage, facilitate, and expedite replication studies. We provide entrepreneurship scholars with ten best practices for conducting replication studies, as well as recommendations to other stakeholders to steer away from the replication “crisis” plaguing other research domains. As they say, it takes a village.

Knowledge Accumulation in Entrepreneurship

“…A third article by Wurth et al (2022) examines the literature from an entrepreneurial ecosystems perspective. Finally, Crawford et al (2022) recount their experiences in attempting to replicate studies of new venture emergence that used PSED-type data.…”

“…Despite help in terms of data files and advice from many of the authors of the original studies, only six of the 19 studies could be exactly replicated. Crawford et al (2022) translate the challenges they faced into a set of "best practices" for facilitating study transparency and replicability and discuss the relevance of such practices for research stakeholders (i.e., authors, editors, journals, academic institutions, and the scholarly field at large). Specifically, they stress the need for published papers-especially those using publicly available datasets-to provide precise documentation of the selection and construction of variables extracted from the original data.…”

“…In conducting, documenting, analyzing, and discussing their 19 replication attempts, Crawford et al (2022) propose several steps authors should take that will facilitate the replication of their work. As such, their paper serves as a guide for scholars and other stakeholders in entrepreneurship research (e.g., authors, editors, journals, or academic institutions) on how studies can be made more transparent and hence more easily analyzed and replicated, for further knowledge accumulation across a variety of domains.…”

“…Crawford et al (2022) contribute a methodological note to this issue by describing the challenges they encountered when trying to replicate published papers in entrepreneurship research. They argue for the importance of knowledge advancement and accumulation through constructive replication, where specific features of a published study are altered, while maintaining the original research question and setting (e.g., Markin et al, 2021).…”

“…Given the extensive amount of variable manipulation that is often necessary to achieve specific research purposes, they also note that published studies need to document how outliers were handled and how/why cases were excluded from the analyses. Since reporting these decisions and practices are expected, the information lacking from published and often widely cited papers noted by Crawford et al (2022) highlights the need for research stakeholders to improve the documentation and storage of datasets and analytical syntax used to create, manipulate, and analyze the data (Chanda & Miller, 2019). Many journals today require or encourage the submission of datasets and analytical material and given the ample opportunities to publish online appendices with published studies, as well as links to free open repositories for datasets and analytical codes (e.g., Zenodo, Open Science Framework, ResearchGate, or data repositories run by publishers), such procedures could be pivotal for further knowledge accumulation in entrepreneurship research.…”