Abstract

Starting in 2012, the LISA 2020 Global Network has grown from two stat labs in developing countries in 2014 to 28 in 2020 and now 36 in 2024 (see chapter 2 of our book for more details). The LISA 2020 Global Network builds partnerships among the University of Colorado Boulder Laboratory for Interdisciplinary Statistical Analysis (LISA) and universities in developing countries to create stat labs that simultaneously develop capacity in statistics and data science and leverage it to solve local development challenges. Shared learning from the stat labs strengthen and help grow the LISA 2020 Global Network. Each stat lab acts as an engine for development by progressing toward three missions:
1. Train statisticians and data scientists to become effective, interdisciplinary collaborators who can move between theory and practice to solve problems for real-world impact;
2. Serve as research infrastructure for researchers and decision-makers to collaborate with statisticians and data scientists to enable and accelerate research and data-based decisions that make a positive impact on society;
3. Teach short courses and workshops to improve statistical skills and data literacy widely.

Between 2019 and 2023, the 36 new LISA 2020 labs reported training 3,625 students to become collaborative statisticians and data scientists. The labs worked on more than 1,827 projects resulting in at least 281 peer-reviewed publications that included stat lab faculty and students as co-authors. These labs taught 654 short courses, workshops, or multi-day workshops taught to 18,868 attendees. So far the Network has published two books: Promoting Statistical Practice and Collaboration in Developing Countries, published in 2022 by Taylor & Francis/CRC and Sustainable Statistical and Data Science Methods and Practices published in 2023 by Springer. A third book, Practical Statistical Learning and Data Science Methods, is in press to be published by Springer later in 2024.

This paper is about the various ways the LISA 2020 labs are educating and training students to become effective interdisciplinary collaborations. We conclude with a list of recommended best practices.