Dr. Daniel McNeish, a quantitative psychologist at Arizona State University, develops accessible statistical tools to help researchers more accurately measure complex human behaviors. His work simplifies the use of advanced statistics, saving time and money while improving research accuracy. By creating free online tools and collaborating with various researchers, McNeish’s contributions benefit both the scientific community and policymakers. He plans to continue developing solutions for emerging data challenges and technologies.
Key findings:
- Statistics can help researchers accurately measure complex human behaviors
- Statistical breakthroughs can save time and money for research teams and funders
- Building freely available, online tools makes the best statistical approaches more accessible to researchers
A single number changed the trajectory of Dr. Daniel McNeish’s life. After achieving a high score on a standardized test in high school, Dr. McNeish noticed a significant shift in the way that guidance counselors and teachers viewed his potential and shaped his path to college. Recognizing how consequential numbers can be, Dr. McNeish became fascinated with the math behind the study of human behavior and currently works as a quantitative psychologist at Arizona State University. For his contributions in over 100 scientific articles and the tools that he has engineered for researchers, Dr. McNeish was nominated by the Society for Multivariate Experimental Psychology as this year’s Early Career Award recipient from the Federation of Associations in Behavioral and Brain Sciences. The Society for Multivariate Experimental Psychology held its meeting on October 9–12 in Ithaca, New York, where Dr. McNeish received his award.
All social sciences share the fundamental challenge of quantifying complex aspects of human behavior. For example, which survey questions capture the true nature of friendship, or depression, or intelligence? And do the same questions work in all cultures, for all genders, or for all ages? As a quantitative psychologist, Dr. McNeish develops statistical approaches to improve the very foundation on which all other knowledge rests.
Psychologists often measure complex behaviors by combining multiple questions into a single score and, ideally, applying statistical tests to validate that these questions accurately capture the underlying phenomenon they hope to study. In the 1990s, the correct calculations might have taken hours of math by hand or computers that were quite slow by modern standards. Today, many researchers continue to be trained to use heuristic shortcuts or workarounds, despite the rapid advances in computing power. As an alternative, Dr. McNeish has built a free, online tool that allows researchers to easily apply the optimal statistics – known as dynamic fit indices – with a few clicks of a button. As he says, “there’s always ways to make things more technical, but how do you make them more approachable?” Making small improvements to the bedrock of psychological knowledge could have widespread benefits for decisionmakers who rely on these findings to support human flourishing.
Dr. McNeish reasons that scientists, practitioners, and policymakers will earn a large return on any investment in statistical tools, especially as advances in computing rapidly erode barriers to creating and sharing accessible tools. The typical structure of academic funding involves a specialist receiving a large grant to answer a specific research question, such as a clinical psychologist running a randomized controlled trial of a new therapy. As a statistician, Dr. McNeish operates differently, regularly collaborating with researchers in different fields to develop statistical advances that help not only his current team, but also many other researchers facing similar challenges. As he says, “A million-dollar study can find policy implications for specific questions, but even $50,000 to $100,000 going to a statistician can create a method or a software that can be used by hundreds or thousands of people.” Some agencies, like the Institute for Education Sciences that has funded Dr. McNeish’s work, have developed specific grants to capitalize on the multiplicative benefits of statistical breakthroughs.
Dr. McNeish sees his role as a “statistical engineer” who works with researchers to build solutions to real-world data challenges. For example, he identified a mismatch in the advice from statisticians, which said you might need at least 30 classrooms in a school-based study to obtain accurate results, with the reality that only half that number may be feasible for researchers to recruit. “As a statistician, my role is not just telling people what they need to do, but trying to say, ‘can we find or create a method to answer your question with the data you have’” In addition to helping researchers obtain more accurate results, the solutions he discovered can save cost in future studies by reducing the amount of money needed to recruit and test many more participants.
Advances in technology have enabled exciting new forms of data collection, such as recording smartphone data or surveying participants during their daily lives. In the coming years, Dr. McNeish will continue engineering statistical solutions to keep pace with these technological advances. He also spent a recent sabbatical learning new coding skills to make his breakthroughs freely available on the internet. All of this work ensures a strong foundation for improving our understanding of human behavior.
Future Directions:
- Develop online tools to help researchers apply rigorous statistics to their data
- Collaborate to engineer new solutions to novel types of data or real-world challenges
