Stats bootcamp is back to answer your statistics questions, including the ones professional training forgot to tell you about. The annual bootcamp from the University of North Carolina (UNC) National Institute of Arthritis and Musculoskeletal and Skin Disease-funded Core Center for Clinical Research (CCCR), housed in the UNC Thurston Arthritis Research Center (TARC), has been a popular program at ACR Convergence for more than a decade.
“Theoretically, ACR/ARP members are exposed to statistics in medical and graduate school, although in reality they probably get, at most, an introduction to the subject,” said Becki Cleveland, PhD, Director of Biostatical Operations at the North Carolina Translational and Clinical Sciences Institute. “Clinicians and fellows and researchers are reading the literature and need to know what the statistical aspects of these projects really mean, whether a study was done properly based on the research question, and whether the correct methods were used. These issues are important in rheumatology and, unfortunately, many may not receive enough statistical training in medical school to be able to make these assessments.”
Dr. Cleveland will open Stats Bootcamp 1: How to Choose the Right Statistical Analysis for Your Research on Sunday, October 26, from 2:45–3:45 p.m. in Room W184B-C of McCormick Place, with a look at how common study designs and different types of variables can help investigators develop useful research questions. The goal isn’t to leave researchers ready to do their own statistical analysis, but with an understanding of which analytical approaches are most appropriate for which kinds of research questions.
“Your research question should lead you into thinking of what the most appropriate analytic approach is to answer your question,” explained Carolina Alvarez, MS, TARC Biostatistician. “The variables you want to study can help guide you to the most appropriate statistical approach to best fit your question.”
The Bootcamp program is designed for both producers and consumers of research, rheumatologists who design and conduct studies, as well as rheumatologists who read and incorporate the results into clinical practice.
“It is important for clinicians to understand these methods, even though they may never use them themselves,” explained Todd Schwartz, DrPH, Professor of Biostatistics at the UNC Gillings School of Global Public Health and Co-Director of the CCCR Methodology Core. “Clinicians are reading papers as the field is being advanced, and it helps to have some idea of the statistical methods involved, things such as latent variables, to advance their knowledge.”
Not sure what a latent variable is? Dr. Schwartz will introduce the concept during Stats Bootcamp 2: Methods for Analyzing Groupings That Cannot Be Directly Observed and Measured: An Introduction to Latent Structures and Latent Variable Models on Monday, October 27, from 4–5 p.m. in Room W184B-C of McCormick Place.
Some measures, such as body mass and blood pressure, are objective and can be measured directly, explained Liubov Arbeeva, MSc, TARC Biostatistician. However, other important concepts like anxiety and pain severity, which matter to both patients and physicians, often cannot be directly observed, i.e., they are latent. These latent variables are typically assessed through more subjective self-reported measures and questionnaires.
Some of these instruments, for example, the Pain Catastrophizing Scale, were developed and validated using factor analysis — a topic that will be discussed during the session. However, self-reported measures, and even lab tests, are prone to measurement errors, natural fluctuations, and over- and under-reporting.
“Modeling unobserved or latent variables is a valuable approach to mitigate misinterpretations that may arise if these issues are overlooked,” Ms. Arbeeva said. “We want readers to feel confident when encountering terms like ‘latent variable’ in research papers and will provide a non-technical overview of the methods contained under the latent structure analysis umbrella.”
On-demand access to recorded presentations from each session will be available to registered participants of ACR Convergence following the annual meeting through October 31, 2026.
Don’t Miss a Session
If you weren’t able to make it to a live session during ACR Convergence 2025 — or you want to revisit a session from the annual meeting — make plans to watch the replay. All registered participants receive on-demand access to scientific sessions after the meeting through October 31, 2026.