Unlocking the World of Statistical Modeling: A Workshop on “Mastering Statistical Modeling in R/Stata for Scientific Publications in the Age of ChatGPT: Uses, Abuses and Real-World Applications"
Today's workshop at the South Asian Institute for Social Transformation (SAIST) was nothing short of fascinating! Instructed by the talented mentor, Juwel Rana sir, from McGill University, Canada, the session went deep into the field of statistical modeling and research methodology.
I had the opportunity to get into a lot of research topics, including from observational studies, cross-sectional studies to cohort studies. Additionally, I gained valuable insights into various regression analysis techniques such as GLM, linear regression, binomial regression, logistics regression and poisson regression. Understanding the specifics of logit for logistics (odds) and log for binomial (probability) was particularly interesting. I learned that for logarithmic measures, we can employ exponential functions. Moreover, I understood the importance of selecting the appropriate regression model based on the rarity of the outcome. For rare outcomes, logistic regression is preferred, while for non-rare outcomes, log binomial/binomial regression is more suitable. In cases where logistic regression or binomial regression is not applicable, modified or robust poisson regression appears as an alternative.
A significant takeaway was the importance on research design and its pivotal role in ensuring the quality of research. Learning about randomized controlled trials (RCTs) and the efficiency they bring to research methodologies was particularly enlightening. Understanding various research models such as causal, predictive and associational further expanded my perspective. I was reminded of the fundamental components of research: formulating a clear research question, designing robust methodologies, conducting accurate data analysis and interpreting results accurately.
Throughout the session, Sir Juwel Rana's lectures were not only informative but also creatively delivered, making complex concepts easy to understand.
Furthermore, I had the opportunity to learn how to implement these models in R and STATA, equipping me with practical skills for data analysis. Additionally, understanding the ethical use of ChatGPT for data analysis added another dimension to my learning experience.
I extend my heartfelt appreciation to the South Asian Institute for Social Transformation (SAIST) for organizing such an insightful and enriching workshop. I'm truly grateful for the opportunity to participate.
