Bayesian and Longitudinal Data Science Lab

(BAL LAB)

Welcome!

Modeling longitudinal data is one of the most active areas of research in social, behavioral, and education sciences because only longitudinal research can provide valuable insights into change and causal relationships, which are inherently linked to policy and decision making. Through longitudinal data, researchers can investigate individual development and the difference in the development among individuals. In addition, researchers can study issues about population heterogeneity, important factors that affect the change trajectories, effective interventions, interrelationships among variables, missing data, etc. 

To draw reliable conclusions with a high external validity, researchers often collect and work with large-scale datasets with complex structures and even untraditional formats. Therefore, many analytical challenges have emerged. Without fully taking the complex data structure into account, longitudinal analyses may lead to biased results and incorrect statistical inferences.

Bayesian methods for complex data analysis has gained increasing popularity, mainly because Bayesian statistics has many advantages over traditional frequentist (i.e., non-Bayesian) statistics. For example, they are powerful to estimate complex longitudinal models, can capture the uncertainty of hypothesis, model, and/or parameters through prior and posterior distributions, and handle nonnormal data and missing data relatively easily.

The Bayesian and Longitudinal Data Science Lab (BAL LAB) is active in developing and refining Bayesian methods and longitudinal data analytical methods. We are interested in address challenges including analyzing nonnormal data, missing data, untraditional data formats, etc. We aim to contribute to fundamental science research as well as to increase the scope of application of statistical analysis to answer important questions about the nature of psychological processes and the factors that shape human behavior. 

Our long-term goal is to reveal the limitation of current models and methods in many disciplines, improve the understanding of scholars of how different data structures lead to different analytical models, broaden the knowledge of intrinsic relationships among meaningful constructs, and raise awareness the critical importance of appropriate data analyses in approaching data collected in social, behavioral, and education sciences.

BAL LAB offers comprehensive training in quantitative methods. We are committed to fostering an academically enriching and supportive climate for all members of our community. We hold the mission of promoting diversity and inclusion, and deeply value the beauty of BALance and symmetry.

This institution will be based on the illimitable freedom of the human mind. For here we are not afraid to follow truth wherever it may lead, nor to tolerate any error so long as reason is left free to combat it.  

– Thomas Jefferson, December 27, 1820