Davood Tofighi, Ph.D. will present Improving Validity and Replicability of Longitudinal Mediation Models in Psychological and Health Sciences.

Abstract:  In this talk, I present an overview of my research. Then I focus on one area of my research that concerns with developing methods to strengthen internal validity and robustness of mediation analysis with application in intervention studies of alcohol use disorder. In any mediation, a critical yet untestable assumption for valid and unbiased estimates of the indirect effects is that there should not be an omitted variable (confounder) that confounds (biases) indirect effect estimates. Because the no–omitted–confounder assumption is untestable, the indirect effect is potentially biased. Researchers recommend conducting sensitivity analysis to quantify bias and assess whether the point and interval estimates of indirect effects would change under varying degrees of bias because of an omitted confounder. I present the latest sensitivity analysis techniques I have developed, termed Correlated Augmented Model Sensitivity Analysis (CAMSA). I discuss how one can use CAMSA to compute biasing effects of confounders on the point and interval estimates of the indirect effects for a longitudinal mediation analysis using an empirical example.

Biography:  Davood Tofighi, Ph.D., is an Assistant Professor of Psychology at the University of New Mexico and a Biostatistician at the Clinical and Translational Science Center in the UNM Health Sciences Center. His research involves developing and evaluating methods that would enhance the robustness and causal inference in mediation analysis, multilevel regression, and structural equation modeling. His substantive research focuses on improving the robustness and replication of statistical methods in health and prevention science and statistical optimization of drug delivery formulation in cancer and COVID-19 research. Currently, he serves as an Associate Editor for the Quantitative Psychology and Measurement specialty section of the Frontiers in Psychology journal. He has served multiple times as a scientific reviewer for NIH Special Emphasis panels. He has over 6,000 citations and an H-Index of 12 according to Google Scholar. His ongoing funding is $4,708,813 and his research has been supported by NASA, NSF, and NIH. He received his Ph.D. in Quantitative Psychology from the Arizona State University. He has an M.A. in Educational Psychology, and M.Sc. and B.Sc. in Industrial and Systems Engineering. 

Zoom link and information:

Link: https://unm.zoom.us/j/93318869319
Meeting ID: 933 1886 9319
Passcode: loganhall