Brett Welch

Hello! I'm Dr. Brett Welch and I am an Assistant Professor at the University of Houston.

I earned a BA in Linguistics and BS in Speech-Language Pathology from the University of Texas at Austin. I went on to earn my MS in Communication Science and Disorders (CSD) from the University of Texas Rio Grande Valley before completing my clinical fellowship. After working in a hospital and starting my own private practice, I began working on my PhD in CSD from the University of Pittsburgh under the guidance of Dr. Leah B. Helou. After earning my PhD, I worked with Dr. Jeffrey Girard as a postdoctoral researcher in the Affective Communication & Computing Lab in the Department of Psychology at the University of Kansas.

Clinically, I specialize in working with the psychological aspects of voice and voice disorders – identity, personality, and trauma routinely come up in my sessions. My line of research stems directly from this clinical background and blends gold-standard clinical voice and speech assessment with rigorous analytic techniques from Quantitative Psychology to interrogate these psycho-social-vocal relationships. My dissertation project integrated contemporary theories and methods used in Personality Psychology to advance our understanding of chronic vocal hyperfunction.

My postdoctoral training further focused on additional statistical modeling techniques, with an emphasis on Bayesian hierarchical linear modeling and dynamic structural equation modeling. Working with my postdoctoral mentor, we are examining how voice and speech behaviors differ in individuals diagnosed with post-traumatic stress disorder (PTSD) and other internalizing psychopathologies (e.g., anxiety, depression). This line of work advances our ability to develop a low-cost, easily deployable screener that can be used with high-risk populations (e.g., military veterans) or in areas with limited access to mental health resources. Additionally, by better understanding how anxiety and depression impact communication behaviors, clinicians will be better able to improve assessment and track treatment progress using objective indicators of mental states.

In addition to pursuing my own research, in my current role I also collaborate with other faculty members and students as a statistical consultant and co-investigator. In this capacity, I also teach statistics and research methods for the PhD students in the Department. I am passionate about mentoring students and helping make statistics more accessible and less scary. I believe that anyone can develop their quantitative skills, and that by combining a solid foundation of key statistical concepts with scientific computing skills, future scientists will be well-poised to rigorously advance their respective areas of research and/or be well-equipped to succeed in industry.

I have gained many data analytic skills during my doctoral and postdoctoral training. I have become extremely proficient using R to clean, organize, and analyze data for projects. I have also taken courses in computer science and have a working proficiency with SQL and Python. Most of my quantitative skillset has focused on multivariate methods (e.g., factor analysis and structural equation modeling), hierarchical/mixed-effects models, Bayesian methods, and emerging skills with dynamic structural equation modeling for intensive longitudinal data. In addition to continuing to expand my own quantitative skillset, I am passionate about mentoring and teaching students to help make statistics more accessible and less scary.