Early Career Impact Awardee – Society of Multivariate Experimental Psychology
Dr. McNeish is a quantitative psychologist whose interest is creating, evaluating, and applying sophisticated statistical methods relevant to behavioral research. He earned his PhD in Measurement & Statistics at the University of Maryland in 2015 and is currently a Full Professor in the Quantitative Psychology program at Arizona State University.
Although quantitative methods research may not always have direct policy implications, the indirect impact can be considerable. Advanced methods unlock insights into behavioral phenomena and outreach efforts can clarify best practices for data analysis to ensure that research with more direct policy questions is methodologically sound.
To this end, Dr. McNeish aims to serve as a bridge between statisticians and behavioral researchers by (a) developing methodological solutions and software that are broadly accessible to both statisticians and psychologists and (b) clarifying best practices for applications of sophisticated quantitative methods to behavioral data.
Dr. McNeish’s research has two primary streams, (a) multilevel, longitudinal, and time-series models and (b) measurement and latent variable models. He also frequently collaborates on research in psychology, public health, education, and psychiatry to strengthen empirical data analyses in projects funded by the US Department of Education, the US Naval Research Lab, NIDA, NIMH, NIMHD, NIAA, NICHD, and the Wellcome Trust.
His recent work has focused on two broad areas, (a) statistically evaluating the quality of measurement scales and (b) statistical methods for intensive longitudinal data that have gained in popularity to accommodate data collected from smartphones and wearable devices.
For evaluating measurement scales, Dr. McNeish created a computational method called dynamic fit index cutoffs that produces custom-tailored cutoffs for evaluating evidence that a scale is plausibly measures its intended construct. Open-source software and a Shiny application (a point-and-click version of an R package that runs over a website; www.dynamicfit.app) have been created to make the method accessible and facilitate implementation.
Within intensive longitudinal data, Dr. McNeish has worked on the dynamic structural equation modeling framework to integrate time-series analysis, multilevel models, and structural equation models for accommodating features of new data structures emanating from smartphone-collected data. He and his students have developed methods in this area for binary and categorical outcomes, missing not at random data, modeling a variance as an outcome, and modeling data from multiple people simultaneously (e.g., romantic partners; mothers and children).
Dr. McNeish’s broader body of work has previously been acknowledged with the APA’s Distinguished Scientific Award for Early Career Contributions in 2023, selection as a Highly Cited Researcher in Psychology/Psychiatry by Web of Science in 2022 and 2023, and election in the 65-member Society for Multivariate Experimental Psychology in 2018.
He has previously received early career awards from APA Division 5 (Quantitative Methods) in 2019, AERA Division D (Statistics) in 2019, the Society for Multivariate Experimental Psychology in 2020, and the Rising Star designation from APS in 2018. He received the APA Division 5 Anastasi Dissertation Award in 2018.
