Udi Alter

Dr. Alter’s research seeks to address common challenges psychology researchers often face. As data and measures become more complex, many conventional statistical techniques do not always handle that complexity well. His research examines novel approaches that account for itand psychometric methods that improve how we measure constructs that we cannot directly observe such as attitudes toward statistics. Consider a concrete example. The more significance tests (e.g., t tests) researchers run, the more likely they are to find a significant result by chance alone (i.e., a Type I error), even when a real effect does not exist. This is the multiplicity, or multiple testing, problem. Of course, some solutions do exist (e.g., the Bonferroni procedure, which adjusts the significance threshold for each test), but they often come with their own set of problems including reduced power to detect true effects. One of Dr. Alter’s projects therefore approaches the multiplicity problem through multilevel modeling and Bayesian statistics. Developing better methods, however, has little impact if they never reach the researchers they were built for. Much of Dr. Alter’s work therefore focuses on knowledge mobilization: open-access papers, free software (e.g., R packages, Shiny apps) that makes modern methods accessible and research on the barriers and incentives that shape how scientists and students adopt better statistical and methodological practices. A further line of research examines how students learn statistics and statistical software: what makes these a source of anxiety for many and which teaching approaches help. This includes topics such as the choice between point-and-click software (e.g., SPSS) and syntax-based languages (e.g., R, Python) and the emerging role of AI tools in learning to program. | Quantitative methods; statistics; equivalence/negligible effect testing; multilevel modelling; Bayesian statistics; the multiplicity problem; item response theory; open science; statistics education; statistical software

Udi Alter is Assistant Professor | Cognition and Neuroscience Research Group Member at MacEwan University. Their stated interests include Dr. Alter’s research seeks to address common challenges psychology researchers often face. As data and measures become more complex, many conventional statistical techniques do not always handle that complexity well. His research examines novel approaches that account for itand psychometric methods that improve how we measure constructs that we cannot directly observe such as attitudes toward statistics. Consider a concrete example. The more significance tests (e.g., t tests) researchers run, the more likely they are to find a significant result by chance alone (i.e., a Type I error), even when a real effect does not exist. This is the multiplicity, or multiple testing, problem. Of course, s. This summary considered all 2 collected sources.