Matthew Sigal, a new lecturer in the Department of Psychology, comes to SFU with a PhD in Quantitative Methods (QM) from York University. He describes QM as “bridging the gaps between statistical theory, research design and methodology, and their applied aspects, such as client consultation and data interpretation.”

Since arriving at SFU, Sigal has been teaching courses in research methods and data analysis for studying psychology. With an interest in data visualization and statistical pedagogy, his teaching helps equip students with skills valuable to a career in research. These classes bridge the gap between theoretical and applied, giving students the tools that they need in research for gathering, analyzing and interpreting their data.

In a recent interview with FASS, Sigal explains what brought him to SFU and inspires his work, and translates the Monte Carlo Simulation covered in his PhD dissertation for a non-Psychology audience.

What brought you to Simon Fraser University?

I must say, as a Battlestar Galactica fan, the possibility to teach in Caprica City was very enticing. However, what truly brought me here was the SFU Psychology Department itself. I've been welcomed into an area that combines my primary interests with my graduate level training, and into a department that supports its people. The feeling of camaraderie really makes a difference.

What inspired you to study psychology?

My first foray into post-secondary education was culinary school, where upon graduating I quickly realized that I did not want to work in restaurants forever. Thanks to a credit transfer program, I enrolled in the Psychology Department of Bishop's University. Being at a small, liberal arts institution allowed me to get to know my professors and catch their excitement for the field of experimental psychology. I remember Dr. Dale Stout lecturing excitedly about statistics, using the names of his various pets as examples, and falling under the spell of research methods, psychometrics, and experimental design during Dr. Stuart McKelvie’s presentations.

How did your interest in quantitative methods and data visualization for psychology research get started?

During my studies at York University, I realized the extent to which the history of statistics and the field of psychology are intertwined. Specifically, how many statistical approaches (such as factor analysis and structural equation modelling) were developed, at least in part, by psychologists struggling to models conceptss like intelligence or extroversion that we cannot measure directly. I found this fascinating and decided to apply to study under Dr. Michael Friendly, a cognitive scientist and expert in data visualization, programming, and statistics. This turned out to be one of the best decisions I have ever made. While quant is an area of psychology, it provides tools to play with data from outside of the discipline, and I have had opportunities to work with all sorts of researchers from a variety of fields.

Is making data accessible an important part of academic life for you? Why?

I truly believe data analysis and computer programming skills training is essential in psychology. We also need to communicate the results of such analyses in such a way that they will be understood. If our results are not accessible, they are easy to dismiss or, worse, misinterpret.

What are you most looking forward to in working at SFU and within the Department of Psychology?

I'm very excited to be involved in a new initiative for the department — the Psychological Methods Consulting Group — where I offer one-on-one statistical consulting, training, workshops, and various online resources. Psychology students are welcome to contact me about participating with the group, and I hope this might ignite the same passion for quantitative methods that doing similar work during my grad studies did for me.

Your PhD title is “Everything on the table: Tabular, graphic, and interactive approaches for interpreting and presenting Monte Carlo simulation data.” What’s the plain language translation for those of us not familiar with this work?

Monte Carlo simulation studies are computer-driven experimental investigations in which certain parameters, such as population means and standard deviations, are known or set a priori, and then used to generate random (but plausible) sample data. These generated data are then used to evaluate how robust a statistical model is under various circumstances. A typical application could determine how well a statistical method performs under pertinent assumption violations, such as when we have a small sample size or heterogeneous group variances. This allows researchers to get a sense of the predictive strength of a model under various messy, real world conditions.
The process of generating and analyzing data is repeated over many independent replications and across many different conditions. While implementing them has been made easier with R packages like SimDesign, the results tend to produce very long (and eye-wearying) tables. A focus of my doctoral work was on improving tabular, graphic, and interactive methods for communicating such findings.