Cause and effect relationships are harder to conclude than most think – especially in the field of social science research. Professor Ken Frank received a three-year, nearly $900,000 grant from the Institution of Education Sciences (IES) that will provide a more precise statistical dialogue about causal inferences among clinicians, researchers, policymakers and the general public.
Frank’s application will ultimately improve the validity of quantitative research. Improving the way we talk about inferences helps stakeholders use evidence to inform practice.
The technique, developed by Frank and this team, will allow researchers to reanalyze quantitative statistical studies in the field of social science and quantify the robustness of their conclusions. For example, researchers will have the ability to determine what percentage of data in a study would have to be due to bias to change an inference from the data.
The project, titled: “Quantifying the Robustness of Causal Inferences: Extensions and Application to Existing Databases,” looks to advance, extend and apply existing statistical techniques to make them more useful for education research and practice. Frank and his team will analyze causal inference — a field of study within statistics and data analysis that aims to understand and quantify cause-and-effect relationships in education policy and practice.
Although quantitative research aims for objectivity and precision, a variety of factors can cause bias.
“When we talk about causal inferences, the paradigm shift is that we think a group of people will make the inference, not the lone scientist who evaluated theory relative to data,” said Frank. “There are a bunch of people who are going talk about how strong the evidence is relative to that theory, and they need language for communicating how confident they are in their inference. We will give them the better language to talk about it.”
Several of the project’s co-investigators are Spartan alumni who worked with Frank as students:
- Guan Saw, Ph.D. ’16 (Measurement and Quantitative Methods)
- Ran Xu, Ph.D. ’16 (Measurement and Quantitative Methods)
- Joshua Rosenberg Ph.D. ’18 (Educational Psychology & Educational Technology)
“I have been very fortunate to work with some great students over the years. This project allows me to continue working with many of them as co-investigators now that they have graduated and established their own careers at various institutions throughout the country and internationally. They are, individually and collectively, a delight to work with,” said Frank.