Workshops

Upcoming Workshops

Statistical Horizons

Sensitivity Analysis for Causal Inference

March 5–6, 2025 (10:30am–12:30pm and 1:00pm–3:00pm Eastern time both days). Virtual.

Description:

In this seminar, you will:

Apply and understand techniques for quantifying the robustness of causal inferences.
- Comparing evidence to a threshold for inference.
- Understanding internal and external validity.
Conduct sensitivity analyses in R or the on-line app http://konfound-it.com (Stata and Excel also available).
Develop a deeper understanding of regression and the counterfactual as well as how threats to internal and external validity compare against the strength of evidence.
Apply sensitivity analysis to a specific problem of interest that may require extensions or adaptations.

MSU Course Drop-In

Zoom Room

April 1, 2025 (possibly April 8) (12:40–3:30pm Eastern time). Virtual.

April 7 & 9, 2025 (10:10–11:20am Eastern time). Virtual.

Check back later for more details and a Zoom link.

June 18–20, 2025

Society for Epidemiological Research

Check back later for a Zoom link.

August 5, 2025 (12:00–4:00pm Eastern time). Virtual.

Check back again soon for news about more workshops!

Past Workshops

Association for Education Finance & Policy (AEFP)

What Would It Take to Change Your Inference? Informing Discourse About Causal Inferences In the Educational Ecosystem

October 11, 2024 (1:00–4:30pm Eastern time). Virtual.

Description:

Statistical inferences are often challenged because of uncontrolled bias. There may be bias due to uncontrolled confounding variables or non-random selection into a sample. We will turn concerns about potential bias into questions about how much bias there must be to invalidate an inference. For example, challenges such as “But the inference of a treatment effect might not be valid because of pre-existing differences between the treatment groups” are transformed to questions such as “How much bias must there have been due to uncontrolled pre-existing differences to make the inference invalid?”

By reframing challenges about bias in terms of specific quantities, this course will contribute to scientific discourse about uncertainty of causal inferences. Critically, while there are other approaches to quantifying the sensitivity of inferences(e.g., Robins and Rotnitzky, 1995; Rosenbaum and Rubin 1983, Rosenbaum, 2000), the approaches presented in this workshop based on correlations of omitted variables (Frank, 2000) and the replacement of cases (Frank and Min, 2007; Frank et al, 2013) have greater intuitive appeal. In this sense the techniques inform a conversation among a broad set of stakeholders in the educational ecosystem, helping the educational community apply evidence to practice.

Academy of Management

What Would it Take to Change an Inference? Sensitivity Analysis to Inform Innovation in the Future

August 9, 2024 (8:00am–11:00am Central time). In-person: Chicago, IL.

Description:

Participants in this PDW will learn several approaches for quantifying the robustness of a causal inference. These provide a more precise language for producers and consumers to talk about potential concerns to inferences (e.g., omitted variables). The Impact Threshold for a Confounding Variable (ITCV) will be introduced to show participants how they can quantify how strong the correlations associated with an omitted variable must be to overturn an inference. The Robustness of Inference to Replacement (RIR) will be introduced to show participants how they can quantify what percentage of cases would have to be replaced with cases for which there was no effect of the predictor of interest to change the inference. Participants will learn to use the on-line application, spreadsheet, and the konfound macros in Stata and R. We will discuss guidelines for the application of sensitivity analysis in management research including how to create and interpret benchmarks based on observed covariates. These approaches allow researchers across a broad range of topic areas to debate the strength of evidence in intuitive and concrete terms necessary to inform innovations in the future.