Relevant Data

I have had the privilege of working with Dr. Tracy Beth Høeg during the Covid era on a couple of papers — one that shed light on post-mRNA myocarditis signals in adolescents, and one on the elusive, non-significant differences in school masking policies. Tracy is a true scientist, physician, and patient safety advocate who courageously dove into the depths of Covid data and brought her findings to light publicly.

After spending the last couple of years co-developing data science notebooks, machine learning models, and LLM-enhanced demo applications, I’ve experienced firsthand both the power—and the tradeoffs—of building software this way. The debate about whether AI actually makes developers more productive is still unsettled. Some claim it dramatically accelerates work. Others argue it simply shifts effort to different parts of the development cycle.

With the long-overdue shake-up at the CDC this week, I’m revisiting and re-packaging my earlier research on the subject—both my independent work and articles co-authored with MDs and PhDs. Enjoy. Relevant Data is a reader-supported publication. To receive new posts and support my work, consider becoming a paid subscriber. Intellectual dishonesty. CDC style. Post-vaccination vs.

Last month, a Minnesota high school softball team won the state championship behind a seven-inning shutout thrown by their pitcher, a biological male competing under the name "Marissa." The team, which had previously struggled, suddenly surged to a 20-2 record. It was the kind of Cinderella story that sports media usually celebrates—except this one hinges on the uncomfortable and increasingly politicized reality of male athletes playing on female teams.

For years, the public has been told that concerns over males competing in girls’ sports are overblown—fringe talking points from politically motivated corners. But the data now speak for themselves. This isn’t rare or isolated. It’s systemic. And until now, no national institution—neither the NCAA, nor the NFHS, nor any major medical or athletic body—has bothered to count the cost.

In Part I of this analysis of the Reason Foundation’s recent report, we looked at overall spending vs. student acheivement. Throwing Money At It: Public Education Spending vs. Outcomes In Part II, let’s break this overall spending down into categories, starting with Non-teaching staff, and see how it relates to outcomes. The chart below from the Mississippi Center for Public Policy nicely combines comparative growth between spending, enrollment, teachers, and Admin.

A new report from the Reason Foundation that comprehensively collected and analyzed public education spending, enrollment, teacher/staff salary, and student outcomes over the past 2 decades has found some incredible insights. They compared all of these data for all 50 states, and were able to asses trends, relationships, and outcomes. The report can be found here, and I recommend saving this one as it is a treasure trove of data.

Social Emotional Learning is all the rage right now within the circles of education policy, research, and funding. Looking at Google Search Trends, we can see that the term reached it’s peak in September 2020. Which is an ironic time to have occurred. We see a slow build-up to August 2020, and then an immediate decline only a month afterwards. What could explain this?

Kelley Kroehnert has been a stalwart Covid data analyst throughout the last few years. Her work on checking CDC’s data and errors has been indispensable. She has recently undertaken the meticulous effort to catalogue the timeline of post-vax myocarditis and CDC’s communication and deliberate disregard for the safety signal, even as clear reports from hospitals were coming in of young males presenting with it. Subscribe to her substack here: Relevant Data is a reader-supported publication.

Introduction: The goal of this modeling project was to predict the outcome of the 2024 Election using public health, demographic, and historical data. The unique approach relies on predictors that are a proxy for public support for the Democratic party within a population. In the U.S., we are faced with a binary choice, Democrat or Republican, and our election is decided by Electoral votes from each state. Therefore, the response metric predicted was simply the margin of victory within a state.