Presenting my research poster

Author: Steicy Lopez | Major: Economics | Semester: Spring 2025

My name is Steicy Lopez and I am a senior studying Economics. I have had the pleasure of working with Dr. Gema Zamarro in the Education Reform department and receive honors college support for Fall 2024 and Spring 2025. I am pursuing graduate school or research positions in economics for my post-graduation plans.

For my honors thesis research, I decided to study school type switches during and after the COVID-19 pandemic. This research was pressing as public schools lost more than 1.2 million students between 2020-22 (Dee, 2023) Yet, we didn’t know much about the types of families who left and whether they had returned. Additionally, we wanted to look further into an enrollment gap pointed out by previous research (Dee, 2023) where students left public schooling but did not reappear in private and homeschooling enrollment numbers. We found that socioeconomic factors, political leanings, and trust in media were more associated with school type switches than perceived risk of COVID-19 or medical conditions. Additionally, non-enrollment in our sample grew by almost 60% from April 2020 to December 2020, mirroring the non-enrollment gap found in administrative data.

For this research, I ran sequentially built regression models of 5 different types and ran a Random Forest algorithm as a robustness check. Models were run by phase of the pandemic, by type of switch, and with up to 15 covariates. Once I had finished collecting all of these results, I felt a wave of panic wash over me wondering how I would explain the data in my paper. Sacrificing detail for conciseness made me worried I would be misleading the reader, or I could be misinterpreted. Additionally, keeping track of over 20 sets of results was exhausting and I kept losing track of the models.

To tackle this problem, I decided to create color coded graphs of my results to get a more comprehensive view. These graphs didn’t make it into my paper and were a significant investment of time. Yet, they were completely worth it because they allowed me to find a story in the data. Then, I wrote down everything I wanted to say about the data, making sure I was citing which model that came from. Once I had a list of results of each model, I was able to see common trends to give the reader key takeaways. This list helped me research potential policy responses and impacts of these results. I realized that once I did this, the way I talked about my thesis changed. Instead of talking about the magnitude of the results or digging into each model, I had a concise narrative to share. I’ve noticed this has been much more impactful including when talking to researchers and people in academia.

The greatest lesson I took from this is just how important scientific communication is. Without clearly linking results to real-world impacts, the value of research risks being overlooked. My biggest piece of advice for those working on their thesis is to really take the time on this link and finding the story in the results. Do this by taking one bite at a time and don’t be afraid to create visualizations/guides just for yourself.

References

Dee, T. S. (2023). Where the Kids Went: Nonpublic Schooling and Demographic Change during the Pandemic Exodus from Public Schools. Urban Institute