Art vibrancy in the State of Virginia

Meet the Data Science for the Public Good team for this project

• Undergraduate Interns: Alicia Bacani​ and David Xing
• Graduate Mentors: Angana Chatterjee and Xiaofan Zhu​
• Faculty Advisors: Michael Cary, Yujuan Gao, and Le Wang

Project description

Arts and culture are vital to strengthening communities through building civic engagement and stimulating economic development. However, understanding the impact of arts vibrancy can limit communication for both society and policymakers. We developed a data-driven framework to help bridge this gap in communication by evaluating arts vibrancy across the state of Virginia.​

We examined how different methods affect the interpretability of the arts vibrancy index. We compared three weighting methods: Equal Weighting, Factor Analysis, and Principal Component Analysis (PCA).​ Our results showed that regardless of the methodology used, the indices remained robust with consistent results and rankings.​ However, the interpretability of the index scores varied across the three methods, as one interpretation could be more complex than another.​ We found that utilizing equal weighting made for the most transparent interpretation compared to factor analysis and PCA. ​Combining our statistical methodologies with visuals, our research serves as a practical tool for policymakers, arts organizations, and community stakeholders to better understand Virginia’s arts ecosystem as a whole.​

Intern insights

Alicia: When I was selected to be a part of a project that involved art, I thought that I would dread the next ten weeks ahead of me. Now, finally nearing the finish line, I could not have been more wrong. 

Throughout my 10 weeks, I learned more and more about how the arts contribute to the economy. While my previous experience in economic work were soley based on robotics or engineering-based projects, this internship taught me that experiences help expand that thinking beyond the knowledge of what we are content with.  

While I was learning R, along with index construction, I noticed how valuable that knowledge would be in any career I choose to be in. Data analytics is everywhere in modern-day society, and any job in CS will most likely require it. Given this experience, I am more than glad to say that I have learned valuable knowledge I will be taking with me to the workforce. 

Additionally, working with my project partner (David) was nice. Considering our dynamic, we both communicated what we were doing throughout our project to ensure no one was lost in the process of our project development. Alongside this communication, we both contributed to the design of the app and how we could engineer this to be more engaging to a wider audience.  

The most important lessons I’ve learned from this internship are to not judge something based on your initial thought processes. If I hadn’t kept an open mind, I am sure that the initial “dreading scenario” may have gotten the best of me, but now that I am at the end, I can connect the dots and start to understand that this experience was more than meant for me. Furthermore, as I move on with my career, I will most definitely be conducting more economic and artistic-based projects/research further into my career.  

David: On my first day of the Data Science for the Public Good program, I was assigned to work on a project where I was to create an “arts vibrancy index.” My initial thoughts were ones of much confusion. How could data be connected to art? 

The main goal of this project was to compare three different methods of constructing an index and see which method would make for the most interpretable and communicable index. Learning more about index construction as a general statistical methodology was my favorite part of working on this project, but learning more about the value of art in society also added to the experience.  

I did not just learn more about statistics and data analytics, though. I realized more about just how important the aspect of communication truly is. Perhaps I can understand my own work and my own results, but can others also understand? I realized that I could answer just about any question I had regarding my work, but were my answers clear enough to the ones who were asking the questions? Throughout this project, I was particularly able to learn more about how to appeal to a variety of audiences. With some audiences, it may be appropriate to get all technical, but other audiences may just need a general idea. 

I now come out of these ten weeks with not just more knowledge in the statistics field, but also more knowledge about the value of art and its impact on society. I also leave having a much better sense of how to communicate across a wide range of audiences. 

Graduate mentor insights

Angana Chatterjee

This summer I served as a graduate mentor on the Art Vibrancy in Virginia project, alongside co-mentor Xiaofan Zhu and a team of undergraduate researchers, in partnership with the SMU DataArts Research Team and the Kohl Centre. Our goal was to build a Virginia-specific, county-level version of the national Arts Vibrancy Index, a composite measure of nonprofit arts organizations, arts-related employment, and government grant funding. This is designed to make rural counties visible in a framework where they currently aren't. I took on this project because it let me apply my training in economics to a question with real stakes for arts policy in Virginia. 

A large part of my work was methodological. The national AVI uses factor analysis to weight its thirteen component variables, but factor analysis behaves unpredictably with the kind of data Virginia produces: many rural counties report zeros across entire categories, and the method can amplify sparse non-zero signals in ways that distort rankings. I, along with Xiaofan, guided the team through building two parallel indices: one equally weighted, one factor analytic.  The framing of the comparison is not a contest between methods but as a robustness check. The key finding was that roughly 95% of counties shifted by five ranks or fewer between the two approaches, with disagreements concentrated in exactly the sparse rural counties where zero-inflation creates instability. That result told us something important: the index is measuring something real, not an artifact of how we built it. 

Our final analysis confirmed a widening gap in arts vibrancy between Virginia's metro and rural areas, driven largely by how grant funding is distributed. What I'll carry forward is a sharper sense of how to position empirical work for non-technical audiences without sacrificing precision. The AVI project reminded me that a ranking is only as useful as the story someone can tell with it, and that story must be honest about what the data can and cannot support. 

What our stakeholders had to say about their experience

It was a pleasure working with David, Alicia, and everyone on the team. We greatly appreciate the thoughtful development and innovative approaches the team brought to calculating and interpreting the AVI, and assessing the robustness of different methodologies.

We were especially impressed by how the team worked with the complex data, addressed its limitations, sought out local partners, and delivered an effective presentation and poster within a short period of time. The AVI is one of our key products, and your work has provided us with a revised framework and several promising ideas for further testing and expansion. You have contributed not only to the continued development of the AVI but also to the broader arts and culture sector by improving how arts vibrancy is measured and interpreted. 

We'd like to sincerely thank all the fellow students, faculty, and staff who supported this collaboration. We are also grateful for the USDA grant that made the program possible.

We look forward to continuing the conversations as we explore scaling this work to the national level.

Wenhua Di
Research Director, SMU DataArts, Meadows School of the Arts
Research Professor, Arts Management and Arts Entrepreneurship & Economics
Southern Methodist University

Click here to view all of the 2026 DSPG projects. >>