# How Do I Refactor This? An Empirical Study on Refactoring Trends and Topics in Stack Overflow

arXiv: https://arxiv.org/abs/2110.12229 · HTML: https://mlsanity.com/paper/2110.12229/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Oct 2021.

## Built on by (most cited first, 1 of 1)

- [What Were You Thinking? An LLM-Driven Large-Scale Study of Refactoring Motivations in Open-Source Projects](https://mlsanity.com/paper/2509.07763/) (2025)

## Abstract (arXiv)

An essential part of software maintenance and evolution, refactoring is performed by developers, regardless of technology or domain, to improve the internal quality of the system, and reduce its technical debt. However, choosing the appropriate refactoring strategy is not always straightforward, resulting in developers seeking assistance. Although research in refactoring is well-established, with several studies altering between the detection of refactoring opportunities and the recommendation of appropriate code changes, little is known about their adoption in practice. Analyzing the perception of developers is critical to understand better what developers consider to be problematic in their code and how they handle it. Additionally, there is a need for bridging the gap between refactoring, as research, and its adoption in practice, by extracting common refactoring intents that are more suitable for what developers face in reality. In this study, we analyze refactoring discussions on Stack Overflow through a series of quantitative and qualitative experiments. Our results show that Stack Overflow is utilized by a diverse set of developers for refactoring assistance for a variety of technologies. Our observations show five areas that developers typically require help with refactoring -- Code Optimization, Tools and IDEs, Architecture and Design Patterns, Unit Testing, and Database. We envision our findings better bridge the support between traditional (or academic) aspects of refactoring and their real-world applicability, including better tool support.