SIDE-lib: A Library for Detecting Symptoms of Python Programming Misconceptions

Abstract

Extensive prior work has identified and described misconceptions held by novice programmers. Much of this prior work has involved at least some automatic detection of potential misconceptions using a variety of methods such as intercepting compiler error messages, pattern matching, and black-box testing. To the best of our knowledge, no independent and flexible tool for automatic detection of misconceptions is currently available to the research community, meaning that detection must be reimplemented from scratch for each new project that aims to understand or support novice programmers using automatic analysis. This is time-consuming work, particularly for misconceptions that require understanding of the context of a program beyond localised syntax patterns. In this paper, we introduce SIDE-lib, a standalone library for detecting symptoms of Python misconceptions. This library is made available with the goal of simplifying and speeding up research on Python misconceptions and the development of tools to support learning. We also describe example use cases for the library, including how we are using it in our ongoing research.