On Designing Programming Error Messages for Novices: Readability and its Constituent Factors

Programming Education & Computational ThinkingUser Research Methods (Interviews, Surveys, Observation)K-12 TeachersOnline Tutors

Title of the Paper

How to Design Error Messages Suitable for Programming Novices: Readability and Its Constituent Factors
(On Designing Programming Error Messages for Novices: Readability and its Constituent Factors)

Bibliographic Information

  • Research Domain: Programming Education, Human-Computer Interaction
  • Keywords: Compiler Design and Implementation, Compiler Error Messages, Programming Education, Novice Programmers, Readability, HCI Theoretical Concepts, Information Design Guidelines, Programming Language Design, CS1 Students, Error Message Improvement

Research Background and Problem

  • Identified Problem: The quality of error message design has not received sufficient attention, especially for novices. These messages are often perceived as obscure, filled with jargon and technical challenges, hindering learning.
  • Significance: Programming error messages are fundamental interactive elements for students learning to code. Poor readability of these messages can directly impact the learning efficiency and confidence of programming novices, while also increasing the teaching burden on educators.
  • Research Motivation: Although many studies and design recommendations emphasize readability, previous suggestions often lack clear definitions or empirical support. Existing designs have yet to address the core issues affecting message readability, which differs from natural language readability. This paper aims to conduct an in-depth investigation into the readability of programming error messages, responding to ongoing demands in the field and providing a practical foundation for improving error message design.

Solution

  • Proposed Method:

    1. Investigate the readability of programming error messages and its influencing factors through three experiments:
      • Experiment 1: Identify factors that may affect the readability of error messages.
      • Experiment 2: Collect feedback from programming novices on factors influencing message readability through open-ended surveys.
      • Experiment 3: Quantitatively confirm the specific impact of these factors.
    2. Define four core factors related to information design: length (whether verbose), jargon (whether containing difficult terminology), sentence structure (whether clear), and lexical difficulty (whether using complex vocabulary).
  • Innovative Contributions:

    1. For the first time, clearly define the readability of programming error messages and test relevant factors.
    2. Propose specific design guidelines to directly optimize error message design: de-jargonization, use of complete sentences, simplified vocabulary, and concise wording.
    3. Demonstrate that these design factors significantly impact the readability of error messages, enhancing the effectiveness of error messages in programming environments.
  • Key Techniques and Implementation Steps:

    1. Reasonable grouping and questionnaire design during experimental data collection to cover the most common error messages.
    2. Conduct multi-national, cross-disciplinary experimental analyses to ensure strong external validity of the data.
    3. Use statistical analysis techniques (e.g., Kruskal-Wallis, Pearson correlation coefficients) to validate conclusions.

Research Results

  • Specific Findings:

    1. Identified four key factors affecting the readability of programming error messages: message length, use of jargon, clarity of sentence structure, and simplicity of vocabulary.
    2. Experimental results demonstrate that these four factors significantly influence the readability and comprehensibility of error messages.
    3. Proposed four specific design guidelines for error messages: de-jargonization, use of complete sentences, simplified vocabulary, and concise wording.
  • Advantages and Improvements:

    1. Provides clear, modern evidence for designing programming error messages, improving feedback mechanisms to make learning more efficient.
    2. Offers special optimization suggestions for non-native English users, supporting internationalized curricula.
    3. Combines large-scale cross-regional data to ensure broad applicability of research findings.
  • Experimental Results:
    Across all tests, strong correlations were observed between the identified factors and readability (e.g., lexical correlation r=0.68). Simpler, jargon-free messages were more appealing to novices, and the proposed recommendations significantly improved message effectiveness.

  • Limitations and Future Directions:

    1. Sample selection bias exists, with cultural and linguistic backgrounds influencing results.
    2. Potential factors affecting readability have not been exhaustively identified, requiring further exploration.
    3. Future research should develop metrics for automatically evaluating the readability of programming error messages.
    4. Test the specific impact of newly designed guidelines on learning outcomes.

Summary and Application

This paper provides a systematic research foundation for designing programming error messages, summarizing four key elements for efficient design through experimental validation. These findings have broad value in education and programming language design. Future work can build on this foundation to develop more automated tools or metrics to further optimize programming error message design.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/47768/2021

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3411764.3445696
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2021
emoji_events
Award
No award tagged
group
Authors
7 authors
sell
Subtopics
Programming Education & Computational Thinking, User Research Methods (Interviews, Surveys, Observation)
work
Professions
K-12 Teachers, Online Tutors
article
Content Status
Full text indexed
hub
Related Papers
0 related papers