Replication Package for `Code Review Comprehension: Reviewing Strategies Seen Through Code Comprehension Theories`
Bibliographic record
Abstract
# Replication Package for `Code Review Comprehension: Reviewing Strategies Seen Through Code Comprehension Theories` ## Abstract Modern code review is a critical component of the software development process for many projects. Despite its importance, there is limited understanding of the cognitive processes that enable reviewers to analyze code and provide meaningful feedback. Comprehending the code changes under review is both the most essential competence for reviewers and their primary challenge. To address this gap, we observed and interviewed ten experienced reviewers as they performed 25 code reviews from their review queue. Using Letovsky’s model of code comprehension, we performed a theory-driven thematic analysis to investigate how reviewers apply code comprehension to navigate changes and provide feedback. Our findings confirm that code comprehension is fundamental to the code review process. We extend Letovsky’s model to propose the Code Review Comprehension Model, demonstrating that code review, like code comprehension, relies on opportunistic strategies. These strategies typically begin with a context-building phase, followed by code inspection involving code reading, testing, and discussion management. Reviewers construct a mental model of the changeset as an extension of their understanding of the overall software system and use mental representations of expected and ideal solutions to interpret and evaluate the proposed changes. Based on our findings we propose the adoption of more human-centric code review tools and practices that support the strategies employed by reviewers and allow them to externalize and share their mental model of the PR and its alternatives with other developers. ## Structure ``` README.md Study-Design/ interview_structure.pdf demographics_survey.pdf participant_consent.pdf Data-Analysis/ transcripts/ P1R1.rtf P1R2.rtf ... codes/ coding_schema_2.pdf coding_schema_theory-driven.png codebook.pdf codes.docx co_author_review_1/ coded_transcript_P4R1.pdf coauthor_notes_P4R1.txt ... co_author_review_2/ coded_transcript_P2R2.pdf coauthor_notes_P2R2.txt transcripts/ P1R1.rtf P1R2.rtf ... ``` ## Contents of the Replication Package ### `Study-Design` Contains documents that outline the study’s methodology and design: - `interview_structure.pdf`: Structure and sample questions for the review&interview sessions. - `demographics_survey.pdf`: Survey capturing demographic data of participants. - `participant_consent.pdf`: Consent form signed by study participants before the recorded sessions. ### `Data-Analysis` Contains files used and produced in the analysis. - `transcripts/`: Raw transcripts from code review sessions (e.g., `P1R1.rtf` for Participant 1, Review 1). - `coding/`: Coding schema and list of final codes and themes. - `coding_schema_2.pdf`: The coding schema created based on Piglet's model of cognitive devlopment and Letovsky's comprehension model. - `coding_schema_theory-driven.png`:The initial coding schema created based off Letovsky's comprehension model and other theories as reported in the paper. - `codebook.pdf`: A complete list of codes created in the analysis - `co_author_review_1/`: Documents from the first co-author review, including coded transcripts and reviewer notes. - `co_author_review_2/`: Documents from the second co-author review, including coded transcripts, reviewer notes, and coding schema for knowledge and mental models. - `transcripts/`: Raw transcripts from code review sessions (e.g., `P1R1.rtf` for Participant 1, Review 1).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.150 | 0.700 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.006 | 0.008 |
| Science and technology studies | 0.004 | 0.005 |
| Scholarly communication | 0.006 | 0.006 |
| Open science | 0.006 | 0.007 |
| Research integrity | 0.003 | 0.007 |
| Insufficient payload (model declined to judge) | 0.091 | 0.022 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".