pith. sign in

arxiv: 2504.17510 · v3 · pith:GEUQBN3Enew · submitted 2025-04-24 · 💻 cs.SE

Psychological Safety Framework in Pull-based Open Source Projects

Emeralda Sesari , Federica Sarro , Ayushi Rastogi This is my paper

Pith reviewed 2026-05-22 18:30 UTC · model grok-4.3

classification 💻 cs.SE
keywords psychological safetyopen source softwarepull requestssustained participationcontributor retentionGitHubcode review
0
0 comments X

The pith

Repositories with higher psychological safety retain more contributors over time, though prior participation predicts future activity even more strongly.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper develops a framework to measure psychological safety from how people interact in pull requests on GitHub. It finds that repositories with higher safety scores keep contributors active both within the next year and over four to five years. A sympathetic reader would care because open-source projects depend entirely on volunteers who can stop contributing at any time if interactions feel risky. The authors extract ten observable behaviors from 60,684 pull requests across 26 popular repositories to build a repository-level safety index and test it with logistic regression. They also show that whether someone has contributed before is a stronger signal of continued involvement than the safety measure itself.

Core claim

This paper introduces a theory-informed framework for measuring psychological safety through pull request data and provides empirical evidence of its relevance in sustaining participation within open-source development. Contributors are more likely to remain active in repositories with higher levels of psychological safety. Psychological safety is positively associated with both short-term and long-term sustained participation. However, prior participation emerges as a stronger predictor of future engagement, reducing the effect of psychological safety when accounted for.

What carries the argument

A repository-level psychological safety index built from ten observable variables extracted from pull request interactions, which operationalizes behaviors that signal safety during code review and serves as the key predictor of sustained contributor participation.

If this is right

  • Higher psychological safety increases the odds of both short-term and long-term contributor retention in a repository.
  • The ten-variable framework enables large-scale measurement of safety using existing pull request data without needing surveys.
  • Prior participation history outweighs the safety index when predicting who will stay active.
  • Open-source projects could sustain their contributor base by encouraging interaction patterns that build psychological safety during reviews.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same variable-extraction method could be tested on issue trackers or other collaborative platforms to see if the safety measure holds outside pull requests.
  • Projects might run experiments changing review comment guidelines and then track resulting shifts in the safety index and retention rates.
  • The stronger predictive power of prior participation suggests that early positive experiences for new contributors could be especially important for long-term engagement.
  • Observational safety measures like this one might be compared against traditional team settings to explore differences in how safety forms without formal roles.

Load-bearing premise

The ten observable variables extracted from pull-request interactions validly and reliably capture psychological safety in the open-source context.

What would settle it

A direct survey of contributors from the 26 repositories asking about their sense of psychological safety, then checking whether responses align with the computed index and with actual short-term and long-term participation records.

read the original abstract

Psychological safety refers to the belief that team members can speak up or make mistakes without fear of negative consequences. While it is recognized as important in traditional software teams, its role in open-source software development remains understudied. Open-source contributors often collaborate without formal roles or structures, where interpersonal relationships can significantly influence participation. Code review, a central and collaborative activity in modern software development, offers a valuable context for observing such team interactions. This paper introduces a framework grounded in psychological safety theory to identify behaviors that signal PS in pull request interactions. We operationalize these behaviors using 10 observable variables derived from 60,684 PRs across 26 popular GitHub repositories and construct a PS index at repository level. We then empirically test the relationship between this index and contributors' short-term (within 1 year) and long-term (over 4-5 years) sustained participation using three logistic regression models. Contributors are more likely to remain active in repositories with higher levels of psychological safety. Psychological safety is positively associated with both short-term and long-term sustained participation. However, prior participation emerges as a stronger predictor of future engagement, reducing the effect of psychological safety when accounted for. This study introduces a a theory-informed framework for measuring psychological safety through pull request data and provides empirical evidence of its relevance in sustaining participation within open-source development.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper introduces a theory-grounded framework for measuring psychological safety in pull-based open source projects. It identifies 10 observable variables from PR interactions, aggregates them into a repository-level PS index using data from 60,684 PRs across 26 GitHub repositories, and tests associations with short-term (1 year) and long-term (4-5 years) sustained contributor participation via logistic regression models. The central findings are that higher PS levels predict greater retention, though prior participation is a stronger predictor.

Significance. If the PS index is shown to validly measure the intended construct, the work would provide a scalable empirical bridge between organizational psychology and software engineering, offering evidence on factors sustaining OSS participation. The large-scale PR dataset and dual short/long-term outcome focus are strengths. However, without construct validation the incremental value over simpler activity measures remains unclear.

major comments (3)
  1. [Section 3 (Framework and variable operationalization)] The operationalization of the 10 variables (Section 3) claims grounding in psychological safety theory but provides no validation such as factor analysis, correlation with Edmondson’s PS scale, contributor self-reports, or inter-rater reliability on coded interactions. This is load-bearing for the claim that the repository-level index captures psychological safety rather than correlated factors like response latency or project activity volume.
  2. [Section 4 (PS index construction)] The aggregation rule or weights used to combine the 10 variables into the PS index (Section 4) are unspecified. Different aggregation choices could materially alter the index values fed into the regressions, undermining reproducibility and the interpretation of the reported positive associations.
  3. [Section 5 (Empirical analysis and regressions)] The logistic regression results (Section 5) indicate prior participation reduces the PS effect, yet the manuscript does not report full model specifications, controls for repo popularity or contributor tenure, effect sizes, or incremental R² / predictive power of the PS index over baseline models. This prevents assessment of whether PS adds explanatory value beyond prior activity.
minor comments (2)
  1. [Abstract] The abstract contains a repeated article: 'introduces a a theory-informed framework'.
  2. [Limitations and future work] Clarify in the limitations section how findings generalize beyond the 26 popular repositories sampled.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We appreciate the referee's detailed and constructive feedback on our manuscript. The comments highlight important areas for improving the clarity, reproducibility, and validity of our proposed psychological safety framework. We address each major comment below, indicating the revisions we plan to make in the next version of the manuscript.

read point-by-point responses

  1. Referee: [Section 3 (Framework and variable operationalization)] The operationalization of the 10 variables (Section 3) claims grounding in psychological safety theory but provides no validation such as factor analysis, correlation with Edmondson’s PS scale, contributor self-reports, or inter-rater reliability on coded interactions. This is load-bearing for the claim that the repository-level index captures psychological safety rather than correlated factors like response latency or project activity volume.

    Authors: We thank the referee for raising this critical point regarding construct validity. The 10 variables were selected based on a direct mapping from Edmondson's psychological safety theory to observable behaviors in pull request interactions, such as timely responses indicating support and acceptance of contributions signaling inclusivity. However, we acknowledge that the manuscript does not include empirical validation steps like factor analysis or correlation with established scales, as the study utilizes large-scale archival GitHub data rather than primary surveys. Inter-rater reliability does not apply directly since the variables are derived from automated extraction of PR metadata and comments rather than manual coding. We will revise the manuscript to include a more explicit discussion of the theoretical derivation in Section 3 and add a limitations subsection addressing the need for future validation studies, including potential self-report correlations. We maintain that the framework provides a novel, scalable approach, but agree that additional validation would strengthen the claims. revision: partial

  2. Referee: [Section 4 (PS index construction)] The aggregation rule or weights used to combine the 10 variables into the PS index (Section 4) are unspecified. Different aggregation choices could materially alter the index values fed into the regressions, undermining reproducibility and the interpretation of the reported positive associations.

    Authors: We apologize for the lack of detail in describing the PS index construction. In the current version, the repository-level PS index is computed as the mean of the 10 normalized variables (each scaled to [0,1] based on their distribution across repositories), with equal weights assigned to each variable reflecting their theoretical equivalence in the framework. We will update Section 4 to fully specify this aggregation method, including the normalization procedure and justification for equal weighting, to enhance reproducibility. revision: yes

  3. Referee: [Section 5 (Empirical analysis and regressions)] The logistic regression results (Section 5) indicate prior participation reduces the PS effect, yet the manuscript does not report full model specifications, controls for repo popularity or contributor tenure, effect sizes, or incremental R² / predictive power of the PS index over baseline models. This prevents assessment of whether PS adds explanatory value beyond prior activity.

    Authors: We agree that the regression analysis section would benefit from greater transparency. We will expand Section 5 to report the complete model specifications, including all control variables such as repository popularity (measured by stars or forks) and contributor tenure. We will present effect sizes using odds ratios and include model comparison metrics, such as changes in pseudo-R² or AIC, to demonstrate the incremental predictive power of the PS index over baseline models that include only prior participation. These additions will allow readers to better evaluate the unique contribution of psychological safety. revision: yes

standing simulated objections not resolved
  • Performing factor analysis, correlating with Edmondson’s PS scale, or collecting contributor self-reports would necessitate new primary data collection, which is outside the scope of the current archival study using existing GitHub PR data.

Circularity Check

0 steps flagged

No significant circularity in the empirical measurement and test chain

full rationale

The paper grounds its 10 variables in external psychological safety theory, extracts them from PR interaction logs to form a repository-level index, and then applies logistic regression to test associations with separate short- and long-term participation outcomes while controlling for prior participation. This is a standard empirical pipeline with distinct measurement and outcome constructs; no equation or step reduces the reported associations to the inputs by construction, no self-citation is load-bearing for the central claim, and the analysis remains falsifiable against the observed data without tautological renaming or fitted prediction presented as novel.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 1 invented entities

The central addition is the operationalization of psychological safety theory into ten PR-derived variables and a repository index; the main empirical step is logistic regression on participation outcomes. Details on variable selection, aggregation weights, and controls are absent from the abstract.

free parameters (1)
  • Aggregation rule or weights for PS index
    Abstract does not specify how the ten variables are combined into the repository-level index.
axioms (1)
  • domain assumption Psychological safety theory can be directly mapped to observable behaviors in open-source pull-request interactions
    The framework is grounded in theory but the translation to the OSS setting is assumed without stated justification in the abstract.
invented entities (1)
  • Psychological Safety Index no independent evidence
    purpose: Repository-level summary score of psychological safety derived from PR interaction variables
    Newly constructed metric; no external validation or independent evidence mentioned in the abstract.

pith-pipeline@v0.9.0 · 5768 in / 1534 out tokens · 98618 ms · 2026-05-22T18:30:17.117935+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

61 extracted references · 61 canonical work pages

  1. [1]

    Software e ngineering team diversity and performance,

    V . Pieterse, D. G. Kourie, and I. P . Sonnekus, “Software e ngineering team diversity and performance,” in Proceedings of the 2006 annual research conference of the South African institute of compu ter scientists and information technologists on IT research in developing countries, 2006, pp. 180–186

    work page 2006

  2. [2]

    Bringing the human factor to softw are eng

    L. Fernando Capretz, “Bringing the human factor to softw are eng.” IEEE Softw., vol. 31, no. 2, pp. 104–104, 2014

    work page 2014

  3. [3]

    Psychological safety and learning behav ior in work teams,

    A. Edmondson, “Psychological safety and learning behav ior in work teams,” Administrative science quarterly , vol. 44, no. 2, pp. 350–383, 1999

    work page 1999

  4. [4]

    What Google Learned From Its Quest to Build the Perfect T eam,

    “What Google Learned From Its Quest to Build the Perfect T eam,” https://www.nytimes.com/2016/02/28/magazine/what-google-learned-from-its-quest-to-build-the-perfect-te am.html, 2016, [Accessed 01-02-2025]

    work page 2016

  5. [5]

    Exploring psychological safety in softwar e engineer- ing: Insights from stack exchange,

    B. S. D. Santana, S. Freire, L. Cruz, L. Monte, M. Mendonca , and J. A. M. Santos, “Exploring psychological safety in softwar e engineer- ing: Insights from stack exchange,” in Proceedings of the XXXVII Brazil- ian Symposium on Software Engineering , ser. SBES ’23. New Y ork, NY , USA: Association for Computing Machinery, 2023, p. 503– 513

    work page 2023

  6. [6]

    Psychological safety in the software work environment,

    B. Santana, S. Freire, J. A. M. Santos, and M. Mendonc ¸a, “ Psychological safety in the software work environment,” IEEE Software, vol. 41, no. 4, pp. 86–94, 2024

    work page 2024

  7. [7]

    Psychological safety and norm c larity in software engineering teams,

    P . Lenberg and R. Feldt, “Psychological safety and norm c larity in software engineering teams,” in Proceedings of the 11th International W orkshop on Cooperative and Human Aspects of Software Engin eering, ser. CHASE ’18. New Y ork, NY , USA: Association for Computing Machinery, 2018, p. 79–86

    work page 2018

  8. [8]

    It is giving major sa tisfaction: Why fairness matters for developers,

    E. Sesari, F. Sarro, and A. Rastogi, “It is giving major sa tisfaction: Why fairness matters for developers,” arXiv preprint arXiv:2410.02482, 2024

    work page arXiv 2024

  9. [9]

    ”was my contribution fairly reviewed?

    D. M. German, G. Robles, G. Poo-Caama˜ no, X. Y ang, H. Iida , and K. Inoue, “”was my contribution fairly reviewed?” a framewo rk to study the perception of fairness in modern code reviews,” in Proceedings of the 40th International Conference on Software Engineering , 2018, pp. 523–534

    work page 2018

  10. [10]

    Psychological Safety Short Course,

    “Psychological Safety Short Course,” https://handbook.gitlab.com/handbook/leadership/emotional-intelligence/psychological-safety-short-cour se/#what-impacts-psychological-safety, 2021, [Accessed 01-02-2025]

    work page 2021

  11. [11]

    The role of psycho logical safety in promoting software quality in agile teams,

    A. Alami, M. Zahedi, and O. Krancher, “The role of psycho logical safety in promoting software quality in agile teams,” Empirical Softw. Engg. , vol. 29, no. 5, Jul. 2024

    work page 2024

  12. [12]

    How do team cohesion and psychological saf ety impact knowledge sharing in software development projects?

    A. K. Kakar, “How do team cohesion and psychological saf ety impact knowledge sharing in software development projects?” Knowledge and Process Management, vol. 25, no. 4, p. 258–267, 2018

    work page 2018

  13. [13]

    Perceptio ns of diversity on git hub: A user survey,

    B. V asilescu, V . Filkov, and A. Serebrenik, “Perceptio ns of diversity on git hub: A user survey,” in 2015 IEEE/ACM 8th International W orkshop on Cooperative and Human Aspects of Software Engineering . IEEE, 2015, pp. 50–56

    work page 2015

  14. [14]

    Understanding sustained parti cipation in open source software projects,

    Y . Fang and D. Neufeld, “Understanding sustained parti cipation in open source software projects,” Journal of Management Information Systems , vol. 25, no. 4, pp. 9–50, 2009

    work page 2009

  15. [15]

    Underst anding the motivations, participation, and performance of open sourc e software developers: A longitudinal study of the apache projects,

    J. A. Roberts, I.-H. Hann, and S. A. Slaughter, “Underst anding the motivations, participation, and performance of open sourc e software developers: A longitudinal study of the apache projects,” Management science, vol. 52, no. 7, pp. 984–999, 2006

    work page 2006

  16. [16]

    Motivation of s oftware devel- opers in open source projects: an internet-based survey of c ontributors to the linux kernel,

    G. Hertel, S. Niedner, and S. Herrmann, “Motivation of s oftware devel- opers in open source projects: an internet-based survey of c ontributors to the linux kernel,” Research policy , vol. 32, no. 7, pp. 1159–1177, 2003

    work page 2003

  17. [17]

    Developer turnov er in global, industrial open source projects: Insights from applying su rvival analy- sis,

    B. Lin, G. Robles, and A. Serebrenik, “Developer turnov er in global, industrial open source projects: Insights from applying su rvival analy- sis,” in 2017 IEEE 12th International Conference on Global Software Engineering (ICGSE) . IEEE, 2017, pp. 66–75

    work page 2017

  18. [18]

    Who will remai n? an eval- uation of actual person-job and person-team fit to predict de veloper retention in floss projects,

    A. Schilling, S. Laumer, and T. Weitzel, “Who will remai n? an eval- uation of actual person-job and person-team fit to predict de veloper retention in floss projects,” in 2012 45th Hawaii international conference on system sciences . IEEE, 2012, pp. 3446–3455

    work page 2012

  19. [19]

    Will my patch make i t? and how fast? case study on the linux kernel,

    Y . Jiang, B. Adams, and D. M. German, “Will my patch make i t? and how fast? case study on the linux kernel,” in 2013 10th W orking conference on mining software repositories (MSR) . IEEE, 2013, pp. 101–110

    work page 2013

  20. [20]

    Wil l they like this? evaluating code contributions with language models,

    V . J. Hellendoorn, P . T. Devanbu, and A. Bacchelli, “Wil l they like this? evaluating code contributions with language models, ” in 2015 IEEE/ACM 12th W orking Conference on Mining Software Reposi tories. IEEE, 2015, pp. 157–167

    work page 2015

  21. [21]

    A study of external c ommunity contribution to open-source projects on github,

    R. Padhye, S. Mani, and V . S. Sinha, “A study of external c ommunity contribution to open-source projects on github,” in Proceedings of the 11th working conference on mining software repositories, 2014, pp. 332– 335

    work page 2014

  22. [22]

    Writing acceptable patches: An empirical study of open source project patches,

    Y . Tao, D. Han, and S. Kim, “Writing acceptable patches: An empirical study of open source project patches,” in 2014 IEEE International Conference on Software Maintenance and Evolution . IEEE, 2014, pp. 271–280

    work page 2014

  23. [23]

    An explorato ry study of the pull-based software development model,

    G. Gousios, M. Pinzger, and A. v. Deursen, “An explorato ry study of the pull-based software development model,” in Proceedings of the 36th international conference on software engineering , 2014, pp. 345–355

    work page 2014

  24. [24]

    Going farther together: The impact of social capital on sustained participation in open source,

    H. S. Qiu, A. Nolte, A. Brown, A. Serebrenik, and B. V asil escu, “Going farther together: The impact of social capital on sustained participation in open source,” in 2019 IEEE/ACM 41st International Conference on Software Engineering (ICSE) , 2019, pp. 688–699

    work page 2019

  25. [25]

    Pull req uest gov- ernance in open source communities,

    A. Alami, R. Pardo, M. L. Cohn, and A. Wasowski, “Pull req uest gov- ernance in open source communities,” IEEE Transactions on Software Engineering, vol. 48, no. 12, pp. 4838–4856, 2021

    work page 2021

  26. [26]

    What are they talking about? analyzing code reviews in pull-based de velopment model,

    Z.-X. Li, Y . Y u, G. Yin, T. Wang, and H.-M. Wang, “What are they talking about? analyzing code reviews in pull-based de velopment model,” Journal of Computer Science and Technology, vol. 32, pp. 1060– 1075, 2017

    work page 2017

  27. [27]

    What make long term contributors : Will- ingness and opportunity in oss community,

    M. Zhou and A. Mockus, “What make long term contributors : Will- ingness and opportunity in oss community,” in 2012 34th International Conference on Software Engineering (ICSE) . IEEE, 2012, pp. 518–528

    work page 2012

  28. [28]

    On the shoulders of gian ts: A new dataset for pull-based development research,

    X. Zhang, A. Rastogi, and Y . Y u, “On the shoulders of gian ts: A new dataset for pull-based development research,” in Proceedings of the 17th international conference on mining software repositories, 2020, pp. 543– 547

    work page 2020

  29. [29]

    A dataset for pull-based dev elopment research,

    G. Gousios and A. Zaidman, “A dataset for pull-based dev elopment research,” in Proceedings of the 11th W orking Conference on Mining Software Repositories, 2014, pp. 368–371

    work page 2014

  30. [30]

    Social networkin g meets software development: Perspectives from github, msdn, stack exchan ge, and topcoder,

    A. Begel, J. Bosch, and M.-A. Storey, “Social networkin g meets software development: Perspectives from github, msdn, stack exchan ge, and topcoder,” IEEE software , vol. 30, no. 1, pp. 52–66, 2013

    work page 2013

  31. [31]

    Understanding th e factors that impact the popularity of github repositories,

    H. Borges, A. Hora, and M. T. V alente, “Understanding th e factors that impact the popularity of github repositories,” in 2016 IEEE international conference on software maintenance and evolution (ICSME) . IEEE, 2016, pp. 334–344

    work page 2016

  32. [32]

    What’s in a github star? und erstanding repository starring practices in a social coding platform,

    H. Borges and M. T. V alente, “What’s in a github star? und erstanding repository starring practices in a social coding platform, ” Journal of Systems and Software , vol. 146, pp. 112–129, 2018

    work page 2018

  33. [33]

    4.5 million (suspected) fake stars in github: A growing spiral of popularity contests, scams, and malware,

    H. He, H. Y ang, P . Burckhardt, A. Kapravelos, B. V asiles cu, and C. K¨ astner, “4.5 million (suspected) fake stars in github: A growing spiral of popularity contests, scams, and malware,” 2024. [ Online]. Available: https://arxiv.org/abs/2412.13459

    work page arXiv 2024

  34. [34]

    No ri sk, no reward: Towards an automated measure of psychological safe ty from online communication,

    S. A. Ferguson, G. V an de Zande, and A. Olechowski, “No ri sk, no reward: Towards an automated measure of psychological safe ty from online communication,” in Extended Abstracts of the CHI Conference on Human Factors in Computing Systems , 2024, pp. 1–7

    work page 2024

  35. [35]

    The effects of m ild task conflict and relationship conflict on psychological saf ety and team effectiveness: Team-level analysis,

    O. Seon Y ong, K. Y eong Sik, and K. In Hye, “The effects of m ild task conflict and relationship conflict on psychological saf ety and team effectiveness: Team-level analysis,” Korean Jounal of Industrial Organizational Psychology, vol. 32, no. 1, pp. 83–106, 2019

    work page 2019

  36. [36]

    Optimizing team confli ct dynamics for high performance teamwork,

    T. A. O’Neill and M. J. McLarnon, “Optimizing team confli ct dynamics for high performance teamwork,” Human Resource Management Review, vol. 28, no. 4, pp. 378–394, 2018

    work page 2018

  37. [37]

    Statistics for c linicians: An introduction to logistic regression

    M. M. Wiest, K. J. Lee, and J. B. Carlin, “Statistics for c linicians: An introduction to logistic regression.” Journal of Paediatrics & Child Health, vol. 51, no. 7, 2015

    work page 2015

  38. [38]

    Poster: The impact of group im balance on logistic regression analyses with assessment data,

    A. Alkhalaf and B. Zumbo, “Poster: The impact of group im balance on logistic regression analyses with assessment data,” in ITC 2016 Conference, 2016

    work page 2016

  39. [39]

    Effect of zero imputation methods for log-t ransformation of independent variables in logistic regression,

    S. Y . Park, “Effect of zero imputation methods for log-t ransformation of independent variables in logistic regression,” Communications for Statistical Applications and Methods , vol. 31, no. 4, pp. 409–425, 2024

    work page 2024

  40. [40]

    Application of robust regression in translational neuros cience studies with non-gaussian outcome data,

    M. Malek-Ahmadi, S. D. Ginsberg, M. J. Alldred, S. E. Cou nts, M. D. Ikonomovic, E. E. Abrahamson, S. E. Perez, and E. J. Mufs on, “Application of robust regression in translational neuros cience studies with non-gaussian outcome data,” Frontiers in Aging Neuroscience , vol. 15, p. 1299451, 2024

    work page 2024

  41. [41]

    Logistic regression in rare events data,

    G. King and L. Zeng, “Logistic regression in rare events data,” Political analysis, vol. 9, no. 2, pp. 137–163, 2001

    work page 2001

  42. [42]

    Predictive per formance of logistic regression for imbalanced data with categorica l covariate

    H. A. A. Rahman, Y . B. Wah, and O. S. Huat, “Predictive per formance of logistic regression for imbalanced data with categorica l covariate.” Pertanika Journal of Science & Technology , vol. 29, no. 1, 2021

    work page 2021

  43. [43]

    Imbalance effects on clas sification using binary logistic regression,

    H. A. Abd Rahman and B. W. Y ap, “Imbalance effects on clas sification using binary logistic regression,” in Soft Computing in Data Science: Second International Conference, SCDS 2016, Kuala Lumpur , Malaysia, September 21-22, 2016, Proceedings 2 . Springer, 2016, pp. 136–147

    work page 2016

  44. [44]

    Aggregation of binary evaluations: a borda-like approach,

    C. Duddy, A. Piggins, and W. S. Zwicker, “Aggregation of binary evaluations: a borda-like approach,” Social Choice and W elfare, vol. 46, pp. 301–333, 2016

    work page 2016

  45. [45]

    Should we care (more) abou t data aggre- gation?

    K. Gr¨ undler and T. Krieger, “Should we care (more) abou t data aggre- gation?” European Economic Review , vol. 142, p. 104010, 2022

    work page 2022

  46. [46]

    Injury sever ity analysis: comparison of multilevel logistic regression models and ef fects of collision data aggregation,

    T. Usman, L. Fu, and L. F. Miranda-Moreno, “Injury sever ity analysis: comparison of multilevel logistic regression models and ef fects of collision data aggregation,” Journal of Modern Transportation , vol. 24, pp. 73–87, 2016

    work page 2016

  47. [47]

    Understanding

    J. Sheoran, K. Blincoe, E. Kalliamvakou, D. Damian, and J. Ell, “Understanding” watchers” on github,” in Proceedings of the 11th working conference on mining software repositories , 2014, pp. 336–339

    work page 2014

  48. [48]

    Do women suffer from network closure? the mo derating effect of social capital on gender inequality in a project-b ased labor market, 1929 to 2010,

    M. Lutter, “Do women suffer from network closure? the mo derating effect of social capital on gender inequality in a project-b ased labor market, 1929 to 2010,” American Sociological Review , vol. 80, no. 2, pp. 329–358, 2015

    work page 1929

  49. [49]

    Disruptive diver sity and recur- ring cohesion: Assembling creative teams in the video game i ndustry, 1979-2009,

    M. d. V aan, B. V edres, and D. C. Stark, “Disruptive diver sity and recur- ring cohesion: Assembling creative teams in the video game i ndustry, 1979-2009,” 2011

    work page 1979

  50. [50]

    Scientific collaboration networks. ii. s hortest paths, weighted networks, and centrality,

    M. E. Newman, “Scientific collaboration networks. ii. s hortest paths, weighted networks, and centrality,” Physical review E , vol. 64, no. 1, p. 016132, 2001

    work page 2001

  51. [51]

    Developer onboarding in github: the role of prior social links and lang uage expe- rience,

    C. Casalnuovo, B. V asilescu, P . Devanbu, and V . Filkov, “Developer onboarding in github: the role of prior social links and lang uage expe- rience,” in Proceedings of the 2015 10th joint meeting on foundations of software engineering , 2015, pp. 817–828

    work page 2015

  52. [52]

    How early par - ticipation determines long-term sustained activity in git hub projects?

    W. Xiao, H. He, W. Xu, Y . Zhang, and M. Zhou, “How early par - ticipation determines long-term sustained activity in git hub projects?” in Proceedings of the 31st ACM Joint European Software Enginee ring Conference and Symposium on the F oundations of Software Engineering, 2023, pp. 29–41

    work page 2023

  53. [53]

    A large scale stud y of long- time contributor prediction for github projects,

    L. Bao, X. Xia, D. Lo, and G. C. Murphy, “A large scale stud y of long- time contributor prediction for github projects,” IEEE Transactions on Software Engineering, vol. 47, no. 6, pp. 1277–1298, 2019

    work page 2019

  54. [54]

    On the effect of disc ussions on pull request decisions

    M. Golzadeh, A. Decan, and T. Mens, “On the effect of disc ussions on pull request decisions.” in BENEVOL, 2019

    work page 2019

  55. [55]

    Women’s participation in open source software: A survey of the lit- erature,

    B. Trinkenreich, I. Wiese, A. Sarma, M. Gerosa, and I. St einmacher, “Women’s participation in open source software: A survey of the lit- erature,” ACM Transactions on Software Engineering and Methodology (TOSEM), vol. 31, no. 4, pp. 1–37, 2022

    work page 2022

  56. [56]

    Climate coach: A dashboard fo r open- source maintainers to overview community dynamics,

    H. S. Qiu, A. Lieb, J. Chou, M. Carneal, J. Mok, E. Amspoke r, B. V asilescu, and L. Dabbish, “Climate coach: A dashboard fo r open- source maintainers to overview community dynamics,” in Proceedings of the 2023 CHI Conference on Human Factors in Computing Syst ems, 2023, pp. 1–18

    work page 2023

  57. [57]

    Almost there: A study on quasi-contributors in open source software proje cts,

    I. Steinmacher, G. Pinto, I. S. Wiese, and M. A. Gerosa, “ Almost there: A study on quasi-contributors in open source software proje cts,” in Pro- ceedings of the 40th international conference on software e ngineering, 2018, pp. 256–266

    work page 2018

  58. [58]

    Work pract ices and challenges in pull-based development: The contributor’s p erspective,

    G. Gousios, M.-A. Storey, and A. Bacchelli, “Work pract ices and challenges in pull-based development: The contributor’s p erspective,” in Proceedings of the 38th international conference on softwa re engi- neering, 2016, pp. 285–296

    work page 2016

  59. [59]

    Influence of socia l and technical factors for evaluating contribution in github,

    J. Tsay, L. Dabbish, and J. Herbsleb, “Influence of socia l and technical factors for evaluating contribution in github,” in Proceedings of the 36th international conference on Software engineering , 2014, pp. 356–366

    work page 2014

  60. [60]

    The promises and perils of mining github,

    E. Kalliamvakou, G. Gousios, K. Blincoe, L. Singer, D. M . German, and D. Damian, “The promises and perils of mining github,” in Proceedings of the 11th working conference on mining software repositor ies, 2014, pp. 92–101

    work page 2014

  61. [61]

    Supplementary materials for the study on psychologic al safety sustains participation in pull-based open source projects,

    “Supplementary materials for the study on psychologic al safety sustains participation in pull-based open source projects,” 2025. [ Online]. Available: https://figshare.com/s/567fcbccc4fd4b75007d

    work page 2025