Elamur Research⌕
☰
home / papers / 2026 / Gendered Stylistic Variation in Brazilian Portuguese Google Play Reviews: A Large-Scale Study

Gendered Stylistic Variation in Brazilian Portuguese Google Play Reviews: A Large-Scale Study

Paper · 2026 · Tiago de Melo

Summary

Research on gender and app-store reviews has mostly used English data. This study examines Brazilian Portuguese at scale. The author collected 76,695,564 pt-BR Google Play reviews of 96 popular apps from 2011 to 2025. Google Play has no gender metadata, so each reviewer's first name was matched against IBGE name frequencies from the 2010 Census. Labels were assigned only when one sex dominated the name. This conservative scheme produced about 22.25M reviews with high-confidence male or female labels. Ambiguous and unknown names were kept as separate categories.

The analysis covered three areas:

  • Style: review length, emojis, letter elongation, repeated punctuation, capitalization, and politeness, laughter and informal markers.
  • Lexical diversity: MTLD.
  • Ratings: the 1–5 star distribution.

Reviews with women's names were clearly more expressive, with roughly 60% more emojis per token and more elongation and repeated punctuation. Review length and MTLD were nearly the same for both groups. Men gave proportionally more 1-star ratings, and women gave more 5-star ratings.

The work offers a Portuguese-language view of digital sociolinguistic behavior. The author cautions that name-based inference does not reflect gender identity and that the results are correlational.

Key points

  • Corpus: 76.7M reviews filtered by the Google Play lang field (pt-BR). Of these, 22,254,455 received gender labels: 12,917,778 male and 9,336,677 female.

  • Gender inference: the first alphabetic token of userName was looked up in IBGE 2010 Census frequencies. Names were labeled female if p_fem ≥ 0.95 and male if p_fem ≤ 0.05. All other names were labeled ambiguous or unknown.

  • Lexical diversity: MTLD (τ = 0.72) was computed on concatenated reviews per group. Values were almost equal, with a slight edge for men. Both groups showed a sharp dip in 2022, followed by partial recovery.

  • Emoji analysis: emoji use was compared with Jaccard overlap of each group's 50 most frequent emojis and the functional taxonomy of Barbieri et al. Associations were measured with odds ratios and Woolf 95% confidence intervals. The ad hoc marker lexicon is to be released with the dataset.

  • Ratings:

    MeasureWomenMen
    5-star share74.21%72.07%
    1-star share10.91%13.84%
    Mean score4.3164.213

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 works

Cite

Cite as Tiago de Melo. 2026. Gendered Stylistic Variation in Brazilian Portuguese Google Play Reviews: A Large-Scale Study. Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 2.

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{melo2026gendered,
  title = {{Gendered Stylistic Variation in Brazilian Portuguese Google Play Reviews: A Large-Scale Study}},
  author = {Tiago {de Melo}},
  booktitle = {Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 2},
  year = {2026},
  url = {https://aclanthology.org/2026.propor-2.32/}
}

References

19 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Francesco Barbieri, German Kruszewski, Francesco Ronzano, and Horacio Saggion. 2016. How cosmopolitan are emojis? exploring emojis usage and meaning over different languages with distributional semantics. In Proceedings of the 24th ACM international conference on Multimedia, pages 531–535. find
  2. Jacek D abrowski, Emmanuel Letier, Anna Perini, and Angelo Susi. 2022. Analysing app reviews for software engineering: a systematic literature review. Empirical Software Engineering, 27(2):43. find
  3. Rafael Dias and Ivandré Paraboni. 2020. Cross-domain author gender classification in brazilian portuguese. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 1227–1234. find
  4. Gustavo Paiva Guedes, Eduardo Bezerra, Lilian Ferrari, and Fellipe Duarte. 2016. Gender differences in the use of portuguese in social networks: Evidence from liwc. In Proceedings of the 22nd Brazilian Symposium on Multimedia and the Web, pages 339–342. find
  5. IBGE. 2016. Um brasil de marias e josés: Ibge apresenta banco de nomes com base no censo 2010. Agência de Notícias do IBGE. Acesso em: 2026-01-13. find
  6. Os Keyes. 2018. The misgendering machines: Trans/hci implications of automatic gender recognition. Proceedings of the ACM on human-computer interaction, 2(CSCW):1–22. find
  7. Moshe Koppel, Shlomo Argamon, and Anat Rachel Shimoni. 2002. Automatically categorizing written texts by author gender. Literary and linguistic computing, 17(4):401–412. find
  8. Philip M McCarthy and Scott Jarvis. 2010. Mtld, vocd-d, and hd-d: A validation study of sophisticated approaches to lexical diversity assessment. Behavior research methods, 42(2):381–392. find
  9. Helena Mihaljević, Marco Tullney, Lucía Santamaría, and Christian Steinfeldt. 2019. Reflections on gender analyses of bibliographic corpora. Frontiers in big Data, 2:29. find
  10. Ehsan Noei, Daniel Alencar Da Costa, and Ying Zou. 2018. Winning the app production rally. In Proceedings of the 2018 26th ACM joint meeting on european software engineering conference and symposium on the foundations of software engineering, pages 283–294. find
  11. Ehsan Noei and Kelly Lyons. 2022. A study of gender in user reviews on the google play store. Empirical Software Engineering, 27(2):34. find
  12. Dennis Pagano and Walid Maalej. 2013. User feedback in the appstore: An empirical study. In 2013 21st IEEE international requirements engineering conference (RE), pages 125–134. IEEE. find
  13. Francisco Rangel, Paolo Rosso, Martin Potthast, and Benno Stein. 2017. Overview of the 5th author profiling task at pan 2017: Gender and language variety identification in twitter. Working notes papers of the CLEF, 48. find
  14. Lucía Santamaría and Helena Mihaljević. 2018. Comparison and benchmark of name-to-gender inference services. PeerJ Computer Science, 4:e156. find
  15. Victor Savicki and Merle Kelley. 2000. Computer mediated communication: Gender and group composition. CyberPsychology & Behavior, 3(5):817–826. find
  16. Mariana Sturzeneker, Maria Clara Crespo, Maria Lina Rocha, Marcelo Finger, Maria Clara Paixão de Sousa, Vanessa Martins do Monte, and Cristiane Namiuti. 2022. Carolina’s methodology: building a large corpus with provenance and typology information. In DHandNLP@ PROPOR, pages 53–58. find
  17. Adile Uka and Maria Berger. 2024. Could style help plagiarism detection?-a sample-based quantitative study of correlation between style specifics and plagiarism. In Proceedings of the 16th International Conference on Computational Processing of Portuguese-Vol. 2, pages 103–108. find
  18. Renata Vieira, Ana Paula Banza, Ana Sofia Ribeiro, Cassia Trojahn, Fernanda Olival, Helena Cameron, Herminia Vilar, Ivo Santos, Joaquim Santos, Maria Gonçalves, and 1 others. 2022. Digital humanities and portuguese processing: a research pathway. find
  19. Barnet Woolf. 1955. On estimating the relation between blood group and disease. Annals of human genetics, 19(4):251–253. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/melo-2026-gendered-stylistic-variation-brazilian-portuguese/