본문 바로가기
  • Home

How Translation Discourse Changes across Goodreads Ratings: Computational Evidence from Korean Literature in English

  • The Journal of Translation Studies
  • Abbr : JTS
  • 2026, 27(3), pp.169~191
  • DOI : 10.15749/jts.2026.27.3.006
  • Publisher : The Korean Association for Translation Studies
  • Research Area : Humanities > Interpretation and Translation Studies
  • Received : August 4, 2026
  • Accepted : September 15, 2026
  • Published : September 30, 2026

Chang-Soo Lee ORD ID 1

1한국외국어대학교

Accredited

ABSTRACT

This study examines how readers discuss translation in Goodreads reviews of Korean literature translated into English and how these discussions vary across star ratings. The dataset comprises 143,806 reviews of 44 translated Korean literary works. Reviews explicitly referring to translation or translators were identified through three node families (TRANSLATION, TRANSLATE, and TRANSLATOR). Binomial logistic regression showed that the frequency of translation discourse differed significantly across ratings: translation was mentioned in 12,855 reviews (8.9%), most often in two-star reviews, with its proportion declining steadily from two to five stars. Rating-specific collocational networks further revealed a corresponding shift in lexical associations. Lower-rated reviews more often linked translation to uncertainty, loss, inadequacy, and linguistic difficulty, whereas higher-rated reviews increasingly associated it with fluency, aesthetic appreciation, successful mediation, and named translators. These patterns were gradual rather than categorical, as criticism remained present in highly rated reviews and positive evaluations also appeared in lower-rated ones. The findings show that readers' overall evaluations are associated both with whether translation is mentioned and with the language used to discuss it, while confirming that Goodreads star ratings cannot be treated as direct measures of translation quality or translation-focused evaluation.

Journal Copyright Policy

No CCL information provided

Citation status

* References for papers published after 2025 are currently being built.

This paper was written with support from the National Research Foundation of Korea.