Analysis of Translation Technique and Quality of Recruitment Terminology in LinkedIn's Automatic Translation of Job Vacancy Posts

This study analyzes the translation technique and quality of recruitment terminology in LinkedIn's automatic translation of English job vacancy posts into Indonesian. It aims to identify the distribution of recruitment terminology across functional categories, analyze the translation techniques applied by LinkedIn's Neural Machine Translation (NMT) feature, and evaluate the resulting translation quality in terms of accuracy, acceptability, and readability. This study employed a qualitative descriptive design using content analysis and Focus Group Discussion (FGD). The data consisted of 103 recruitment terminology units collected from English-language job vacancy posts published by companies operating in Indonesia on LinkedIn between April and May 2026. Recruitment term categories were classified based on Herold and Roedenbeck's (2025) framework, translation techniques were analyzed using Molina and Albir's (2002) classification, and translation quality was assessed using Nababan, Nuraeni, and Sumardiono's (2012) Translation Quality Assessment model. The findings show that HR Management is the most dominant recruitment term category (53.4%), Established Equivalent is the most frequently applied translation technique (64.6%), and an original finding of this study is a combination technique in which two techniques operate simultaneously within a single term, identified in 13 data units. The overall translation quality obtained a mean score of 2.59, categorized as High, although acceptability received the lowest score among the three quality dimensions. The findings suggest that translation quality is closely related to how standardized a term's professional domain already is in Indonesian professional discourse, while more technical or internationally circulated categories remain more vulnerable to quality variation.