The Accuracy and Consistency of Machine Translation Engines: Case Study of Translating Economic & Statistical Texts from English into Arabic

Abstract

Objectives: This study assesses the precision and consistency of the Google Translate engine and tool in translating statistical and economic texts from English into Arabic. Methods: The descriptive analytical approach was employed in this project, and the study sample consists of five texts extracted from statistical and economic research. Results: The study revealed that the Google translation tool faced numerous challenges and contained lexical, grammatical, and consistency errors. Besides, a grand challenge might arise from the idioms and abbreviations, as Google Translate would use word-for-word translation and render them inaccurate meaning. In some cases, the abbreviations in the source text (ST) would be displayed as they are in the target text (TT), In other cases, Google Translate would do a deep search to render their meanings in the TT. Further, Google Translate has a significant problem rendering technical and specialized terms into the TT, and in some cases, it doesn’t consider the contexts and has a lexical choice challenge. Finally, Google Translate renders some terms with different meanings within the same text. Conclusions: Idioms and abbreviations present a grand challenge, as Google Translate would use word-for-word translation and render them in an inaccurate meaning. In some cases, the abbreviations in the ST are rendered as they are in the TT, and in other cases, Google Translate would do a deep search to render their meanings in the TT.