Examining HSK 3.0 Before Implementation: A Validity-Oriented Review of Linguistic Features and Task Structure

Abstract

The Hanyu Shuiping Kaoshi (HSK; 汉语水平考试, “Chinese Proficiency Test”) has long served as one of the most widely used standardized tests of Chinese language proficiency for non-native speakers (Peng et al., 2021; Teng, 2017; Wang, 2018). Since its introduction in the 1980s, the test has undergone several revisions in response to changing views of language ability, proficiency standards, and assessment design (Teng, 2017). The current version, HSK 2.0, was released in 2009 and implemented in 2010, consisting of six levels, which its developers claimed aligned with the Common European Framework of Reference for Languages (CEFR) from A1 to C2 (Hanban, 2010). The main test assesses listening and reading at all levels and adds writing from HSK 3 onward, while speaking is offered through a separate, optional test, the HSKK (Hanyu Shuiping Kouyu Kaoshi; 汉语水平口语考试, “Chinese Proficiency Speaking Test”). The listening and reading sections rely predominantly on selected-response item formats, while writing tasks range from sentence construction at lower levels to short topic-based responses at higher levels. Over time, a growing body of research has raised concerns about aspects of HSK 2.0, including the interpretability of its proficiency levels, its representation of communicative ability, and its usefulness for broader score-based decisions (Fu et al., 2014; Peng et al., 2021; Su & Shin, 2015; Teng, 2017; Wang, 2018). These concerns have kept questions of test validity central to discussions of Chinese proficiency assessment.