Open AccessAn Examination of the New General Service List
Author(s) -
Tim Stoeckel
Publication year - 2019
Publication title -
vocabulary learning and instruction
Language(s) - English
Resource type - Journals
eISSN - 2187-2759
pISSN - 2187-2767
DOI - 10.7820/vli.v08.1.stoeckel
Subject(s) - coca , word (group theory) , interim , computer science , selection (genetic algorithm) , word lists by frequency , service (business) , word list , american english , section (typography) , natural language processing , artificial intelligence , linguistics , history , class (philosophy) , marketing , philosophy , business , archaeology , sentence , operating system
The New General Service List (NGSL; Browne, Culligan, & Phillips, 2013b) was published on an interim basis in 2013 as a modern replacement for West’s (1953) original General Service List (GSL). This study compared GSL and NGSL coverage of a 6-year, 114-million word section of the Corpus of Contemporary American English (COCA), and used COCA word frequencies as a secondary data source to identify candidates for addition to the NGSL. The NGSL was found to provide 4.32% better coverage of the COCA than the GSL. Moreover, several candidates were identified for inclusion to the NGSL: three are current members of the NGSL’s companion list, the New Academic Word List (Browne, Culligan, & Phillips, 2013a); five are words whose usage has increased in recent years; and five are individual types that appear to have been miscategorized during the original development of the NGSL. Because NGSL word selection was based on not only empirical but also subjective criteria, the article calls for the addition of annotations to the NGSL to explain decisions regarding low-frequency NGSL constituents and high-frequency non-constituents.