Knowing ‘help’ does not guarantee knowing ‘helpless’, vocabulary study of Chinese and Japanese learners finds

Knowing the English word help does not reliably mean knowing helpless, according to a new study of 427 university-level English as a foreign language learners in China and Japan. The research, published in the Asian-Pacific Journal of Second and Foreign Language Education, tested how well learners could recall the meanings of base words and their derived forms within the same word family, and found that knowledge of a base word extended to its derivational forms in only about two thirds of cases overall.

The study was conducted by Joshua Matthews, Brett Milliner, Yusen Wang and Kriss Lange, who administered a short 24-item meaning-recall vocabulary test to tertiary-level learners: 277 from a large university in south-western China and 150 from one rural and two metropolitan universities in Japan. The test covered six high-frequency word families built on the verbs help, accept, collect, develop, operate and use. For each family, learners were asked to translate the base word and three derivational forms into their first language, with the derived forms sampled from high, mid and low frequency ranges of the British National Corpus. An example from the help family included helpful as a high-frequency form, helpless as a mid-frequency form and helplessness as a low-frequency form.

The meaning-recall format required learners to retrieve and express word meanings without recognition cues. Each test item presented the target word in a simple sentence offering no semantic clues, and test takers had to produce an acceptable translation in Chinese or Japanese. Responses were scored as correct or incorrect by trained bilingual raters, with a secondary rater independently double-scoring a random 20 percent of the data. Inter-rater reliability was Cohen’s kappa of 0.71 for the Japanese dataset and 0.88 for the Chinese dataset, and internal consistency for the combined dataset was acceptable, with Cronbach’s alpha of 0.84.

The central measure was the conditional probability that a learner knew a derivational form given knowledge of the corresponding base word, calculated as the proportion of participants who knew the derived form among those who knew the base. Across the 18 base-derivational pairs, the overall mean conditional probability was 0.663. For the Chinese subgroup it was 0.681, and for the Japanese subgroup 0.623. In no case was a derivational form more consistently known than its corresponding base word, a directional asymmetry the authors interpret as evidence that knowledge flows from simpler to more complex forms rather than the reverse.

Frequency emerged as a strong organising pattern. Across the full sample, the probability of knowing a derivational form given knowledge of its base word was highest for high-frequency forms, at a mean of 0.87, and fell to 0.62 for mid-frequency forms and 0.50 for low-frequency forms. The pattern held in both groups: Chinese learners showed means of 0.88, 0.65 and 0.52 across the high, mid and low bands, while Japanese learners showed 0.84, 0.56 and 0.46. The lowest individual probabilities among Chinese learners were observed for low-frequency forms such as collectable, usable, helplessness and inoperable, where the conditional probability ranged from 0.29 to 0.54, meaning that base-word knowledge frequently did not extend to these items.

Not every item followed the frequency trend. Among Japanese learners, the pairs collect/collection and collect/collector reached conditional probabilities of 0.901 and 0.975 respectively, well above the group’s overall average. The authors suggest that established loanword forms in Japanese, such as borrowed versions of collection and collector, may help explain this stronger performance, although they stress that this interpretation is tentative and based on limited retrospective inspection of learner responses rather than a systematic examination of loanword effects.

The second research question compared the two learner groups directly. Using a mixed-effects logistic regression model that accounted for the repeated-measures structure of the data, with participant as a random intercept and Tukey-adjusted pairwise comparisons, the researchers found statistically significant differences between Chinese and Japanese learners for 15 of the 18 derivational forms. Chinese learners showed higher meaning-recall knowledge for acceptable, unacceptable, acceptability, redevelop, helpful, helpless, helplessness, usage, inoperable and operational, while Japanese learners outperformed on collection, collectable, collector, developer and usable. No significant differences emerged for development, operation or user.

These group differences should be read with caution. The study did not include a shared external proficiency measure, and scoring procedures differed across the two languages. Teacher estimates and locally administered vocabulary levels tests suggested the Chinese participants were at approximately the B2 CEFR level, with an estimated average text-range vocabulary size of about 3,677 word families, while the Japanese participants were at approximately B1, with about 3,168 word families. Because different test instruments were used in the two contexts, these estimates are approximate indicators rather than directly comparable measures. The authors therefore treat the cross-group differences as descriptive patterns within the present dataset rather than definitive evidence about underlying causes, which may include overall proficiency as well as L1-specific influences.

Retrospective inspection of erroneous responses offered tentative diagnostic insights. When translating developer, a notable number of Chinese participants added suffixes analogous to the English -er to the translated base word, producing incorrect forms rather than the conventional translations. For usable, many Chinese participants did not distinguish it from useful. The authors argue that such patterns illustrate how formative meaning-recall tests can reveal recurring misconceptions that base-word testing alone would obscure, giving teachers material for targeted instruction, such as explicitly addressing the difference between usable and useful.

The pedagogical upshot is that testing only base words can give a misleading picture of learners’ lexical knowledge. If a learner knows the base form of a high-frequency word, teachers can be more confident, though not certain, that the learner also knows related high-frequency derivational forms. For mid- and low-frequency derivations, however, the overlap is substantially weaker, so these forms deserve direct assessment and instruction. The authors note limitations: frequency is not fully separable from item-level characteristics such as semantic transparency, some items like usage and inoperable are not consistently classified as members of the same word family across lexical resources, and the operate family showed consistently low knowledge for both groups, possibly because forms such as operational and inoperable are semantically opaque, though this explanation remains speculative. Whether the findings generalise to other EFL contexts is a matter for future research, and the complete test instruments and scoring criteria are available as supplementary materials for classroom use.

Subject of Research: Base-word and derivational form vocabulary knowledge among Chinese and Japanese EFL learners

Article Title: Is knowing help the same as knowing helpless? A formative vocabulary assessment of Chinese and Japanese EFL learners’ knowledge of derivational forms within word families

Article References: Matthews, J., Milliner, B., Wang, Y., & Lange, K. (2026). Is knowing help the same as knowing helpless? A formative vocabulary assessment of Chinese and Japanese EFL learners’ knowledge of derivational forms within word families. Asian-Pacific Journal of Second and Foreign Language Education, 11(1), Article 89. https://doi.org/10.1186/s40862-026-00467-9

Image Credits: AI Generated

DOI: 10.1186/s40862-026-00467-9

Keywords: vocabulary acquisition, derivational morphology, word families, EFL learners, meaning-recall assessment, formative assessment, conditional probability, frequency effects, Chinese learners, Japanese learners, second language vocabulary, CEFR proficiency