Assessment in Education: Principles, Policy & Practice ( IF 2.7 ) Pub Date : 2021-12-27 , DOI: 10.1080/0969594x.2021.1999209 Melissa R. Hunte 1 , Samantha McCormick 1 , Maitree Shah 2 , Clarissa Lau 1 , Eunice Eunhee Jang 1
ABSTRACT
Children’s oral language proficiency (OLP) is integral for developing literacy skills. Storytelling or retelling is often used by parents and educators to elicit children’s OLP, yet it is less commonly used for assessment purposes. Leveraged by natural language processing and machine learning, this study examined the extent to which computational linguistic and acoustic indices predict human ratings of children’s (n=184 aged 9 to 11) OLP using two story retell stimuli presented in written and aural forms. Human raters scored children’s OLP on five oral proficiency criteria: vocabulary, grammar, idea development, task-fulfilment, and speech delivery, using a 4-point scale, and linguistic and acoustic features were used to predict each criterion. Results showed the efficacy of automated indices to predict human scores of children’s OLP. This study calls for attention to discrepancies in human and machine speech delivery scores and stimulus effects on story retelling performance among children of different language backgrounds.
中文翻译:
研究 NLP 驱动的语言和声学特征在预测儿童口语能力人类分数方面的潜力
摘要
儿童的口语能力 (OLP) 是培养读写技能不可或缺的一部分。父母和教育工作者经常使用讲故事或复述来引出儿童的 OLP,但它不太常用于评估目的。利用自然语言处理和机器学习,本研究使用两种以书面和听觉形式呈现的故事复述刺激,检查了计算语言和声学指数预测儿童(n = 184 名 9 至 11 岁)OLP 的人类评分的程度。人工评估员使用 4 分制根据五个口语能力标准对儿童的 OLP 进行评分:词汇、语法、思想发展、任务完成和言语表达,并使用语言和声学特征来预测每个标准。结果显示了自动化指数在预测儿童 OLP 的人类分数方面的功效。