Related references
Note: Only part of the references are listed.
Article
Psychology, Educational
Gary J. Ockey et al.
Summary: Most second language assessment researchers agree that interactional competence (IC) is crucial in oral communication assessment. However, assessing IC is challenging due to the need for an interlocutor. Using a Spoken Dialogue System (SDS) as a test taker's partner could mitigate these challenges. A study comparing an SDS with a human peer partner found that the SDS condition resulted in more ratable IC features, lower scores for most IC features, and more positive perceptions of the rating process.
LANGUAGE ASSESSMENT QUARTERLY
(2023)
Article
Psychology, Educational
Barry O'Sullivan
Summary: This paper highlights the rapid changes in technology and partial validation efforts as issues of concern, and suggests important considerations in terms of validation. The author predicts that technology will bring radical changes to the practice of language testing.
LANGUAGE ASSESSMENT QUARTERLY
(2023)
Article
Psychology, Educational
Daniel R. Isbell et al.
Summary: The application of remote proctoring in language testing raises important issues of fairness and justice, including concerns about construct-irrelevant responses, technological biases, and access to suitable technology and physical space for remote proctoring.
LANGUAGE ASSESSMENT QUARTERLY
(2023)
Article
Psychology, Educational
L. Hannah et al.
Summary: Using machines for scoring writing has significant implications for formative assessment, but the validity of machine scoring in the K-12 context needs further research.
LANGUAGE ASSESSMENT QUARTERLY
(2023)
Article
Computer Science, Artificial Intelligence
Yigal Attali et al.
Summary: This paper presents an interactive reading task approach based on transformer-based deep language modeling for generating reading comprehension assessments. Through a large-scale pilot test, the feasibility of this approach for automatic creation of complex educational assessments has been demonstrated.
FRONTIERS IN ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Debaditya Chakraborty et al.
Summary: This study compared the predictive capabilities of interpretable and noninterpretable machine learning models, revealing that tree-based ensemble models can perform similarly to deep learning models in structured hydro-climatological datasets. Using a newly developed sequential transfer-learning technique, the tree-based ensemble model was able to impute missing climate data at various levels. The eXML framework quantified the global importance of hydro-climatic variables and identified transition points of climate variables for daily ETo rates.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Education & Educational Research
Jing Xu et al.
Summary: This study found that the reliability of the automarker in online oral English test is good, but it tends to be more lenient towards low-proficiency speakers. The uncertainty measure named Language Quality, which indicates the confidence of speech recognition, was found to be useful in predicting reliability and identifying abnormal speech.
ASSESSMENT IN EDUCATION-PRINCIPLES POLICY & PRACTICE
(2021)
Article
Linguistics
Xiaoming Xi et al.
Article
Computer Science, Artificial Intelligence
L Breiman
