A Bibliometric Analysis of Peer Assessment in Online Language Courses

Abstract

As a popular strategy in collaborative learning, peer assessment has attracted keen interest in academic studies on online language learning contexts. The growing body of studies and findings necessitates the analysis of current publication trends and citation networks, given that studies in technology-enhanced language learning are increasingly active. Through a bibliometric analysis involving visualization and citation network analyses, this study finds that peer assessment in online language courses has received much attention since the COVID-19 outbreak. It remains a popular research topic with a preference for studies on online writing courses, and demonstrates international and interdisciplinary research trends. Recent studies have led peer assessment in online language courses to more specific research topics, such as critical factors for improving students’ engagement and feedback quality, unique advantages in providing online peer assessment, and designs to enhance peer assessment quality. This study also provides critical aspects about how to effectively integrate educational technologies into peer assessment in online language courses. The findings in this study will encourage future studies on peer assessment in online learning, language teaching methods, and the application of educational technologies.

A meta‐analysis of the effects of augmented reality technologies in interactive learning environments (2012–2022)

“…For specific dimensions and elements in e‐learning environments, Aparicio et al [4] proposed a conceptual framework listing particular elements in people, technologies, and services involved in technological integration. While their original framework adopted these three dimensions for naming elements of each component, Lin and Yu [50] proposed influencing factors for technological integration. Characteristics of technologies, participants, and learning contents could impact the effectiveness of educational technologies [50], which corresponded to the three components in Aparicio et al's framework [4].…”

“…While their original framework adopted these three dimensions for naming elements of each component, Lin and Yu [50] proposed influencing factors for technological integration. Characteristics of technologies, participants, and learning contents could impact the effectiveness of educational technologies [50], which corresponded to the three components in Aparicio et al's framework [4].…”

“…Different studies adopted diverse outcome indicators, such as knowledge acquisition outcomes, emotions, and skill performances. With the interdisciplinary trends of educational technology studies [50], effectiveness measurements have been introduced from multiple research areas. Applying augmented reality has become an emerging trend since the 2020s [21], and the application contexts broadly diversified.…”

A bibliometric analysis of artificial intelligence chatbots in educational contexts

“…Figure 1 demonstrates the theoretical model applied to this study, which was extended from Aparicio et al ’s holistic framework of e-learning in 2016 to analytical dimensions and influencing factors of the three components. As proposed by Lin and Yu (2023), functional characteristics of technologies, technological advancements and interdisciplinary influencing factors should be considered. According to these three dimensions, we would first consider technological characteristics, functions and how the effectiveness could be evaluated.…”

Modelling Sign Language with Encoder-Only Transformers and Human Pose Estimation Keypoint Data

“…Along with a general increase in cultural and disability awareness, and focus on improving accessibility for Deaf communities, the research community started to embrace the multimodal aspects of sign language as well as the needs of the people it ostensibly intends to help [24]. At the very least, this amounts to a recognition that sign language goes beyond hand gestures [25,26], which has led to models increasingly utilising the whole signing space [27] and not just data derived from hands-although this practice does still continue largely via wearable technology e.g., [1,[28][29][30][31][32][33][34][35][36], which has long been met with disdain by the Deaf community [37,38]. As with models that derive input data from video feeds that cover the signing space, those that incorporate data from human pose estimation keypoints can also inherently use the extra information available, though to what extent depends on the keypoints being used.…”