The Relative Contribution of Language Complexity to Second Language Video Lectures Difficulty Assessment

Abstract

Although core in the teaching of academic language skills, little research to date has investigated what makes video-recorded lectures difficult for language learners. As part of a larger program to develop automated videotext complexity measures, this study reports on selected dimensions of linguistic complexity to understand how they contribute to overall videotext difficulty. Based on the ratings of English language learners of 320 video lectures, we built regression models to predict subjective estimates of video lecture difficulty. The results of our analysis demonstrate that a 4-component partial least square regression model explains 52% of the variance in video difficulty and significantly outperformed a baseline model in predicting the difficulty of videos in an out-of-sample testing set. The results of our study point to the use of linguistic complexity features for predicting overall videotext difficulty and raise the possibility of developing automated systems for measuring video difficulty, akin to those already available for estimating the readability of written materials.

Enriching Library Holdings for English Language Learners: Applying Traditional Readability Formulas and Modern Cohesion Methods to Film

“…For example, traditional formulas were largely designed for native English‐speaking audience but have been adopted to serve ELLs. Further, the modern methods RDL2 and CEFR Complexity Score along with recent developments by Alghamdi et al (2022), while designed for ELLs, group all ELLs irrespective of linguistic background into one group regardless of possible user background influences. Moreover, both traditional measures and recent advancements often neglect user aspirations and motivations in ELL development.…”

What makes video‐based academic lectures difficult for language learners to comprehend? The role of multimodal complexity

“…The findings showed that lexical complexity was significantly associated with L2 learners' perception of video difficulty and that lexical frequency alone explained approximately 40% of the variance in learners' assessment of video lecture difficulty. Using a wide range of acoustic, phonological, lexical, syntactic, and discourse complexity indices, Alghamdi et al (2022) developed a partial least squares regression model to predict video difficulty. The results obtained demonstrated that a significant proportion of the variance reported in terms of video lecture difficulty ( R 2 = 0.52) could be explained in terms of acoustic, lexical, and syntactic complexity.…”

“…By investigating the relationship between lexical complexity (frequency, density, and diversity) and language learners' perception of video difficulty, Alghamdi et al (2023) found that lexical frequency was a strong predictor of instructional video difficulty. In particular, Alghamdi et al found that B1 English language learners perceived instructional videos as being more difficult if they contained many words that were not among the first 3000 most common words in the British National Corpus (BNC)/Corpus of Contemporary American English (COCA) list.…”

Leveraging ensemble machine learning and multimodal video complexity for better prediction of video difficulty in second language

“…In a recent study, Alghamdi et al (2022) examined the role of variation in pitch, speech rate, articulation rate, pausing, and phonotactic probability on video lecture difficulty assessment. Results of their study showed that only variation in pitch was a significant predictor of video lecture difficulty and that video difficulty decreases when the speakers have high pitch variation.…”

“…One way to help determine the appropriateness of videos for language learners is by assessing video content difficulty. While numerous predictive measures of content difficulty have been developed for written and spoken learning materials, little research has been done on video content difficulty prediction (but see Alghamdi, Gruba, Marsi, & Velloso, in press;Alghamdi, Gruba, & Velloso, 2022). The primary objective of this study is to investigate the usefulness of ensemble machine learning approaches for predicting video lecture difficulty.…”