Contextualised judgements: A comparison of the rating criteria used to judge oral presentations in higher education and speaking performances in the TOEFL iBTTM 
Ana Maria Ducasse, RMIT University, Melbourne, Australia
Annie Brown,  ACER, Camberwell, Australia
https://doi.org/10.58379/JWID1797
Volume 9, Issue 2, 2020
Abstract: This study investigates assessment within oral academic assessment tasks, specifically focusing on the criteria used by discipline specialists and comparing them with those used to assess performance on TOEFL iBTTM speaking tasks. Three pairs of tutors from three faculties took part in verbal report sessions where they watched, rated and discussed the performances of ten native and ten non-native students completing first-year university oral assessment tasks in their discipline. The verbal report sessions were audio recorded, transcribed and segmented into meaning-based units prior to thematic analysis. The features which emerged from the analysis were compared with those described within the TOEFL speaking rubrics. The analysis found that while there were some broad similarities in the focus there were also marked differences. Two of the three TOEFL strands (delivery and content) were well-represented in the academic tasks assessments rubrics and tutor discussion. However, the quality of the non-native students’ language was only of concern when it was perceived as interfering with the student’s ability to communicate. An additional focus in the assessment of university tasks was the use of academic skills, prompts and aids, non-verbal communication and engagement with the audience. ​
Keywords: TOEFL iBTTM speaking, speaking rubrics, oral assessment tasks, academic skills, non-verbal communication