An examination of the effectiveness of embedded audio feedback for English as a foreign language students in asynchronous online discussions
This study examined the effect of asynchronous embedded audio feedback on English as a Foreign Language (EFL) students' higher-order learning and perception of the audio feedback versus text-based feedback when the students participated in asynchronous online discussions. In addition, this study examined how the impact and perceptions differed when the instructor providing the feedback was a nonnative English-speaking teacher (NNEST) versus native English-speaking teacher (NEST) (Pasternak & Bailey, 2004). A quasi-experimental design was used with audio feedback and text-based feedback as a within-subject factor, instructors' language background (NNEST and NEST) and students' level of language proficiency (high and low) as the between-subjects main factors. The students were assigned to the levels of language proficiency (high and low) and two types of instructors (NNEST and NEST), but all of them experienced audio feedback and text-based feedback. ^ To accomplish this, an examination of the students' weekly online postings across the three time periods (pretest, posttest 1, and posttest 2) and the perceptions of the technique were carried out. Two instruments were used to examine the effect of embedded audio feedback (a) the scoring rubric (Ertmer & Stepich, 2004), and (b) the audio feedback survey to examine students' responses to audio and text-based feedback (Ice, 2008). Specifically, for this study, the EFL students' weekly scores indicating the quality of online discussion posting for audio feedback and text-based feedback delivery methods and their perceptions on the survey were used as dependent variables. The three independent variables of this study were: (a) students' level of language proficiency; (b) embedded audio feedback versus text-based feedback; and (c) nonnative (NNEST) or native English-speaking (NEST) instructors who were providers of feedback. The quantitative data were analyzed with descriptive statistics, logistic regression analysis, a Wilcoxon Signed Rank Test, an independent t-test, a mixed-effect ANOVA, and the two-way between-groups ANOVA. The results indicated the effectiveness of audio feedback and text-based feedback to promote EFL students' higher-order learning and to increase perceived effectiveness of both types of feedback. The results also indicated that there were no significant differences between the groups (NNEST and NEST) and the student's levels of language proficiency (high and low) on the increased quality of the students' online postings and their perceptions of audio feedback. However, the effect of audio feedback on the quality of online posting was different because it depends on the students' level of language proficiency. In this study, the students at the low level of language proficiency were more likely to drop the course and/or received the low scores on their online postings. However, the students at the low level of language proficiency perceived that the audio feedback helped them retain the course information more than the text-based feedback. Finally, the students in the NEST group perceived higher motivation and retention than the students in the NNEST group. The study has implications for instructors and designers in creating online learning environments as it relates to asynchronous online discussions that include EFL students.^
Jennifer C. Richardson, Purdue University.
Education, English as a Second Language|Education, Instructional Design|Education, Technology of