The purpose of this study was to investigate how an instructor's accent influences students' learning achievement. Furthermore, this study also explored how students' accent preference may affect their learning. Unlike native voices, accented voices were not natural to the native speakers; therefore, it required more cognitive resources for processing the information, compared to native voice, which reduces the quality of students' learning experience (Mayer, Sobko, & Mautone, 2003). However, this did not explain how students' accent preference might influence their learning achievement. In order to address this unique and challenging issue, the research needed to compare the students' accent preference to their achievement scores by determining at what level of accent the non native voice causes an increase in cognitive load.
The study was experimental research. The study had three parts; survey, instruction, and assessment. Before the experiment, participants completed a short survey about their general knowledge of statistics, familiarity with multimedia learning, and accent perceptions. During the experiment, participants were randomly assigned to view a short multimedia instruction explaining how to use the software program, SPSS (Statistical Package for the Social Sciences). The instructions were given in either a native voice or in one of four different accented voices (mild and heavy European accents and mild and heavy Asian accents). After listening to the instructions, participants completed a short assessment. The average completion time, including survey, instruction, and assessment, was 25 minutes. The quantitative data were analyzed by both One-way ANOVA and Two-way ANOVA. Of the 192 participants, 187 were undergraduate students and 5 were graduate students, all from Ohio University.
The research found that; (a) there was no significant difference between students' achievement scores when given native voice instruction and those with accented voice instruction; (b) students' perception toward accent did not influence their learning, in general; (c) only students who, prior to the instruction, said they disliked Asian accents and received instruction from an instructor with an Asian accent showed lower assessment performance (this was not the same for European accents); (d) there was a prior knowledge effect in both the native and European accented instructional voice groups, but not with the Asian accented instructional voice group; (e) the duration of time for completion of the study is influenced with (“lower” or “higher”) achievement scores in the European and Asian accented voice groups, but not with the native voice group; (f) according to the instructor evaluation rating report, not only were the native voice rating scores significantly higher than both the European and Asian voice ratings, but also, European voice ratings were significantly higher than Asian voice ratings.
A possible explanation for the no significant difference among different accented voice instructions is listening adaption. Unlike the previous study from Mayer and his associates, in which it took 140 seconds for the instruction, this study instruction time was 10 to 17 minutes. Furthermore, the combination of social identification and having greater experience with Asian instructors might have accounted for the difference between European and Asian accents on a student's learning. The most challenging aspect and notable limitation of this study was that the instruction was scripted rather than being recorded from a natural instructional setting; therefore, all other aspects of communication, such as grammar, vocabulary, and speaking style were excluded.