Introduction

Second-language instruction has seen a renewed dedication to both communicative abilities and discrete language skills. While speaking teachers do well to include communicative activities in each class session, they are also encouraged to focus on particular features of spoken language in order to improve students’ intelligibility and comprehensibility in speech (Levis & Grant, 2003). These features include segmentals (e.g., vowel and consonant production), and suprasegmentals (e.g., word level stress and intonation). Along with a targeted focus on pronunciation, classrooms also feature opportunities for learners to take responsibility for their own learning (Lazarton, 2001). Students can apply strategies and techniques to their own learning styles and can control other aspects of their own learning. In addition, students are much more likely to be motivated when they have a voice in the evaluation of their progress. By allowing students to exercise choice in learning their second language, teachers can improve students’ chances for success.

As a result of the need to address discrete skills in pronunciation and an increasing desire to involve students in their own learning, it has been suggested that teachers engage students in self-monitoring through the use of video (Celce-Murcia, Brinton, Goodwin, & Griner, 2010). This self-monitoring can occur outside the classroom and be directed with specific guidelines. It can also involve transcription in order to identify pronunciation errors. However, the actual practice of self-evaluation through video has yet to be empirically investigated as a means for improving the pronunciation ability of second language learners. For this reason the current study aimed to investigate to what extent students’ focused self-monitoring strategies improve their oral performance.

Method

The study was conducted by comparing the improvement of two sets of oral presentations (presentation skills and pronunciation) between am experimental group and a control group. Students were assigned to the control group (i.e., no self-evaluation treatment) or the experimental group (i.e., with self-evaluation treatment) depending on the class section in which they were placed by the intensive English program (IEP) administration at a Southwestern American University. Students included 27 full-time ESL learners (L1 of Arabic or Chinese) between the ages of 18 to 25 (21 males and 6 females) from two intact advanced listening and speaking classes (6 hours per week). Based on the in-house placement test, these students had high-intermediate to advanced levels of proficiency in English..

Based on the IEP curriculum, two sets of oral presentations (pre- and posttests) were given and video-recorded in Weeks 8 and 15 of the 15-week semester. To prepare, both the control group and the experimental group received two sets of 20-minute lessons inserted into their IEP classes: one in Week 7 focusing on presentation skills (hand gestures and eye contact) and another in Week 14 focusing on pronunciation (sentence stress and lexical stress). For each set of presentations, both groups gave a practice talk (pretest) after the lesson and received written feedback from their teachers. Participants in the experimental group also watched the videos of their own speech (posted on YouTube) and self-evaluated their performances using forms provided; the control group did not do so. Finally, participants of both groups gave a final presentation as the posttest.

The rating included two types of measurements. The first measurement involved two analytical rubrics scoring presentation skills and pronunciation. Each of these rubrics featured three separate sub-criteria (1 to 5). These sub-criteria scores were summed and averaged to create a final composite score for both constructs of presentation skills and pronunciation. Final composite scores ranged from 1 to 5 and were utilized for statistical analysis. The second measurement included two sets of 7-point Likert Scales for comprehensibility and oral proficiency. The presentations were randomly ordered and coded for both groups so that raters were not aware of the nature of each speech. Each presentation was rated by two linguistically-trained raters separately and high inter-rater reliability was achieved (Pearson Correlation Coefficients were between .79 and .98 for all constructs rated).

Qualitative data were also collected via self-evaluation forms provided to student participants during the study. Answers to both multiple-choice questions regarding use of suprasegmental features and open-ended questions regarding personal views on improving use of suprasegmentals were analyzed in order to highlight patterns among participants’ self-evaluation.

Results

Results indicated a more general effect in the experimental group. Students who engaged in self-monitoring showed substantial improvement from pretest scores in oral proficiency during week 14 to posttest scores in week 15. This group also showed more long-term improvement (i.e., week 7 to week 15) for oral proficiency scores. However, improvements in measures of comprehensibility, pronunciation, and presentation skills were not noticeable. The open-ended responses from experimental group participants on self-monitoring forms for pronunciation helped to shed light on the manner in which students internalized the self-monitoring process. Responses were often off-topic or quite simple. For example, when responding to the question, “What ways can you change your stress to help your presentation,”one student responded, “Speak little more.” Another student responded, “Know the background.” However, particular individual responses helped to highlight the more strategic approach some students utilized in self-evaluating speech performance for use of suprasegmental features. In response to the same question, another student responded, “To choose a few word to stress on it.” Qualitative and quantitative results conjunctively provided insight into the lack of direction students may have had during self-monitoring. Still, with the guidance provided, the experimental group improved greatly in comparison to the control group in oral proficiency.

The results suggest that improving pronunciation and presentation skills can be quite challenging when using self-monitoring as a tool for instruction. Additionally, the effects of self-monitoring on oral proficiency indicated that self-monitoring may provide stronger effects when utilized over a longer period of time. Furthermore, student responses showed evidence of a need for more structured and guided training in self-monitoring. These responses provided a mixed message. Some students showed an ability to analyze their speech through more specific responses, while most students provided more general responses. Overall, student responses indicated a raised awareness of suprasegmental features in pronunciation.

Implications

Important implications for future research resulted from this research. The focus of the current study centered on suprasegmental features of English. Though these features are considered highly important in determining the intelligibility of an L2 speaker, segmental features might also be of interest. Investigating the effect of self-monitoring for these features might contribute greatly to a more in-depth understanding of self-monitoring for pronunciation. For example, a study might investigate the extent to which students’ self-monitoring for vowel accuracy affects their oral performance. Additionally, there is a clear need to incorporate more training during the treatment portion of the study. Because students indicated a lack of understanding of how to engage in self-monitoring, it may be beneficial to incorporate sessions for students to be oriented to the nature of self-monitoring and, particularly, to the language features in question.

Along with implications for further research, implications for instruction also emerged. Through our research, it became clear that the task of identifying issues with features like sentence stress and word stress are not necessarily easy for students. Teachers must acknowledge the demand these self-monitoring tasks place on their students. However, students’ inability to thoroughly self-monitor their own pronunciation of suprasegmental features also helps to highlight the need for teachers to engage students in practices with evaluation for these features before students can be expected to successfully self-evaluate on their own. Simply providing guidelines via a worksheet does not suffice. Consequently, teachers can develop materials for in-class activities that require students to develop skills in self-monitoring such as videos of prior student presentations, group discussions about the strengths and weaknesses of a particular student’s segmental or suprasegmental pronunciation, and guided practice evaluating one’s own pronunciation during class time. These measures will likely improve the chances for students’ successful self-monitoring both during class and after. Though training for self-monitoring must occur, our results indicate that simply beginning the process of self-monitoring can help to improve students’ abilities in oral performance.

References

Celce-Murcia, M., Brinton, D., Goodwin, J., & Griner, B. (2010). Teaching pronunciation: A reference and course text (2nd ed.). Cambridge: Cambridge University Press.

Lazarton, A. (2001). Teaching oral skills. In M. Celce-Murcia (Ed.), Teaching English as a second or foreign language (pp. 103–116). Boston, MA: Heinle & Heinle.

Levis, J. M., & Grant, L. (2003). Integrating pronunciation into ESL/EFL classrooms. TESOL Journal, 12, 13–19.

Nicholas Velde is a current graduate from the MA TESL program at Northern Arizona University. He teaches as an English instructor at Melikşah University in Kayseri, Turkey and is interested in vocabulary and pronunciation research, strategy instruction, and task-based language teaching.

Yuan Zhuang is a doctoral student in the program of applied linguistics at Northern Arizona University. She received her MA in TESOL from Missouri State University and MA in English language and literature from Liaoning Normal University, China. Her research interests include L2 pronunciation and phonology, ESL listening and speaking, SLA, and translation and social culture.

Okim Kang, PhD is assistant professor of applied linguistics at Northern Arizona University. Her research focuses on L2 pronunciation, language attitudes, oral proficiency assessment, and ITA issues. She is the winner of the TOEFL Outstanding Young Scholar 2013 award from Educational Testing Service.