
R. L. Lu et al.
pants were a class of sophomores enrolled in a teacher education program and seeking certification to teach
English at elementary level (pre-service English teachers). Twenty-five students in this class were randomly as-
signed to the experimental group while the other 20 students were assigned to the control group.
Native English speakers who served as peer English instructors for this project were selected from the College
of Coastal Georgia in the United States. Selection of the peer instructors were conducted through an application
process that was targeted towards pre-service teachers seeking certification in English or language arts. Five in-
structors were chosen at random upon initial screening of basic qualifications.
The 25 Chinese students that were identified in the experimental group were randomly assigned into groups
of five and assigned to a single American peer instructor. However, interaction and instruction occurred on a one
to one basis within the group.
2.3. Procedure
The treatment of this study consisted of a 10-week educational project that includes weekly synchronous video-
conferencing sessions between Chinese pre-service English teachers and their American peers. The Chinese
project participants were assigned a speech topic for each conferencing session which was conducted and eva-
luated in a similar manner to the IELTS (International English Language Testing System) oral examination.
On each videoconferencing session, American peer instructors “called” their cohort of 5 Chinese students.
The live video conferencing sessions had scripted open ended prompts/questions for the American instructors to
ask of their Chinese students. These sessions were recorded and later evaluated by the project personnel. Imme-
diate peer feedback on how well each student performed during the conversation and how to improve oral skills
were provided by the American instructors to their Chinese students. These live video conferencing sessions
took place once a week. Each session lasted 100 minutes (equal to one block course schedule), and every Chi-
nese participant had approximate 20 minutes to talk/converse with their American peer instructors.
While the experiment group was having live videoconference with their American peers, the control group
took the English speaking course delivered by the course instructor in a regular classroom setting, which also
lasted 100 minutes. The topic of each class session was same as that of the experiment group.
Both groups were pre- and post tested on their oral English skills in four dimensions: Fluency & Coherence,
Lexical Resource, Grammatical Range & Accuracy, and Pronunciation. Additionally, attitudinal surveys on par-
ticipants’ confidence in various aspects of teaching English language in future were administered to all partici-
pants at the end of the treatment to determine if significant difference occurred between the groups.
2.4. Materials and Instruments
The digital nature of this project required several technological items. American peer instructors were required
to have a personal computer with internet access, web camera, microphone and access to a web conference and
course management platform (Skype and Vodburner). The treatment group had the synchronous videoconfe-
rencing sessions at the university computer lab with same facilities and access.
Chinese students’ oral skills as reflected in the pre-and post tests were measured on a speaking assessment
rubric with four dimensions (Fluency & Coherence, Lexical Resource, Grammatical Range & Accuracy, and
Pronunciation) which was based within the oral examination components and assessment criteria of the IELTS
examination. The total score of the pre-and post test measure is 100 points, with each dimension weighing 25
points. Items on the attitudinal survey measuring participant confidence in teaching English in future were de-
rived and modified from Jeong-Ah Lee’s surveys on English Teaching Confidence, Attitudes toward English
and English Language Speaking Proficiency [10]. The survey items were scored in 5-point Likert Scale, with 1
indicating lowest confidence, and 5 the highest.
2.5. Data Collection and Analysis
Data collection started with the pre-test of English oral skills for all participants at the beginning of the treatment.
The post-test along with an altitudinal survey on participants’ confidence in teaching English were administered
one week after the treatment ended.
One-way Analysis of Co-Variance (ANCOVA) was conducted to examine students’ performance on the pre-
and post-tests to estimate the difference between the groups on the post-test after adjusting for differences on the