Pumping Up Pronunciation

This review ascertains student opinion of the CD ROM "Pronunciation Power" (English Computerized Learning). The survey analyzes a sample of 51 ESL students at North Seattle Community College in Seattle, Washington.

Overall, the data is remarkably uniform, pointing to many favorable ratings of "Pronunciation Power". The practical significance of the results for language laboratories are discussed and some suggestions are made about potential future improvements of "Pronunciation Power". I designed the survey instrument specifically with ESL students in mind. A review of the literature found abundant materials for evaluating computer assisted language learning (CALL) software but most of it is written for native English speakers. The vocabulary of the survey instrument used here is simplified so ESL students can respond more easily.

The subject sample consists of 51 ESL students from North Seattle Community College in Seattle, Washington. Ages range from 19 to 60 with a mean of 29. Nationalities included: Japanese 29%; Korean 12%; and Vietnamese 10%. Gender divided into 61% female and 39% male. I categorized the subjects as beginners if they were not yet admitted into an Associate of Arts (AA) or Associate of Science (AS) degree program, intermediate if they are currently in an AA or AS program, and advanced if they are beyond an AA or AS program. Subjects are 55% beginners, 27% intermediates, and 18% advanced. After all data was collected, however, the three English proficiency levels only significantly varied from each other on survey statement number 2 - "Text suits your reading ability".

Subjects were limited to those who previously had at least three hours of experience with ESL software. Therefore all the subjects have computer skills and were familiar with ESL software before trying "Pronunciation Power".

The survey consists of a Likert-type rating scale, composed of 30 Likert-type items, on a five point continuum. Subjects are asked to rate each statement with a response as follows: 1 = strongly disagree, 2 = disagree, 3 = undecided, 4 = agree, and 5 = strongly agree All subjects answer thirty questions after using "Pronunciation Power" for 30 to 60 minutes.

Table 1 (page 29) shows the survey statements and their corresponding sample means and standard deviations (SD) for the composite scores. The three groups (beginning, intermediate, and advanced) differ little from each other except on Statement 2, "Text suits your reading ability." On statement 2 the groups rate accordingly: beginner = 4.04 with SD = .74, intermediate = 4.29 with SD = .73, and advanced = 4.78 with SD = .44.

The two statements which score highest are:

• Options, menus, and choices are easily found, and;
• Exercises teach pronunciation skills.

ESL students find that "Pronunciation Power" has clear navigational options and teaches pronunciation. The two statements which score lowest are: There are choices for different level students, and program re-teaches subjects after errors. ESL students appear undecided on whether or not "Pronunciation Power" gives different choices to different student levels or re-teaches subjects after errors.

Students rated this program highly. We will be using this rating system in future assessments of software suitability in the American Language Review. Too often student opinions are not sought and their input can be a valuable tool to acquire a fresh perspective on ESL software selection