Controlled Study

Evaluation of the AlphabiTunes Computer Program for Teaching Beginning Literacy
by Dr. Ian J. Cameron Faculty of Education University of Victoria Victoria, B.C.

Over a two year period, experimental groups, control groups, and placebo groups of kindergarten students were used to evaluate the effectiveness of a computer program designed to teach sight-sound correspondence. Three categories of results were thought to be of interest: the actual results of the testing, the experiences of the experimental classes with regard to the program and the use of computers in general, and the results of native Indian students

Introduction
One of the most contentious issues in education concerns the most effective method of teaching reading. Chall titled her 1967 book on the topic Learning to Read: the Great Debate, and the debate is on-going some 25 years later. Phonics has been condemned by some as a tool for maintaining the status quo with regard to socio-economic inequalities, and as a tool to oppress minorities (Weaver, 1991) and there exists a fairly substantial body of literature, as cited by Johnson and Bauman (1984), to the effect that a concentration on phonics as a reading strategy may negatively affect comprehension ability and perhaps affective aspects of reading as well. On the other hand, numerous studies have concluded that phonics is indispensable in word recognition (Smith, Rhinehart & Thomas, 1993). Smith, Rhinehart & Thomas (1993) surveyed teachers in the 50 U.S. states, and concluded that, “whole language will fail because of the negative attitude towards phonics, which most kids need in order to learn to read.” More recently authorities on reading including Tomkins (1997) and Heilman, Blair and Rupley (1998) stated that phonics is essential in teaching beginning readers, although there are various opinions as to the most effective methods. The idea that phonics plays an important role in beginning reading seems to be mirrored by teachers. Shaffer, Campbell and Rakes (2000) surveyed 208 teachers in three south-eastern states and discovered that almost all felt that phonics was very important, that most students did not have strong phonics skills, and that they had not been well prepared to teach phonics. This suggests that a method that would assist teachers in providing phonics instruction would be timely. The purpose of this study was to examine the effectiveness of the AlphabiTunesTM computer program in teaching sight-sound correspondence to beginning readers.

The Program
AlphabiTunes™ is a CD ROM computer program. According to the developer, AlphabiTunes turns the computer into a virtual teacher that encourages students to work independently and teaches them new concepts by means of stories. - The Tree of Knowledge needs help recovering its lost letters. So the student must travel to distant lands to return the letters to the tree. Along the way, the student will learn a unique story about each letter that presents the letter’s sound, shape and the distinct features of vowels. Games following each letter add depth to the learning. (Pringle, 2001) The program runs on either Windows or Macintosh, and requires 64 K RAM, and a processor speed of 133 MHz. It consists of three modules, each with about nine letters and three numerals. The program tracks student progress.

Method
The study was an experimental design, conducted over two years. In the first year, three kindergarten classes were assigned to each of three groups: experimental, control, and placebo. Existing classes were used, but all students were assigned randomly to classes. Further, classes were assigned to experimental, placebo and control groups entirely randomly. Socio-economic status was controlled for by selecting a school from an upper-middle class, a middle class, and a lower-middle class catchment area for each group. The program was still being developed, so Module One was tested first, and then Module Two. Pre-tests were conducted during an interview with each student. The student was asked to identify the letters that made up the first module. First the letters were placed on the table and each one named in turn by the researcher. The students were asked to point to the letter and say its sound. Then the process was reversed. The researcher said the sound, and the student was asked to point at the appropriate letter and name it. As some letters have more than one sound, the possible score was 35. Students in the three experimental classes used the first module for about 1 hour per week for six weeks. Teachers of the three control classes were asked to involve their students in whatever reading-related activities they thought appropriate, and to report on their activities anecdotally at the conclusion of the study. The amount of time spent in reading activities by the three teachers varied from 15 minutes to 25 minutes per day. Only one of the teachers reported spending extensive time with sight-sound correspondence. Teachers of the three ‘placebo’ classes were asked to spend at least 15 minutes per day specifically on sight-sound activities with the letters found in the first module. These teachers reported having done so, and while the activities varied, all were activities commonly used in primary grades. It should be noted that this was not a true placebo treatment, as the activities might be expected to have positive results. Post-testing for module one was carried out in exactly the same way as pre-testing. At the same time, pre-testing for module two was conducted. In the second year, the program became a victim of its own success. Teachers who taught the experimental classes in year one were so enthusiastic about the results that it was difficult to find schools willing to provide control or placebo classes in year two. Further, the participating district requested that two ‘special’ classes be included, both as experimental classes. One of these classes consisted of native Indian students, the other of English As a Second Language (ESL) students. These classes are anomalies in the district: ordinarily students are heterogeneously grouped at all grade levels. Finally an agreement was reached that saw five experimental classes, three control classes, and three placebo classes take part in the study. As the computer program was now complete, all three modules were tested, using the same procedure as in year one. The possible score was 88.

Results: Year One: Analysis of Co-variance for Module One

Source Type III Sum of Squares df Mean Square F Sig Eta Squared
Corrected Model 3990.190 3 1330.063 72.453 .000 .631
Intercept 2198.556 1 2198.556 119.763 .000 .485
Mod 1 Pre 3430.037 1 3430.037 186.846 .000 .595
Group 251.456 2 125.728 6.849 .001 .097
Error 2331.413 127 18.358      
Total 87001.000 131        
Corrected Total 6321.603 130        
Year One: Analysis of Co-variance for Module Two
Source Type III Sum of Squares df Mean Square F Sig Eta Squared
Corrected Model 9493.117 3 3164.372 79.908 .000 .680
Intercept 1562.077 1 1562.077 39.4463 .000 .259
Mod 2 Pre 7951.107 1 7951.107 200.786 .000 .640
Group 122.752 2 61.376 1.550 .217 .027
Error 4474.797 113 39.600      
Total 106763.00 117        
Corrected Total 13967.915 116        
* indicates significant at .05 level

Discussion

The results indicate this program is effective. Teachers of the experimental classes reported that the students enjoyed using it, an impression that was shared by the researcher and his assistants when conducting post-tests, and that it was easy to run, once the initial bugs were ironed out. One particular strength identified by the teachers was that the program really does provide individualized instruction, so students who have mastered a certain letter can move on to the next letter and then to the next module. Under regular classroom conditions individualized instruction is very difficult to achieve, especially when the tasks involve letter-sound correspondence. The teachers who taught placebo classes reported that their students enjoyed the sight-sound instruction, but that it was sometimes difficult to maintain the attention of the students who had mastered the sounds on which they were working. They reported finding it difficult to individualize instruction. Students who had mastered a letter had to continue taking part in activities involving that letter, or engage in an activity of a different type. They could not work on another letter. Further, when most of the class had mastered a letter, teachers had to move on to another one, with some children still not having achieved mastery. This problem illustrates one of the advantages of the computer program. Of particular interest were the results with native students. The class of 13 native students had a pre-test average of 18.77, the lowest in the study, and a post-test average of 40.58, slightly lower than the average post-test score of other experimental classes, but the average gain for the class was one of the highest in the study. Considering the problems faced by native students in British Columbia with regard to literacy, this is a notable result.

Conclusion

There seems to be a growing consensus that sight-sound correspondence is an important part of teaching children how to read. Current teaching methods, in particular the trend away from ability grouping, make it difficult to meet the needs of children with varying abilities and degrees of readiness when it comes to teaching the letter and their sounds. Unlike many skills that are taught in schools, sight-sound correspondence requires mastery or it is of little use. As is the case with many skills, some children will achieve mastery quickly, and will be bored as others struggle to attain it. The AlphabiTunes program seems to offer a solution to this problem. The program seems to teach sight-sound correspondence, and it does so in an individualized fashion. This program can make a real contribution to the teaching of beginning reading, especially, it would seem with native students.

References

Chall, J. (1967). Learning to Read: the Great Debate. New York: McGraw-Hill.

Heilman, W.H., Blair, T.R., Rupley, W.H. (1998). Principles and Practices of Teaching Reading. (9th ed). Columbus, OH, Merrill.

Johnson, D.D. and J.F. Bauman (1984). “Word identification”in P.D. Pearson ed. Handbook of Reading Research 583 - 608. New York: Longman.

Pringle, B. (2000) Introduction to AlphabiTunes. Victoria, B.C.

Shaffer, G.L., Campbell, P., & Rakes, S. (2000). Investigating the status and perceived importance of explicit phonics instruction in elementary classrooms. Reading Improvement, 37(3), 109 - 118.

Smith, P.K., Rhinehart, S.D., & Thomas, K.F. (1993). Perceptions and reactions of language arts teachers. Reading Horizons, 34(1), 42-54.

Tomkins, G.E. (1997). Literacy for the Twenty First Century. Columbus, OH. Merrill.

Weaver, C. (1991). The Hidden Agenda of Intensive, Systematic and Extensive Phonics. (ERIC No. ED 334558)