"Mini" ILS Improves Students' Language Mechanics Skills

Afew years ago, School District No. 42, in Nashua, New Hampshire, faced a problem that confronted almost every other district in the country&emdash;insufficient funds. This held especially true for computer software. Beginning in the early 90's, the district implemented a plan to retrofit its 17 schools with some of the latest information technologies. The plan included placing a computer lab at each facility and installing networked software packages. One of the first labs to be completed was at the inner-city Ledge Street Elementary School, which has 650 students in grades one through six with diverse socio-economic backgrounds. The lab contained 15 Macintosh SE 20MB workstations connected to a Mac SE 30 80MB server, using AppleTalk networking software and Farallon PhoneNet connectors. Initial software included HyperCard 2.0 and Microsoft Works 2.0, an integrated word processor, database, spreadsheet and drawing package. Administrators Weigh Costs With available software money nearly depleted, the prospect of finding networked software for 15 machines seemed rather discouraging. Ledge Street administrators had learned that Integrated Learning Systems (ILS) ranged from $5,000 to $10,000 or more. "We were looking for lab software that would address an academic need of the majority of our fourth to sixth graders and still be within our financial reach," says Dr. Ray Van Gr'esbeck, then a sixth grade teacher and computer coordinator at the school. After seeing an advertisement in a technology journal, Van Gr'esbeck purchased Perfect Copy, a language skills development package from Logicus Inc., of Schomberg, Ontario, Canada. With the program, up to 200 students can develop basic language mechanics skills at their own pace and at different levels of difficulty. It includes a management system, so both teachers and students can monitor progress. Van Gr'esbeck refers to Perfect Copy as a "MILS" (Mini-Integrated Learning System), which he defines as software that gives students some control over the direction and complexity of the subject material being presented. The package consists of nearly 500 articles covering the following skills: punctuation, pronouns, abbreviations, apostrophes, capitals, commas, numbers, irregular verbs, subject/verb agreement and more. Over 200 grammatical rules are incorporated for review as students correct the intentional errors made in each article. Three levels of help are offered to the learner: the first shows which lines in the article contain errors; level two highlights words that need to be corrected; and the third models the corrected sentence. Each clue level may be disabled by the teacher. When an article has been successfully completed, the program displays a graphic message. Van Gr'esbeck says that students of all achievement levels were fully engaged in the exercises. One of the highlights of Perfect Copy, according to Ledge Street teachers, is a text editor that enables them to create their own articles. For example, notes Van Gr'esbeck, an article could be written about the American Revolution for fifth-grade students that features historical facts plus language mechanics errors that need to be corrected. "The student receives a double learning situation." Tests Measure Progress But administrators wanted to know if the youngsters were truly absorbing the information in articles. With the help of Logicus, in the fall of 1993 they arranged for a study to measure the language mechanics skills of 79 students (in grades four and six). Instructors first pre-tested the students using a subtest of the California Achievement Test and the Learning Styles Inventory from Price Systems, Inc., of Lawrence, Kan. The experimental group worked with Perfect Copy Version 1.1.2, while the control group learned through traditional drills and lectures. After 13 weeks of daily instruction, an alternate form of the CAT was administered and, according to Van Gr'esbeck, those who used the MILS performed noticeably better than the others. (Experimental group: mean=52.68, standard deviation=19.16 / Control group: M=42.97, SD=16.80) Van Gr'esbeck adds that results showed the software worked equally well for all learning styles and for males and females. "We were quite pleased with the outcome of the study," he concludes. "This was well worth the relatively small investment."

This article originally appeared in the 05/01/1995 issue of THE Journal.