The effects of Thinking Maps on reading scores of traditional and nontraditional college students

Marjann Kalehoff Ball


The purpose of this study was to determine whether or not a thinking skills program, specifically Thinking Maps, interwoven with an existing college reading course, would have a significant effect on reading scores. Investigation was also conducted to assess any effects Thinking Maps would have on traditional and nontraditional college students. The study was conducted over 2 semesters at a junior college with participants from college reading classes to obtain a sample of 92 students to form a control and an experimental group. At the beginning of each semester a questionnaire was administered to the participants so that identification could be made of the status of the students as traditional or nontraditional. The Stanford Diagnostic Reading Test (SDRT), Blue Level, Form G, was given to the two groups at the beginning of the semester. Instruction for both the control and the experimental groups was as identical as possible with the same instructor. The only difference was that Thinking Maps, a thinking skills program based on eight fundamental thinking processes represented and activated by semantic maps, were used as tools of learning. The SDRT provides information on a student's skills in reading comprehension, vocabulary, phonics, structure, fast reading, scanning, and word parts. At the end of the semester the SDRT, Form G, was given to the two groups, and the data obtained were analyzed to determine if the treatment of mapping had resulted in a significant difference in posttest reading scores. Further analyses were made to ascertain if a difference occurred when status was considered and if there were a significant interaction between treatment and status. Determination was also made on which variables of fast reading, phonics, comprehension, scanning, structure, vocabulary, and word parts the mapping/no mapping groups differed. Necessary calculations were made using the multivariate analysis of covariance (MANCOVA) using the Wilk's lambda criterion. Follow-up univariate analyses were utilized to clarify any significant multivariate results. Statistically significant main effects were found for treatment ( p <.01). There were no significant main effects for status, nor was there a significant interaction between treatment (mapping/no mapping) and status (traditional/nontraditional). Significant differences at the.01 level were found for five out of seven subtests of fast reading, comprehension, structure, vocabulary, and word parts with the mapping group outperforming the no mapping group on each of the five variables. The findings of the univariate treatment by status analysis of covariance were consistent with the results of the multivariate analysis which found only the main effects for treatment were statistically significant. No statistically significant effects were found for status nor was interaction between treatment and status statistically significant. It may be concluded that mapping made a significant difference on reading test scores. Whether a person is characterized by age, social roles assumed, or other criteria as traditional or nontraditional made no significant impact on reading test scores.