Teaching in a Multimedia Computer Environment: A Study of the Effects of Learning Style, Gender, and Math Achievement
Yea-Ru Chuang, Fu-Jen Catholic University Taipei, Taiwan

Abstract
The major goals of this research were to investigate the presentation effects of text, oral narration, and computer animation implemented in an instructional lesson, and to examine individual differences which affect students' learning in a multimedia computer environment. For the purposes of the study, four courseware versions, namely, animation+text, animation+voice, animation+text+voice, and free choice version, were developed to study the interface effects. Subjects' personal characteristics, including FI/FD learning style, gender, and math achievement, were examined in this study. Study results indicated that subjects performed significantly better on the posttest in the animation+text+voice version, which was also the favorite interface design chosen by most of the subjects. It was also found that the animation+text+voice interface effect was only strong for FI subjects, males, or students with low math achievement.

1. Introduction
In a relatively short period of time, computer technology has increasingly changed the ways we teach and learn. Because of the technological advancements, the instructional software developed and used in computer assisted learning has become more sophisticated. Newly-developed instructional software that integrates text, sound, and computer animation now can present material to students in a multimedia form that may maximize its effectiveness.

Dual-code theory provides theoretical support of the use of verbal (such as text) and nonverbal (such as animation) codes in lesson presentations. Studies performed by Mayer and Anderson ( 1991, 1992) and Mayer and Sims ( 1994) confirm the dual-code theory. Their studies reveal that computer animation and oral narration are most effective when they occur contiguously in time or space. Other studies have also verified positive learning effects in certain circumstances where computer animation is used in courseware (Park and Gittelman, 1992; Rieber, 1991; Baek and Layne, 1988; Baggett, 1984). However, a number of questions remain to be answered.

Although previous studies have indicated that learning is enhanced when computer animation is combined with text or oral narration, few studies have been conducted to examine the three factors simultaneously. Thus, the first question to be answered in this study was: What is the most effective way to combine text, voice, and computer animation in the presentation of instructional material? Text only? Animation only? Text plus animation? Voice plus animation? Text plus voice and animation? Which element or combination of elements will contribute most to the student's understanding and, therefore, enhance the learning outcome?

Our second question was: What type of student benefits from each of the three presentation methods? The study of Mayer and Sims ( 1994) reveals that the contiguity effect (computer animations and oral narrations presented simultaneously) is strong for high- but not for low-spatial ability students. What about other individual characteristics? Our study has tried to examine more individual differences which affect students' learning from visual and verbal instruction in a computer environment. Three personal characteristics are examined: Field Independence/Field Dependence (FI/FD) learning style, gender, and math achievement. Thus, the second goal of our study was to find out if there exist significant differences in learning styles between FI and FD subjects, between males and females, or among subjects with different math aptitude, in a multimedia learning environment.

Finally, most of the previous studies have concentrated on the learning effects of various presentation interfaces. Few of them have examined the subjects' preferences among the interfaces. Therefore, the third goal of our study was to find out what kinds of interface designs were favored by the subjects.

2. Methods
2.1 Subjects And Independent Variables
The subjects were 175 seventh grade students who came from eight classes of a rural junior high school in Taipei county of Taiwan.

Field dependence/independence is one of the ways to determine individual cognitive style preferences. It distinguishes individuals in terms of the way in which they analyze information. The field dependent individuals rely more on external references, and focus on individual parts of an object. They tend to solve problems through common sense and intuition and use a trial-and-error approach. At the opposite pole, the field independent persons rely more on internal references, perceive objects as a whole, and tend to reduce problem situations to a set of underlying causal relationships (Ayersman, 1993; Schiff, 1980). Before the experiments, an embedded figure test was administered to roughly 330 students from the 8 classes to determine their FI/FD learning style. Students whose embedded figure test scores were within the top 25% of each class were identified as FI subjects. Those with test scores within the lowest 25% of each class were identified as FD subjects. In all, 175 students were selected to participate in this study.

Among the 175 subjects, 89 subjects were in the FI group (50.86%). Their average score on the embedded figure test was 11.73. The FD group consisted of 86 subjects (49.14%), and their average score on the embedded figure test was -1.68. A significant difference in the average scores was found between the FI and FD groups from an ANOVA analysis (MS=7864.47, F = 705.63, p<.001), which revealed that the two groups did have significantly different learning styles in terms of field dependence/independence.

Based on their final grades in math from the previous semester, subjects were divided into three groups with different math achievement. Students whose math scores were among the top 25% of each class were classified as the high math achievement group. There were 50 subjects (28.57%) in this group with an average math score of 85.32, based on a 100 point scale. Students whose math scores were within the lowest 25% of each class were classified as the low math achievement group. There were 57 subjects (32.57%) in this group with an average math score of 27.02. The rest of the students of each class were classified as the average math achievement group. There were 68 subjects (38.86%) in this group with an average math score of 63.79. The average math scores among these three groups were found significantly different (F=346.29, p<.001).

Among the 175 subjects, 90 students (51.43%) were males, while 85 students (48.57%) were females.

2.2 Lesson Content And Courseware Versions
The experimental courseware used for this study covered the topics of "forces" and "the results of forces" in the Physics domain. These lessons involved the concepts of motion and trajectory. Previous studies have shown that understanding such concepts can be enhanced by computer animation ( Park and Hopkins, 1993; Rieber, 1991). The lesson material was divided into five major parts: 1) force equilibrium, 2) the motion of forces, 3) the calculation of the results of forces, 4) the introduction of resultant forces through parallelograms, and 5) review and practice questions. All instructions were presented at an introductory level. Approximately 30 minutes were needed to complete a lesson.

About the author...

In this study, computer animations were mainly used to illustrate the moving trajectory of a stationary object when pushed by two forces either in the same or in the opposite direction. For example, one of the instructional screens showed an object being pushed in a line by two unequal forces from opposite directions. The object would start to move toward the direction of the larger force.

Subjects from the eight classes were randomly assigned to one of the four lesson versions in the experiment.

2.3 Dependent Variables
A ten-item paper-pencil type of posttest was administered to subjects to measure the learning outcomes. A questionnaire developed by the researcher was used to collect the data concerning subjects' preferences among the interfaces. Subjects took the posttest and filled out the questionnaire immediately after they completed the instructional lesson. It took about 10 to 15 minutes to finish the posttest and the questionnaire.

2.4 Research Questions
Four research questions were issued in this study.

1. Were there significant differences in the effects that the four courseware versions had on learning?
2. For each courseware version, did a) FI, FD subjects, b) males and females, or c) the three groups with different math achievement differ on the posttest scores?
3. For a) FI subjects, b) FD subjects, c) males, d) females, e) high math achievement students, f) low math achievement students, or g) average math achievement students, were the posttest scores significantly different for the four courseware versions?
4. What was the favorite interface design for each type of learner?

3. Results and Discussion
Significantly different learning effects were found among the four presentation interfaces, F=4.20, p<.01. Subjects in the animation+text+voice group scored significantly higher on the posttest (Least Square Means, LSM=71.47) than those in the animation+voice (LSM=64.01) or the animation+text (LSM=59.72) groups. No significant difference was found on the posttest scores between the animation+text+voice and free choice (LSM=69.40) groups. When subjects in the free choice group were asked, "During the lesson, which type of presentation interface did you use most of the time?" 32 out of 43 (74.4%) subjects indicated that they used animation+text+voice type most of the time. This might be the reason that no significant difference was found between animation+text+voice and free choice groups. Based on the results stated above, it could be concluded that among the animation+text, animation+voice, and animation+text+voice interface designs, the animation+text+voice type achieved the best result in terms of learning effects. Table 1 presents the results of ANOVA analysis on the learning effect for different groups of subjects.

Table 1. Results of ANOVA on the posttest scores.

Significant differences on the posttest score were found between males and females, F=7.36, p<.01, as shown in Table 1. The male subjects performed better than the female subjects. However, posttest scores were different only under the animation+test+voice interface. No significant difference on the posttest scores between males and females was found in the animation+text, the animation+voice, or the free choice versions. Study results also indicated that significant differences on the posttest among the four courseware versions were found only for males, F=3.00, p<.05. Male subjects in the animation+text+voice or the free choice versions performed significantly better on the posttest than those in the animation+text or the animation+voice versions. There was no significant difference in the posttest score for female subjects among the four courseware versions. It was concluded that the animation+text+voice interface effect was strong only for the male subjects, but not for the female subjects.

Significant differences were also found on the posttest among the three various math achievement groups, F=15.07, p<.001. Subjects with higher math achievement also had better posttest scores. However, such results were found only in the animation+text or animation+voice versions, but not in the animation+text+voice or the free choice versions. The data shown in Table 3 indicates that only subjects with low math achievement had significantly different posttest scores among the four courseware versions, F=5.79, p<.01. No significant posttest score difference was found either for subjects with high math achievement or for subjects with average math achievement. Low math achievement subjects performed significantly better in the animation+text+voice or the free choice versions among four courseware versions.

Given every possible combination of the presentation methods among text, voice, and computer animation, subjects in the animation+text, animation+voice, and animation+text+voice groups were asked: "If you could choose the presentation interface, what would be your favorite design?" Seventy-six out of the 131 students (58%) indicated that they preferred the animation+text+voice type. Thirty-five subjects (26.7%) said they would choose animation+text as the presentation interface. In the free choice group, subjects were entitled to change their interface during the lesson. However, only one out of 43 persons said that during the lesson, he or she changed the presentation interface frequently. The other 42 subjects chose one presentation format and stuck with it throughout the lesson. Thus, it was not surprising to find that 83.7% (36 out of 43 subjects) of the subjects indicated that they didn't particularly appreciate the choice of presentation interfaces in a lesson.

4. General Discussion
Many studies investigating animation effects in learning have focused on a comparison of the learning effects of computer animation (with or without text) to static graphics ( Rieber, 1996; Park and Gittelman, 1992; Rieber, 1991; Rieber, 1989; Baek and Layne, 1988; Rieber and Hannafin, 1988). Some studies have examined the presentation order between computer animation and verbal narration ( Mayer and Anderson, 1991, 1992; Mayer and Sims, 1994). The present study examined three media factors -- namely, text, voice, and computer animation -- because using the media in combination is more consistent with current presentation strategies applied in today's computer-assisted instruction. Results of the present study revealed that concurrent use of text, voice, and animation in an instructional interface resulted in a significantly better learning outcome when compared to using animation with text alone or animation with voice alone. The animation+text+voice design was also the favorite interface chosen by most of the subjects in this study. On the practical side, these findings are worthy of consideration when integrating multimedia into the presentation interface of a computer-based lesson.

In response to the question, "What type of student benefits from each of the presentation methods?" the results of this study indicated that the animation+text+voice effect was strong only for FI subjects, males, or students with low math achievement, but not for FD subjects, females, or students with high or average math achievement. The findings contribute a fuller understanding of learning in the multimedia environment.

Unexpectedly, we found that few subjects preferred the free choice of presentation interfaces. Due to the limited choices of interface designs provided by this study, further research is recommended to clarify the issue.

The results of this study also indicated that the subjects in the animation+text+voice version got the highest posttest scores. Furthermore, FI students performed better than the FD students. Male subjects scored higher than female subjects. In addition, students with higher math achievement had better performance on the posttest than those with lower math achievement. However, we must be careful in drawing generalizations here because of the nature of the topics covered in the example lessons. Traditionally, Physics is a more spatial and male-oriented field. Subjects' learning outcomes may have been influenced by the knowledge domain. Therefore, further research within different disciplines is also recommended.

5. References 

2. Baek, Y. K. and Layne, B. H. (1988). Color, graphics, and animation in a computer-assisted learning tutorial lesson. Journal of Computer-Based Instruction, 15(4), 131-135.

3. Baggett, P. (1984). Role of temporal overlap of visual and auditory material in forming dual media associations. Journal of Educational Psychology, 76(3), 408-417.

4. Mayer, R. E. and Anderson, R. B. (1991). Animations need narrations: An experimental test of a dual-coding hypothesis. Journal of Educational Psychology, 83(4), 484-490.

5. Mayer, R. E. and Anderson, R. B. (1992). The instructive animation: Helping students build connections between words and pictures in multimedia learning. Journal of Educational Psychology, 84(4), 444-452.

6. Mayer, R. E. and Sims, V. K. (1994). For whom is a picture worth a thousand words? Extensions of a dual-coding theory of multimedia learning. Journal of Educational Psychology, 86(3), 389-401.

7. Park, O. and Gittelman, S. (1992). Selective use of animation and feedback in computer-based instruction. Educational Technology Research & Development, 42(4), 27-38.

8. Park, O. and Hopkins, R. (1993). Instructional conditions for using dynamic visual displays: A review. Instructional Science, 22, 1-24.

9. Rieber, L. P. (1989). The effects of computer animated elaboration strategies and practice on factual and application learning in an elementary science lesson. Journal of Educational Computing Research, 5, 431-444.

10. Rieber, L. P. (1991). Animation, incidental learning, and continuing motivation. Journal of Educational Psychology, 83(3), 318-328.

11. Rieber, L. P. (1996). Animation as feedback in a computer-based simulation: Representation matters. Educational Technology Research & Development, 44(1), 5-22.

12. Rieber, L. P. and Hannafin, M. (1988). Effects of textual and animated orienting activities and practice on learning from computer-based instruction. Computers in the Schools, 5(1/2), 77-89.

********** End of Document **********

IMEJ multimedia team member assigned to this paper	Yue-Ling Wong