Teaching writing strategies through multimedia authorship (1).

This study investigates the possibility that multimedia authorship may help students become more proficient writers. Four classes of 8th-grade students in an urban setting participated in the study, which took place over seven successive class periods. Students were given a 7-paragraph article on teen smoking to use in developing a nonlinear presentation with HyperStudio[TM]. Organizational skills necessary for successful writing were emphasized throughout the unit. The students wrote an essay on an assigned topic in a counterbalanced manner both before and after they completed the multimedia project. There was a significant increase in students' scores on the organizational quality of the essays from the pretest to the posttest, especially for students who received low scores on the preessay. Data on the multimedia projects created by the students are also presented, and the relationship between multimedia composition and traditional writing is discussed.

**********

Multimedia is playing an increasingly important role in education, and one educational use of multimedia is for students to create their own multimedia documents. The similarity between this process (i.e., multimedia composition) and writing is quite apparent, for both multimedia authorship and writing involve many of the same basic cognitive processes--for example, planning, transforming, evaluating, and revising (Hayes, 1996; Hayes & Flower, 1980; Lehrer, 1993). In addition, computers and word processing have made it relatively easy to include media other than text (e.g., images, graphs, animation, and audio) in "written" documents, and such possibilities may be changing the way people communicate with one another. This possibility, along with the similarity between multimedia authorship and writing, provides the theoretical context for the present study.

The purpose of the study is to investigate the possibility that multimedia authorship may help students become more proficient writers. Students who learn to create multiple representations of their ideas by developing multimedia projects may be able to apply those same strategic thinking skills to other forms of communication, especially writing. Thus, the main concern of the study is whether teaching students about multimedia design and having them develop their own multimedia project will enable them to better organize their ideas in traditional text writing.

The representation of knowledge is fundamental to the writing process. One reason students struggle with writing is that they are unable, for one reason or another, to tap into their own knowledge resources in a manner capable of being understood by others. They are unaware that others may not understand the information presented in their writing in the same way that they themselves understand it--that is, they do not realize that knowledge can be represented and organized in different ways, and that different people often represent the same information differently. They also do not understand that writing is about communicating ideas and not about the written end product. Students must learn to organize the ideas and information they wish to communicate to their readers. Therefore, how students go about representing knowledge and how that knowledge is organized are important concepts when dealing with the teaching of writing.

Being able to represent knowledge in a variety of ways, including visual-spatial, "are essential for understanding the message of the text" (Hayes, 1996, p. 5). Although proficient writers are able to describe visual images and sounds with text, such linguistic representations carry different types and amounts of information than actual images and sounds, which are represented in different ways than text (Bower, 1972; Larkin & Simon, 1987; Paivio, 1971). Information coded in multiple ways (e.g., text and visual) tend to be learned better than information coded in only a single medium (Jacobson & Spiro, 1995; Paivio, 1971). Thus, helping students learn how to represent information in different media, as occurs when they develop multimedia products, has the potential for facilitating students' writing.

A major element of an effective teaching approach to writing is having students see writing as problem solving (Flower & Hayes, 1981; Hayes, 1996; Hayes & Flower, 1980). Writing is part of a larger process in which the writer must articulate a problem, establish a position about the problem, locate and organize information, create an effective argument, and select appropriate support (Flower & Hayes, 1981). Writing, however, is an ill-defined problem (Hayes, 1996; Petraglia, 1995) that can be solved in many different ways with a multitude of different strategies and methods. Solutions are necessarily dependent on the writer's knowledge, perspectives, organizing principles, and schemas. Understanding these elements, and the processes that make use of them, helps students create documents that are purposeful and functional.

Thinking skills utilized during the writing process, however, are not isolated cognitive processes. If a student learns how to become a proficient problem solver within the realm of the writing process, then those thinking processes and strategies should improve, to some extent, the cognitive and problem-solving abilities of the student within other, related situations. Thus, it is reasonable to investigate the possibility that multimedia composition provides a useful vehicle for teaching writing, both in the traditional sense of text writing and in the new possibility of using multimedia documents as a vehicle for "written" communication.

Developing multimedia documents can support writing in a number of ways. Lehrer (1993; Lehrer, Erickson, & Connell, 1994) used the Hayes and Flower (1980) model of writing in their research on hypermedia design. Multimedia composition has a number of important educational advantages: It can be used to encourage students to engage in active problem solving and strategic thinking about how to represent their ideas in appropriate media, how to construct meaningful and accurate links to other ideas, how to think about relationships among ideas (Lehrer et al., 1994), and how to communicate those ideas to other people. Students must (a) define the problem, (b) identify goals, (c) gather, select, organize, link, and display information, (d) evaluate their work, (e) revise plans, (f) prioritize tasks, and (g) make a variety of decisions in order to complete the project. Perhaps most important, multimedia composition can ease the transition from concepts and images to words (Halio, 1996) by helping students communicate their thoughts, not only through words, but also through music, pictures, colors, and the nonlinearity that text on paper does not allow.

If students develop multiple representations of concepts by developing multimedia products, then hopefully they will be able to apply those representations flexibly to other situations. To the extent that successful multimedia design is similar to successful written composition in requiring multiple cycles of drafting and revision, then students may be able to be taught the cognitive processes used in both forms of communication concurrently.

With very few exceptions, however, research on multimedia composition has been concerned with students learning subject matter (e.g., history) rather than communicating with other people (e.g., writing). We have found only one study (Halio, 1996) directly concerned with multimedia authorship and writing.

There is evidence, however, that multimedia projects that require active participation and engagement in authentic problem solving have the potential to enhance students' motivation and learning (Hay, Guzdial, Jackson, Boyle, & Soloway, 1994; Jonassen, Pech, & Wilson, 1998; Lehrer et al., 1994; Liu & Pedersen, 1998). For example, Lehrer et al. (1994) conducted a study in which ninth graders created hypermedia presentations on American history for other students. They found that students' time on-task increased significantly over the course of successive design projects. In addition, the design process helped students to internalize various design skills. Students reported increases in mental effort and involvement, interest, planning, collaboration, and individualization.

Nicaise and Crane (1999) studied a graduate course in which students learned about educational theory by designing and creating a hypermedia chapter for an Internet based book. Qualitative data provided both students' and teachers' perspectives regarding the course. Most of the students were highly satisfied with the course. Although six of the 12 students in the course demonstrated an in-depth understanding of the material, transferring this understanding to other situations, four of the remaining six students held major misconceptions about the topic they were studying. The authors suggest that one of the major limitations of the course was a necessity to devote too much time to learning how to use the technology.

Research focusing on using multimedia in an English curriculum generally has been concerned with learning from or creating hypertext documents, having students create multimedia projects rather than traditional research papers, or using multimedia to add additional meaning to the written text. Halio (1996), for example, had students in her first-year college composition course add sounds and graphics to their essays in the hope that thinking about which graphics and sounds to include would improve the quality of their written narratives. The essays of some students clearly benefited from multimedia composition, but other students had difficulty integrating graphics and sounds into their essays in a meaningful way. These latter students generally either collected their images and sounds before beginning the written part of their essay or wrote complete drafts of their essay before thinking about which sounds and images to add.

Other research within the English curriculum includes action research by practicing teachers. Much of this research, however, is anecdotal and focuses on student motivation and the end project. An example of such research includes Haviland and McCall (1999) who sought to revise the traditional research project by using HyperStudio[TM]. They found that using technology within the classroom enhances presentations, fosters collaboration, and serves multiple learning styles. At present, no research has been published dealing specifically with the use of multimedia authorship to supplement teaching students the cognitive processes involved in the writing process.

METHODOLOGY

This study involved a collaborative effort among the researchers, the 8th-grade English teacher, and the school's computer teacher. The 8th-grade English teacher gave her students an English-Language-Arts practice test at the beginning of the school year. (2) Based on this practice test, she observed that the students had most difficulty with the listening and writing part of the test, especially in understanding how to categorize ideas in an essay into main ideas and then into details that support or are related to these main ideas--that is, how to organize their ideas in meaningful and coherent ways. This problem resulted in the students' thoughts and ideas lacking logical coherence within their written essays. Consequently, this study was based on creating a unit plan that would help teach students organizational skills necessary for successfully taking the ELA examination at the end of the year.

Participants

A total of 86, 8th-grade students in four classes participated in the study during their scheduled English classes. The school is located in a Northeast urban setting. Although technically a magnet school, approximately two-thirds of the students come from the neighborhood with the remaining one-third selected by lottery from those who apply, typically during the elementary school years. The school population is of mixed-ethic background. According to the New York State Report Card (New York State Department of Education, 2001), 81-90% of the student population come from families who are on public assistance. In terms of academic success, in 2000 only 27% of the 8th-grade students at the school passed the English Language Arts (ELA) exam.

Pre/Posttest

Since this study was based upon teaching organizational skills in writing, a pretest and posttest were administered to measure the effectiveness of the instructional cycle. Two essay topics chosen directly from the ELA Examination practice book, Aim Higher (Castro & Shepherd, 1999), were used in a counter-balanced manner on the pre and posttest. The two topics were (a) "A school cafeteria at lunchtime" and (b) "The monkey house in the Zoo." On the pretest, two classes were given the first topic and the other two classes were given the second topic; on the posttest, each class wrote on the topic they had not received on the pretest. This procedure ensures that any difference obtained between the tests cannot be explained in terms of one topic being easier than the other. For both tests, students were given the following instructions:

    You are to write a descriptive paragraph based on the following
    topic. Someone who knows nothing about this topic will be reading
    this paragraph. Therefore, you must include as much information as
    possible about this topic. First create a graphic organizer, of your
    choice, that will help you describe the scene. Do not begin writing
    the paragraph until you have completed the graphic organizer. Using
    your graphic organizer, write the descriptive paragraph on a
    separate sheet of paper.

Following the pre and posttests, students were given three questions to answer to find out if their knowledge about multimedia composition changed as a result of the instructional cycle. These questions were: (a) What else can you include with your text to help clarify or add more meaning to your description that would help your audience understand the scene you are describing?; (b) In what ways is the organization of text articles (such as newspaper or magazine articles) the same as the organization of web pages with the same type of information?; and (c) In what ways is the organization of text articles (such as newspaper or magazine articles) different from the organization of web pages with the same type of information?

Instructional Methods

This study took place during seven, 40-minute English-class periods. The students were in the regular classroom once for an introductory lesson and in the computer lab for the subsequent six class periods. The computer lab consisted of 25 networked iMac computers, and each student had access to their own computer. The lessons were taught by the first author, a graduate student certified in English education. Also present were the students' regular 8th-grade English teacher, the Technology Coordinator for the school, and the second author. All those present worked collaboratively with the students.

Multimedia project and unit plan. Students participating in the study individually completed a multimedia project using HyperStudio[TM], a multimedia authoring program which allows students to create a multimedia document that includes text, pictures, animations, video, sound clips, and internal and external links. Throughout the unit, the HyperStudio[TM] stacks were compared to web pages and web sites: that is, an individual card is like a web page, and the stack, made up of individual cards, functions like a web site.

For the project, students were given a printed copy of an informational article titled "Teen Smoking Rates on the Rise" consisting of 7 paragraphs. The students' objective was to use HyperStudio[TM] to present all the information in the article without altering the meaning or objectivity. They were to include all the text from the original article in their multimedia project and create a nonlinear multimedia presentation of this information.

On the desktop of each computer was a graphics folder containing photographs, pictures, and graphs. All images used in the students' projects came from this graphics folder. If a student had a certain graphic or picture in mind that was not supplied in the folder, they could make a special request, and the teacher would try to fulfill the request by the next day. The reason students were not allowed to choose their own graphics to add to their projects was that in previous research students had been unable to locate suitable images within a reasonable amount of time. Because of time constraints for the cycle, there was not sufficient time for students to search for or create their own graphics/images.

Introductory lesson. The first class of the unit took place in the regular classroom and served as an instructional lesson to teach students the differences between a text document and a multimedia document. Most importantly, the students needed to understand the difference between linear and nonlinear presentations. During this period, it was made clear that the projects were not to be modeled after the text document, for example, a linear presentation. Rather, the project was to be structured in such a way as to force the students to think about how to reorganize and categorize the ideas in the article.

At the beginning of the class period, students received a Venn Diagram with two overlapping circles and given 5 minutes to brainstorm ideas about the similarities and differences between (a) A Web Site and a (b) Text Article/Essay. They were prompted by asking them to think about the following:

    What are the differences and similarities between using the Internet
    (WWW, web pages) and a text article to provide information? For
    instance, imagine a web site that provides information about musical
    instruments, then compare that web site to a book that provides
    information about those same musical instruments: What's the same?
    What's different?

After completing this activity, the students were asked to share their observations and thoughts with the class, as the instructor wrote their ideas on the chalkboard. This procedure enabled the instructor to determine the amount of experience the students had had with the Internet and to help the students better understand the difference between linear and nonlinear documents and how important it was for their own project to be a nonlinear presentation of the text document.

The importance of organization and categorization in the ELA examination, as well as the listening part of the exam were also discussed. A reference titled "Examples for How to Organize and Categorize Information" was distributed and briefly reviewed.

Finally, the students were introduced to HyperStudio[TM] through a projected presentation of a HyperStudio[TM] stack that explained "card," "stack," how navigational buttons work, and so forth. Next, students were shown an example of the type of project they would be completing, using an article directly from their ELA practice books. While the instructor navigated the example project, students were asked questions dealing with: (a) the function of the homepage, (b) the use of color and font size, (c) button navigation, (d) text links, (e) why one should or should not use a picture, (f) the different categories into which the article was split, and (g) the layout of the cards in the stack.

At the end of the class period, the students were handed the teen smoking article and told to read the article. Their homework was to think about the categories into which they would separate the article for preparation of their HyperStudio[TM] projects.

Instruction on subsequent days. All subsequent lessons were held in the computer lab. On Day 2, a projected presentation introduced the students to HyperStudio[TM], with which none of the students were familiar, and showed them how to: (a) create a new document, (b) save the document, (c) create a button, (d) add images, (e) insert a text box, (f) change colors, and (g) cut-and-paste text from a word-processing document into HyperStudio[TM]. Video, sound, and animation were not covered as they were not considered essential elements for completing the project. If students wanted to learn how to use additional features, such as "making a text link," they were taught individually. The students also were given a HyperStudio[TM] tutorial that they could use as a reference. During Days 3-7, students worked on their projects, and the adults present (see above) circulated around the room helping students with questions and problems.

RESULTS

Venn Diagram Activity

A total of 77 of the 86 students completed the Venn-Diagram activity on Day 1. This activity was used to determine students' conceptions of similarities and differences between web pages (multimedia documents) and text articles (books or essays) and their understanding of differences between linear and nonlinear communication. Since students could list several similarities and differences, the 77 students produced a total of 318 responses. These responses were recorded in one of three categories, depending on the part of the Venn Diagram in which the response was written: (a) ways in which a web site/page is different from a text article, (b) ways in which web pages and text articles are similar, and (c) ways in which a text article is different from a web page. The responses were then coded according to the theme that best described the response by first reading all the responses and then grouping them into themes, taking note of all responses that were unique (see Table 1 for the themes represented). Table 1 reveals the following results:

Ways in which a web page/site is different from a text article. The most frequently cited difference (mentioned by 42% of the students) was that web pages can include multimedia; 31% said web pages provide more information; 18% said web pages are faster and easier to use; 18% said web pages contain links to additional information; and 14% said that web pages are always updated.

Ways in which web papers and text articles are similar. Two-thirds (66%) of the students said both web pages and text articles provide the same information, while 31% and 27%, respectively, mentioned that both included pictures and that both contain text.

Ways in which a text article is different from a web page. Nineteen percent said that with text it is harder to find information on a topic and that more work and time is needed to find the information, 10% said that text is limited in that it provides less information, and 9% said that text is boring or not interesting. One student noted that text is linear.

Multimedia Project

A total of 78 of the 86 students completed the HyperStudio[TM] project. While working on the project, many of the students failed to understand that they needed to include the entire teen-smoking article in their multimedia presentation. Some students thought they merely had to summarize what the article was saying and began to put the article into their own words. Other students seemed to stray from what the article actually contained and were adding their own prior knowledge about teen smoking and nonsmoking messages that did not appear in the article.

After this initial confusion, on Day 3 (second day in the computer lab) a decision was made that the project's instructions and objectives needed to be reinforced. The next day, before they continued working on their projects, the students were asked to read over the article and categorize the information. They were reminded of the example HyperStudio[TM] stack presented on Day 1. After categorizing the information by writing on their copy of the article, they were asked to think about what information would go on each card of their stacks. It was announced frequently that they must use the entire article. Some students caught on quickly to the project objectives, although others still seemed lost and needed help with presenting the information in a logical sense.

Students had not been told a specific number of cards to include in their stacks, although the various projects contained between 4 and 9 cards with an average of 5.9 cards per stack. The importance of using a homepage was emphasized frequently, and students were asked to think of their projects as if they were creating web pages for the Internet. The purpose for emphasizing the importance of a homepage was to stress the importance of a nonlinear presentation of information. As indicated in Table 2, 78% of the students who completed the project included a homepage in their stacks, and 55% of the students included navigational buttons on their homepage to each card in their stack.

Organization. Most of the students organized their stacks based upon how the original article was organized, that is, in terms of paragraphs. For example, when the students were asked to categorize the information on their paper copy of the article, some students labeled the seven paragraphs in the article with numbers, which corresponded to how many cards they would include in their stack. Other students categorized the information by the main subjects, such as advertising, Hollywood, statistics, and so forth. For example, two of the paragraphs in the original article are about Hollywood stars, and the students would combine those two paragraphs onto one card in the stack. Many students also did the same for the two paragraphs about advertising.

Linear / nonlinear presentations. As indicated in Table 2, 36% of the completed projects were considered to be a linear presentation. Linear in this context is defined as a project having only one specified way to navigate through the stack. A linear document does not permit the user/reader to look at the information in the sequence they choose. Instead, a linear document "chooses" for them. For example, including only one navigational button per card would be considered a linear presentation.

Some of the students who finished early used the full article, but their presentations were linear--that is, each card in their stacks contained only one link. An examination of the projects of those students that finished early indicates that these students did not spend adequate time categorizing the information. These projects looked as if the text from the article had been put into the stacks randomly without thought or preplanning. One student, for example, placed each paragraph from the article onto individual cards in his stack and then linked the cards together in no particular order with one button on each page that led to the next card in his stack.

Student design concepts. Considering the time constraint of the unit, the students caught on quickly to accepted concepts of multimedia design. The rubric used by the teacher to grade the completed projects, and emphasized by the instructor during the lesson, concentrated on four aspects of design: (a) organization and layout, (b) navigation (buttons and links), (c) use of graphics, and (d) use of text. Projects were scored on a 4-point holistic scale. For example, a score of 4 would be given to projects meeting the following criteria: (a) the entire article was used, (b) the pictures correspond to and enhance the text, (c) the presentation is nonlinear, and, (d) navigation through the stack is easy and logical. Approximately two-thirds (64%) of the students that submitted projects were successful in creating a nonlinear presentation.

Navigation. As demonstrated by their final projects, students' understanding of logical navigation differed drastically. Twenty-seven percent of the projects contained links that were confusing and were hard to navigate. Examples of this include cards that contained no links to other cards within the stack, therefore making it impossible to navigate to all the information within the entire stack. Other examples include the user "getting lost" within the stack and not knowing how to get back to the homepage. Twenty-seven percent of the projects contained approximately the same number of logical and confusing links. Forty-six percent of the projects contained logical and easy to navigate links. Examples of this would include the user knowing exactly where they were inside of the stacks at all times and access back to the homepage was consistent throughout the stack.

Only three students included text links (i.e., a link from a word or words to additional text, which perhaps provides a more detailed explanation) in their projects. Although text links were not specifically taught to the whole class, they were included in the example HyperStudio[TM] project and discussed on the first day. Also discussed during presentation of the example stack was the importance of giving each card in the stack a title to let the user knows exactly what information is being viewed. Only 45% of the projects contained titles on each card in the stack. Many students also had problems with the labeling of navigational buttons. Fifty-one percent of the projects contained buttons that where labeled generically with such terms as: "next page," "back," and "forward." The use of color to help users navigate through the stack also was discussed during the first day, and 24% of the students used color meaningfully--for example, the color of the navigation button corresponded to the background color of the card to which it was linked.

Extra functions. Only two of the four classes experimented with the extra features of HyperStudio[TM], which include animations, transitions, and sound. It was emphasized that students needed to complete their stacks before they could experiment with the extra functions. In all, 22% of the projects contained extra sounds and transitions.

Pre/Postquestionnaire

The students' answers to the three questions on the pre and postquestionnaires were tallied and coded by looking for reoccurring themes and unique responses of the students. No significant differences occurred in the frequency of themes between the pre and postquestionnaire. The following results were found for each of the three questions:

Question 1: What else can you include with your text to help clarify or add more meaning to your description that would help your audience understand the scene you are describing? This question was not answered by 24 students on the prequestionnaire and by 20 students on the postquestionnaire. Of the students who did answer the question, 41 and 45 students, respectively, mentioned they could have added more text details to their essay; and 15 and 13 students, respectively, said they could have added pictures or illustrations. Between one and five students mentioned the following themes on the pre and postquestionnaire, respectively: add charts/graphs/diagrams (2/0), add sound (3/5), add smells (1/0), take them there (1/3), add color (0/1), add others' opinion (1/0), and create a HyperStudio[TM] stack: "just like we did in the computer room" (0/1).

Question 2: In what ways is the organization of text articles (such as newspaper or magazine articles) the same as the organization of web pages with the same type of information? This question was not answered by 35 students on the prequestionnaire and by 21 students on the postquestionnaire. Of the students who did answer the question, 31 and 39 students, respectively, said that the information in text articles and on web pages are the same; 6 and 17 students, respectively, said that both articles and web pages contain pictures; 6 and 14 students, respectively, said that they both contain text; and 7 and 3 students, respectively, said they both have pages. Eight other themes were mentioned with frequencies of 1 to 6 responses.

Question 3: In what ways is the organization of text articles (such as newspaper or magazine articles) different than the organization of web pages with the same type of information? This question was not answered by 42 students on the prequestionnaire and by 26 students on the postquestionnaire. Of the students who did answer the question, 21 and 15 students, respectively, said that web pages and text have different form or format; 6 and 11 students, respectively, said that web pages have links; 6 and 14 students, respectively, said that web pages include multimedia; and 5 and 11 students, respectively, said that web pages have a greater variety of information. Eight other themes were mentioned with frequencies of 1 to 8 responses.

Pre/Postessay Scores

The pre and posttest essays were scored on a four-point scale using a rubric that focused only on the organizational aspect of the essays. All the essays were read and scored at the same time without regard to order (i.e., pre or postessay) and blind as to the identity of the author. Scoring was as follows: a score of four represented an essay with a clear and logical progression of ideas and events; a score of three represented a logical progression of ideas and events, but with minor lapses of organization; a score of two represented some progression of ideas and events; a score of one represented an essay with no sense of strategy or control of organization; and a score of zero represented an incomplete essay.

Fifty-eight students completed both the pre and postessay. The pre and post scores were compared and the results categorized as follows: the student improved one, two, or three points from the score on the pre to the postessay; there was a one or two point decrease from the pre to the postes-say; or there was no difference between the pre and postscore. Forty-five percent of the students improved their essay scores, 36% of the scores remained the same, and 19% of the scores were lower by one or two points on the postessay than they had been on the preessay. This overall difference in the organizational quality of the students' essays was statistically significant across the four classes, [[chi square].sub.6] = 17.1, p < .01.

In one of the classes, however, only one student's score improved, and 41% of the students received a lower score on the postessay than on the preessay. A closer examination of this class reveals that all but 2 of the students received a score of 3 or 4 (the maximum score) on the preessay, leaving little if any room for improvement. Since the results for the other class receiving the same counterbalanced order is similar to the classes receiving the other counterbalanced order, it appears the decline did not result from a differential difficulty in the essays.

Another possibility is that improvement depends on the organizational score received on the preessay, with greater improvement for those receiving lower scores. To test this possibility, the data were tabulated according to preessay scores (Table 3). These differences are significant, [[chi square].sub.2] = 22.0, p < .001, indicating that students who scored 0, 1, or 2 on the preessay had higher organizational scores on the postessay, but no improvement occurred for students receiving scores 3 or 4 on the preessay.

DISCUSSION

This study provides a number of interesting findings regarding students creating multimedia products and the relationship of this process to traditional text writing. The most important finding is a significant increase in the organizational quality of traditional text essays written by students who experienced an instructional unit designed to teach organizational strategies through multimedia composition. This improvement, however, occurred only for students who originally received low-scores on their initial essay. Only 9% of the students who received 3 or 4 points of the 4-point maximum improved, 39% received lower scores on the postessay, and 52% scored the same. This failure to find an improvement in organizational scores for high-scoring students may be the result of the obvious ceiling effect or that these students already had acquired the skill prior to the treatment. The overall improvement, however, suggests that the two tasks (multimedia composition and traditional essay writing) are sufficiently similar for organizational strategies learned during the multimedia design task to be used in the second, essay-writing task.

The Venn-Diagram activity showed that the students had a limited understanding of the differences and similarities between traditional text and web pages. For example, only 18% of the students mentioned that web pages can include links, and only 27% said that both web pages and text articles contain text. This finding indicates that more preparation with students is needed in order for them to have a better understanding of linear and nonlinear communication. The present study suggests that multimedia instruction needs to be introduced to students at the beginning of the year for the full effect to be achieved. Students need to have a strong understanding of multimedia design concepts, especially as they relate to the writing process, and more time needs to be spent with the students on teaching them these concepts and the similarities between the writing process and multimedia composition.

Students would benefit from being introduced to the concepts of organization and categorization before they began working with HyperStudio[TM], preferably at the beginning of the school year. During the present study, students were only handed a ditto sheet that outlined several different ways in which information can be categorized. In addition, students would have found the project's objectives less confusing if they had been more familiar with HyperStudio[TM]. Since this was the first time they had used HyperStudio[TM], once they were in front of the computer, nearly all their attention was focused on learning to use the program rather than on the project. After practice with HyperStudio[TM] and it's functions, students should be given time in the regular classroom to discuss the project and think about how they could use HyperStudio[TM] to present their articles in a nonlinear fashion, which should reduce the confusion that occurred in the present study.

One way in which such preplanning could be to conducted would be to have the students construct a story board of their HyperStudio[TM] project in preparation for the time they spend in the computer lab. For example, the relationship between a HyperStudio[TM] stack and the cards that make up the stack could be demonstrated by giving students index cards and having them use the cards to plan out and storyboard their projects. They would be able to plan the interface of each card, the placement of buttons, and links between cards in the stack. Navigation through the stack would also need to be planned with the index cards, and the students could show this relationship by arranging their cards on their desk or on a larger sheet of paper and draw lines between linked cards. In this way, they would have planned out their presentation, interface, navigation, and organization before they started to use HyperStudio[TM].

Providing students with good examples of multimedia design also would be helpful. The pictures distracted the students at first. Most of the students did not chose pictures and colors to use in their stacks based on the information in the article. Rather, these were chosen on the basis of what they themselves preferred visually. This problem is similar to the difficulty experienced by Halio's (1996) students when they completed either the written text or locating images before trying to integrate the two media. More emphasis needs to be placed on the function of pictures within a document and developing the document in an integrated manner from the beginning. If a model of composition or structured activities were given to the students, they probably would have spent more time thinking about the design, navigation, and interface of their HyperStudio[TM] projects.

This investigation was based upon the notion that the thinking processes involved in writing and multimedia composition are similar. In order to assist students in making comparisons between these two methods of communication, they first must have an understanding of the thinking processes involved in these forms of communication. If students are struggling with the writing process, then teaching them multimedia design concepts and having them become familiar with the important steps that must be taken (i.e., planning, transforming, evaluating, and revising) should help students make the writing process more meaningful.

All things considered, including the short amount of time that was available, the HyperStudio[TM] projects completed by the students were impressive. None of the students had used HyperStudio[TM] before and were unfamiliar with the interface. During the six, 40-minute class periods of the teaching cycle, all the students successfully learned how to create cards, paste text onto their cards, insert graphics, insert navigational buttons, and connect the cards to each other in order to make a complete stack. The students were only lacking a solid foundation of multimedia design knowledge, which takes time to emerge.

The students were truly motivated by HyperStudio[TM]. They enjoyed the program, and even problem students seemed to work hard on their projects. Some students surprised the English teacher with the quality of the work they completed. Those students are usually unmotivated, and their poor class work reflects this lack of enthusiasm for academics. For those students, it appeared that HyperStudio[TM] provided them with the motivation to show that they are capable students. One class period, the teacher announced to the students that they only had two more minutes of class and that they needed to save what they had and quit for the day. Many of the students reacted with displeasure, as if they wanted to stay in the classroom and work on their projects some more. Usually it's the opposite: the students can't wait to leave the classroom. One student was quoted as saying during this cycle, "I just wanted to tell you that this is the most interesting thing I've done in English class all year!"

In many ways, the two-dimensional quality of linear thinking is being replaced with multidimensional thinking. If this prediction is accurate, then helping students learn about the writing process and about communicating ideas by engaging in multimedia design can serve as an important passage towards understanding multidimensional thinking. The present study is a step toward better understanding the nature of multimedia composition. Since the current project lasted for only seven 40-minute class periods, it would be useful to know if the gains in writing ability that were obtained are sustained in the longer term. Research on having students become multimedia authors to teach writing strategies can advance understanding about this topic and help teachers to search for new ways to effectively design such multimedia learning environments.

Table 1 Themes of Responses to Venn-Diagram Activity

                                                Number of  Percentage of
                                                Responses  Students

How Web Pages/Sites Differ from Text Articles
No response or response that did not make
  sense                                         17         22%
More information                                24         31%
Includes multimedia                             32         42%
Always updated                                  11         14%
Different organization / medium                  2          3%
Includes links                                  14         18%
Faster, easier to use                           14         18%
More pictures                                    1          1%
Provides immediate communication                 3          4%
Hard to read                                     1          1%

How Web Pages/Sites are Similar to Text
  Articles
No response or response that did not make
  sense                                         16         21%
Provides same information                       51         66%
Both includes pictures                          24         31%
Both contain text                               21         27%
Both have pages                                  1          1%

How Text Articles Differ from Web Pages
No response or response that did not make
  sense                                         43         56%
Limited text (less information)                  8         10%
You can carry it around                          2          3%
Linear                                           1          1%
Boring, isn't interesting                        7          9%
Has more information                             5          6%
Easier to read                                   3          4%
Words are static                                 2          3%
Harder to find information / more work and
  time                                          15         19%

Note: Seventy-seven students responded to this activity generating a
total of 318 responses. Percentages are the percentage of the 77
students whose response reflected the theme indicated.

Table 2 Number and Percentage of Completed Student Projects with Given
Characteristics

Characteristic                                           Number  Percent

Linear presentation                                      28      36%
Used entire article                                      41      53%
Contained a homepage                                     61      78%
Homepage contained button to each page in stack          43      55%
Each page has all buttons                                17      22%
Buttons are clearly labeled                              33      42%
Buttons are generically labeled                          40      51%
Titles on each page/card                                 35      45%
No titles on pages/cards                                 35      45%
Some pages titled/other pages not titled                  8      10%
Links are confusing (hard to navigate)                   20      26%
Some links are logical, while other links are confusing  18      23%
Links are logical (easy to navigate)                     36      46%
Unrelated titles/button labels                            5       6%
Included text links                                       3       4%
Use of colors is meaningful                              19      24%
Added sounds and transitions                             17      22%

Total number of students in all 4 classes = 86
Total number of projects completed = 78

Table 3 Number of Students Who Score on the Preessay Improved on the
Postessay

Score on   Improved  Same  Declined  Total
Pre-Essay

0-2           24       9      2        35
3-4            2      12      9        23
Total         26      21     11        58

Footnotes

1. This article is based on a master's thesis by the first author under the direction of the second author. An earlier version of the article was presented at the 2002 annual meeting of the American Educational Research Association. The authors wish to thank Anne Michael and Shelley Bernosky of the Buffalo Public Schools for the many contributions they made to the successful completion of this project.

2. In New York State, all 8th-grade students must take an English Language Arts (ELA) examination at the end of the school year. This test assesses the students' skill in the areas of reading, listening, and writing. A variety of resources are provided to assist teachers in preparing students for the examination, including practice tests.

References

Bower, G.H. (1972). Mental imagery and associative learning. In L.W. Gregg (Ed.), Cognition in learning and memory (pp. 51-88). New York: Wiley.

Castro, D.P., & Shepherd, R.D. (Eds.) (1999). Aim higher: English language arts, Level H. Gloucester, MA: Advanced Instructional Materials.

Flower, L.S., & Hayes, J.R. (1981). A cognitive process theory of writing. College Composition and Communication, 32, 365-387.

Halio, M.P. (1996). Multimedia narration: Constructing possible worlds. Computers and Composition, 13, 343-352.

Haviland, V.S., & McCall, M.J. (1999). Transformation through technology: How HyperStudio updated middle school research. English Journal, 89(1), 63-68.

Hay, K., Guzdial, M., Jackson, S., Boyle, R., & Soloway, E. (1994). Students as multimedia composers. Computers and Education, 23, 301-317.

Hayes, J.R. (1996). A new framework for understanding cognition and affect in writing. In C.M. Levy & S. Ransdell (Eds.), The science of writing: Theories, methods, individual differences, and applications (pp. 1-27). Hillsdale, NJ: Lawrence Erlbaum.

Hayes, J.R., & Flower, L.S. (1980). Identifying the organization of writing processes. In L.W. Gregg & E.R. Steinberg (Eds.), Cognitive processes in writing (pp. 3-30). Hillsdale, NJ: Lawrence Erlbaum.

Jacobson, M.J., & Spiro, R.J. (1995). Hypertext learning environments, cognitive flexibility, and the transfer of complex knowledge: An empirical investigation. Journal of Educational Computing Research, 12, 301-333.

Jonassen, D.H., Pech, K.L., & Wilson, B.G. (1998). Learning with technology: A constructivist perspective. Columbus, OH: Merrill.

Larkin, J.H., & Simon, H.A. (1987). Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11, 65-100.

Lehrer, R. (1993). Authors of knowledge: Patterns of hypermedia design. In S.P. Lajoie & S.J. Derry (Eds.), Computers as cognitive tools (pp. 197-227). Hillsdale, NJ: Lawrence Erlbaum.

Lehrer, R., Erickson, J., & Connell, T. (1994). Learning by designing hypermedia documents. Computers in the Schools, 10, 227-254.

Liu, M., & Pedersen, S. (1998). The effect of being hypermedia designers on elementary school students' motivation and learning of design knowledge. Journal of Interactive Learning Research, 9, 155-182.

New York State Department of Education. (2001). The New York state school report card: An overview of academic performance. Albany, NY: The University of the State of New York: The State Education Department.

Nicaise, M., & Crane, M. (1999). Knowledge constructing through hypermedia authoring. Educational Technology Research and Development, 47(1), 29-50.

Paivio, A. (1971). Imagery and verbal processes. New York: Holt, Rinehart and Winston.

Petraglia, J. (1995). Writing as an unnatural act. In J. Petraglia (Ed.), Reconceiving writing, rethinking writing instruction (pp. 79-100). Mahwah, NJ: Lawrence Erlbaum.

SHANNON M. CARLIN-MENTER AND THOMAS J. SHUELL

State University of New York at Buffalo

USA

[e-mail]