Effects of Learner Control and Learning Strategies on English as a Foreign Language (EFL) Learning from Interactive Hypermedia Lessons.

This study investigated the effects of learner control, English learning strategies, and the use of advance organizers on English as a Foreign Language (EFL) learning from interactive hypermedia lessons. The experiment used 150 EFL subjects in Taiwan who studied English using one of several alternate versions of a hypermedia-based interactive video program. A 2 X 2 X 2 factorial experimental design was employed. The independent variables were: (a) variations of control--learner control or program control, (b) the use of advance organizers--presence or absence of advance organizers, and (c) students' English learning strategy ability--high ability or low ability as measured by Oxford's Strategy Inventory for Language Learning (S.I.L.L.). The dependent variables in this study included: (a) scores on immediate recall protocols, and

(b) students' attitudes toward learning English from interactive hypermedia lessons. Results of this study revealed significant effects of the learner control treatment and the use of advance organizers. Learner control was especially beneficial for students with lower ability in English learning strategy use. Students' attitudes were very positive toward learning English from interactive multimedia regardless of the treatments.

The issue of learner control in second language learning has been discussed by several researchers who have emphasized that effective language learning requires the learner's control and active involvement in the learning process (Liu, 1992; Oxford, 1993; Underwood, 1984). For instance, based on Krashen's (1982) assumptions about second language acquisition, Underwood (1984) categorized computer-assisted language learning (CALL) activities with linguistic dimensions by distinguishing communicative CALL from noncommunicative CALL. In communicative CALL, the learner is in control of the learning process, and students relate to subject matter in a personal way creating their own learning experience. In noncommunicative CALL, the program is in control of the learning process.

In the past, the issue of learner control in computer-assisted instruction (CAI) or multimedia has been investigated extensively but researchers have reported ambiguous results. In addition, existing studies have rarely addressed second language learners. Steinberg (1977) argued that many learner control experiments failed to show advantages of learner control because subjects used incorrect learning strategies. In addition to issues associated with learning strategies, researchers have noted that learner disorientation and cognitive overload are two challenges faced by instructional designers of hypermedia lessons (Jonassen, 1989; Kenny, 1992). To cope with this problem, Steinberg (1989) suggested that certain strategies might be embedded in hypermedia to help students.

Past research has demonstrated that advance organizers can have beneficial effects on memory and subsequent achievement (Ausubel, 1968). This view also aligns with the theory of communicative language teaching (Krashen, 1982), which stresses that while learning a second language, learners need to actively interact with the external environment and integrate new information with the information already stored in their memory. Although the effects of advance organizers on learners' comprehension of text have been demonstrated, very few studies have focused on how advance organizers influence EFL learning in a hypermedia-based learning environment.

Another factor that may play a role in students' language learning from multimedia is the language learning strategies that the learners themselves bring to the task. Language learning strategies are the various methods individuals have for perceiving and processing information while reacting to their environment. According to Oxford (1993), there is significant variation across levels in English learning strategy use among EEL students. Do EFL learners with fewer English learning strategies have difficulty learning in hypermedia-based lessons in the target language? This question has not been addressed.

This study investigated the following questions: Does learner control benefit EFL learners using hypermedia-based instructional materials? Do EFL students learning with advance organizers as an embedded learning strategy recall more information than EFL students learning without advance organizers? Do EFL students' English learning strategies influence their recall in hypermedia-based lessons?

Learner Control

Learner control in computer-based instruction has been extensively investigated in the past. Unfortunately, researchers have reported equivocal results. In a few studies, students given control over the amount of material tended to learn more or more efficiently in a CAI or multimedia learning environment (Mayer, 1979; Shyu & Brown, 1993). Although learner control is assumed to be motivating and effective to students, it has been shown most often to be only as effective or even less effective than program-controlled condition. Williams (1993), for example, reviewed over 70 research findings and summed up the ambiguity of learner control research: "Learner control should be better than program control; however, students left on their own do not uniformly make good use of such strategies" (p. 9-10).

In many studies of interactive learning programs, students actually learned less when they were provided choices that would allow them more control over their own instructional strategies (Gay, 1986; Ross & Rakow, 1981). Many other researchers have found no significant differences in achievement (Arnone & Grabowski, 1991; Lopez & Harper, 1989; Williams, 1993).

In summarizing the mixed findings regarding learner control in CAL/ multimedia lessons, researchers have attributed the discrepancies to learner characteristics, such as students' prior knowledge about the learning materials (Ross & Rakow, 1981), perception of control (Gay, Trumbull, & Smith, 1988), relevant ability (Shyu & Brown, 1993), and learning strategies (Hansen, 1982-83). Steinberg (1989) claimed that beginning level students in a particular subject area may lack the subject-specific learning strategies necessary for learner control to be effective. In addition to learning strategies researchers have also noted that learner disorientation and cognitive overload (Jonassen, 1989; Kenny, 1992) are two challenges faced by instructional designers of hypermedia lessons. To cope with this problem, Steinberg (1989) has suggested that certain strategies might be embedded in hypermedia to help students. The present study was an attempt to extend this line and to expand the base of empirical evidence in the are a of learner control and learning strategies in interactive hypermedia lessons.

The Advance Organizer

The concept of the advance organizer was developed by Ausubel (1968), who defined it as "...appropriately relevant and inclusive introductory materials...introduced in advance of learning...and presented at a higher level of abstraction, generality, and inclusiveness.. ."(p. 148). The nature of his theory is that if a new concept is to be understood, it should be related to a concept in the student's cognitive structure, which functions as a conceptual framework for those new concepts. Ausubel (1968) suggested that the use of advance organizers is the most efficient way to relate new concepts to their relevant concepts in the students' cognitive structures and consequently enhance learning and retention.

In the past, researchers have studied extensively the effectiveness of the advance organizer in a variety of disciplines. Nevertheless, reviews of the advance organizer literature revealed that support for such organizers is controversial. After reviewing 135 studies of the facilitative effect of advance organizers on learning and retention, Luiten, Ames, and Ackerson (1980) found a general positive effect for experimental groups over control groups in all content areas examined and with individuals of all grade and ability levels.

Despite the fact that advance organizers have frequently been shown to produce facilitative effects, there has also been research studies (Games, Lindbeck, & Griffin, 1987; Eastman, 1977) revealing that advance organizers do not facilitate learning. In a meta-analysis of the effects of advance organizers, Barnes and Clawson (1975) examined 32 studies incorporating the use of advance organizers and concluded that advance organizers generally did not facilitate learning.

Researchers have pointed out some reasons for the research discrepancies regarding the effectiveness of advance organizers, such as (a) the imprecise construction of advance organizers (Kenny, 1992), (b) failure to provide instruction for students about how to use the advance organizer (Bricker, 1989), (c) ignoring learner characteristics (Diptoadi, 1991), and (d) the lack of objective methods for qualifying and quantifying learning results (Clark & Bean, 1982).

Although the effects of advance organizers on learners' comprehension of text have been demonstrated, very few studies have focused on how advance organizers influence EFL learning in a hypermedia-based learning environment. The present study investigated the effectiveness of advance organizers on EFL students' learning in a computer-based interactive videodisc program while considering both learner characteristics and learner control.

Research Hypotheses

The following research hypotheses guided this study: (a) EFL students learning with learner control treatment in a hypermedia-based interactive multimedia lesson will recall more information than EFL students learning with program control treatment; (b) EFL students learning with the advance organizer as an embedded learning strategy in a hypermedia-based interactive hypermedia lesson will recall more information than EFL students learning without the advance organizer; (c) the provision of the advance organizer as an embedded learning strategy in a hypermedia-based interactive lesson will be more effective for EFL students in the learner control condition than those in the program control condition; (d) EFL students' language learning strategy use will interact with the instructional treatments (learner control vs. program control) in a hypermedia-based interactive lesson in ways that will influence their recall; (e) EFL students learning with learner control treatment will have better attitudes toward the instruction than students with program control treatment.

METHODS

This study employed a 2 X 2 X 2 factorial experimental design with three independent variables and two dependent variables. The independent variables were the variations of control of a hypermedia-based interactive program (learner control or program control), the provision of the advance organizer as an embedded learning strategy (with the advance organizer or without the advance organizer), and learners' level of English learning strategy use (higher or lower). The dependent variables were the amount of information recalled after viewing the hypermedia program and students' attitudes.

Subjects

One hundred and fifty Taiwanese EFL students volunteered to participate in this study. The age of the subjects ranged between 18 and 22 years old with an average of 7 to 10 years of previous English instruction. To ensure that all the subjects had familiarity with the computer system, a 15-minute pre-instructional program was designed for the subjects, which introduced the interactive videodisc system they used. Participation in the whole study involved approximately 100 minutes of each subject's time.

Materials

Five computer-based interactive videodisc programs were developed including one pre-instructional program and four versions of the instructional program corresponding to the four treatment conditions. All programs were developed using a Macintosh Quadra 900 computer and HyperCard (V 2.1)(Apple Computer Inc., 1991). Lehman IVD (V 2.3)(Lehman, 1993), an interactive video controller, was used to control the Pioneer LDV4400 videodisc player which provided video clips as part of the instructional programs.

Content of the instructional program. Many researchers have suggested the use of authentic video materials to foster students' second language learning (O'Banion, Goldman, Sharp, Vye, Bransford, Beaty, & Saul, 1993; Rivers, 1987). For example, students can watch speakers interacting through video materials and observe nonverbal behaviors as well as the types of exclamations and expressions that are used, how people initiate a conversation, how they negotiate meaning, and how they terminate an interactive episode. In this study, four instructional programs were developed by repurposing a commercially available videodisc named In the Holy Land (American Broadcasting Companies, Inc., 1989). Five historical events from the Unit of History were chosen for the lesson materials: (a) Holocaust and Immigration, (b) the UN Partition, (c) Proclaim the State of Israel, (d)

War of Independence, and (e) War and Refugees. A cause-effect relationship among these historical events existed. For each historical event, different factions (Jews or Arabs) have different interpretations. This conflict phenomenon and the cause-effect relationships are particularly good for constructing advance organizers because they match Ausubel's (1968) original definition of advance organizers as "...appropriately relevant and inclusive introductory materials.. .presented at a higher level of abstraction, generality, and inclusiveness..." (p. 148)

The advance organizer. An advance organizer was presented before the lessons in the advance organizer treatment conditions (Table 1). The construction of the advance organizer was based on following procedure (Bricker, 1989): (a) analyzing learning materials; (b) mapping the cognitive structures of learners; (c) determining characteristics of the advance organizer; (d) estimating the readability of the advance organizer; (e) checking the understandability of the advance organizer; (f) assessing the study time of the advance organizer; (g) evaluating the validity of the advance organizer; and (h) revising the advance organizer.

Structure of the instructional program. Although the four treatments were different in design, there were some commonalties. First, each of the four treatments was divided into five lessons: (a) Holocaust and Immigration, (b) the UN Partition, (c) Proclaim the State of Israel, (d) War of Independence, and (e) War and Refugees. Second, each lesson had two main parts: (a) Description of the historical event, and (b) Perspectives. For each part, both a video segment from the videodisc as presented on a separate video monitor and the transcripts of the segment as printed on the computer screen were provided. To understand the program, students had to exercise both their reading ability and listening ability. Additionally, students could get access to a Glossary that listed definitions of words that might be unfamiliar.

Four Treatments of the Instructional Program.

1. Learner control with the advance organizer (LC/AO)--In this treatment, the student went through the program based on his/her own choices, order, and pace. Rather than requiring the student to go through the whole program in the predetermined order, the student was provided a menu from which to select his/her preferred text or video segment. In each instructional segment, the student was able to repeat the segment, stop the segment, or continue at any given point (Figure 1). This design option was helpful for the student who wished to repeat individual sentences or stop the video segment temporarily to check the Glossary. He/she also could view the instructional sequence in any order. The student could go back to any segment he/she wanted to see again, and he/she was allowed to quit the program at any time. The advance organizer was provided before the lesson. The student was informed of the purpose of the advance organizer and how to use it before he/she read the advance organizer to ensure that it had the intended effect.

2. Learner control with no advance organizer (LC/NAO)--This treatment was the same as Treatment 1, but without the presence of the advance organizer.

3. Program control with the advance organizer (PC/AO)--Although the lesson content in this treatment was the same as the learner-controlled treatments, this treatment condition was totally controlled by the program (as specified by the designer). In this treatment, the pace and sequence were fixed and controlled by the computer (designer). Subjects under program control of pacing saw screens only for a set period of time, as determined by the amount of text on a given screen. To estimate the appropriate pace of the program, a reading-pace test was administered to 54 subjects who subsequently participated in the study. Six paragraphs chosen from an English-written text book, with a reading level judged by two composition teachers as equivalent as the lesson materials, were used. After reading each of the six paragraphs, the students were asked to record the time they needed to finish reading and the extra time they needed to understand each paragraph. The results collected from this test were translated into a formula that gave each subject approximately 1.2 seconds to process each word appearing on a screen. Based on these data, it was estimated that it would take about 45 minutes for students to complete the entire program in the program control treatment. As described in Treatment 1, the advance organizer was provided in advance of the lesson.

4. Program control with no advance organizer (PC/NAO)--In this treatment, the student was provided with the same program as treatment 3, but without the provision of the advance organizer.

Instruments

Oxford's Strategy Inventory for Language Learning. Oxford's Strategy Inventory for Language Learning (S.I.L.L.) has been translated into several languages. The Chinese translation of Version 7.0 (Oxford, 1990) was used to measure students' levels of English learning strategy use. This self-report questionnaire consists of 49 statements following the general format "I did such-and-such"; students respond on a five-point Likert-type scale. S.I.L.L. items are grouped according to the six categories in Oxford's (1990) strategy classification systems: (a) memory strategies, (b) cognitive strategies, (c) comprehension strategies, (d) metacognitive strategies, (e) affective strategies, and (f) social strategies. The internal consistency reliability of the S.I.L.L. using Cronbach's Alpha was reported to be .96 for a 1200 subject university sample and .95 for a 483 subject military sample (Oxford, 1990).

Attitude questionnaire. A Likert-scale questionnaire consisting of 16 items was developed. The questionnaire items were divided into three groups according to their purposes. Group one included items asking about students' general attitudes toward using computer-based interactive multimedia to learn English. Items of Group two tried to determine students' continuing motivation toward using interactive multimedia to learn English. Items of Group three, which asked about the presence of the advance organizer, were only for treatments with the advance organizer. Item 11, which asked, "As I worked on this program, I felt I had the control over the lesson," was examined individually.

Written recall protocols. The purpose of the recall protocol was to collect data for investigating students' comprehension of the lesson, either by reading the text on the computer monitor or listening to the people talking in the video. To ensure that each student obtained the same instructions on how to do the recall task, a list of guidelines was generated and printed on the recall sheet for students to follow. To analyze students' recalls, an idea unit analysis was carried out using the content-structure analysis recommended by Meyer (1975), which permits a detailed examination of the types of idea units recalled from a text. Three steps were used to analyze student protocols.

Step 1. The content-structure analysis. Each lesson in this program was transcribed to do the content analysis. To perform the analysis, first of all, the "rhetorical predicates" (Meyer, 1975), which are the sets of rhetorical relationships among ideas throughout the text, were identified. These included: "antecedent/consequent" (cause-effect relationship), "comparison," or "description." This produced a "macropropositional level." For example, in Event 1: HOLOCAUST AND IMMIGRATION [EVENT], the text was held together by the "cause-effect" relationship: "The rise to power of Hitler" led to the "Holocaust," and "Immigration of Jews" was the result of the "Holocaust." In Event 1: HOLOCAUST AND IMMIGRATION [PERSPECTIVES], the content was held together by a "comparison" relationship between the Israeli point of view, which focused on "security," and the Palestinian point of view, which stressed "immense impact." These macropropositional levels provided the skeletal framework for all subsequent levels.

In the next step of the analysis, the subordinating idea units were hierarchically arranged beneath the macropropositions to yield "micropropositional levels" using the rhetorical relationships as cues to guide the analysis. For example, in Event 1: HOLOCAUST AND IMMIGRATION [PERSPECTIVES], "...push ahead with the establishment of..." and "...increasing western sympathy for..." were descriptive points arranged beneath the "immense impact." The information provided in each successive micropropositional level thereafter decreased in its importance to the meaning of the text.

Step II. Scoring. Once the analysis was completed and the hierarchical representation of the text produced, these idea units were listed vertically in descending order and each idea unit was assigned a representative number. This procedure yielded 207 separate idea units, primarily consisting of nouns, verbs, and prepositional phrases. The levels were also assigned point values for scoring purposes in descending order, with the top level being awarded five points, the second highest level awarded four points, and so forth. This process awarded more points for idea units of greater importance to the meaning of the text and fewer points for idea units of less importance. For ease in scoring the recall protocols, the list then was rearranged in the order of their occurrence in the text. This list was then used as the instrument against which the student protocols were scored.

Students' protocols were scored independently by two raters against the list of ideas units. Scoring involved merely awarding the points for each idea unit in students' recall. In order to ensure that raters reliably rated each protocol, a set of scoring rules was developed. For example, points were awarded for alternative wordings of the idea units as long as the meaning was the same. In the scoring process, the raters did not know the treatment conditions of the subjects. Through the use of such rules, near-perfect agreement between raters in scoring protocols for idea units was achieved. The inter-rater reliability was calculated as r=.986. The scores from the two raters then were averaged to yield a score for that particular protocol.

Step III. Data analysis. Two steps were used to analyze the quantitative data collected. The first step was to compute the descriptive statistics (the mean and standard deviation) for the eight groups. Means and standard deviations were computed for English learning strategy use and written recall protocols. Second, factorial ANOVAs were employed to analyze the variance.

Procedure

Prior to the instructional phase of the experiment, all 150 subjects were administered the Chinese version of S.I.L.L. The results of S.I.L.L. were used to determine the attribute variable and level of language learning strategy use. The 60 students with the highest S.I.L.L. scores were assigned to the High Level group. The 60 students with lowest S.I.L.L. scores were assigned to the Low Level group. To minimize the regression of the scores to the mean, 30 students who scored in the middle were eliminated from the experiment. Thus, only 120 out of 150 subjects were included in the experiment. Students from each of the two groups (higher or lower) were randomly assigned to one of the four treatments with equal number of students in each treatment. Participation was voluntary, and 9 out of the 120 students failed to sign up for the experiment. Therefore, a final total of 111 subjects was included in the analysis of the experiment. Because of limited equipment, students participated in the instructional phase o ne-by-one according to appointments scheduled with the researcher. After an individual student finished the interactive hypermedia program, he/she was asked to complete the recall protocol of the lesson content as accurately as he/she could, and the student was told that he/she should write down as many ideas as possible. The attitude questionnaire was given to the student after the recall test. Because of the unbalanced numbers of subjects for each group, analysis of variance (ANOVA) was performed by using a SAS GLM procedure.

RESULTS

The means of the S.I.L.L. scores for High Ability and Low Ability groups were 3.31 and 2.14 respectively. Analysis of variance (ANOVA) revealed that the difference of means between these two groups was significant (F = 2l7.33,p = .0001) (Table 2).

The means of recall scores for different treatments showed that the students in the Learner Control condition (Mean = 23.54) demonstrated superior recall compared to the students in the Program Control condition (Mean = 18.20). The results of the GLM ANOVA showed that there were significant differences of recall scores as a result of variations of control with Learner Control treatment superior to Program Control treatment (F = 8.95, p = .0035). Besides, subjects who experienced the treatment with the advance organizer recalled significantly more than their counterparts who did not have the advance organizer treatment (F=6.23, p=.0142) (Table 3).

The means of recall scores between High Ability and Low Ability groups were almost the same (Mean = 20.93, Mean = 21.05, respectively). The GLM ANOVA showed that Low Ability strategy users recalled as well as High Ability strategy users for the immediate recall protocols (F = 0.00, p = .9947) (Table 3). The results of the GLM ANOVA for recall scores also indicate that there were significant differences as a result of the interaction of the control condition and students' level of language learning strategy use (F = 4.98, p = .0277). Figure 2 shows a graphical representation of this interaction effect. Whereas learners classified as High Ability strategy users performed at a similar level in both the learner control and program control conditions, those classified as Low Ability strategy users performed much better in the learner control condition. However, there were no significant differences in recall scores as a result of the interaction of variations of control and the provision of the advance organizer (F = 0.42, p = .5201).

For the Attitude survey, the GLM ANOVA suggested that subjects who had the Program Control treatment and subjects who had the Learner Control treatment shared similar attitudes toward learning English from hypermedia-based CBIV lessons (F= O.67, p = .4162).

DISCUSSION

The results of this study revealed that EEL students in the Learner Control condition significantly outperformed EEL students in the Program Control condition on the recall test. It is therefore suggested that, for EFL learning, a learner-controlled environment is better than a program-controlled environment. This finding is consistent with existing second/foreign language learning theories such as communicative language learning and Krashen's (1982) assumption of second language acquisition, which emphasize that while learning a second language, learners need to actively interact with the external environment. Specifically, second/foreign language learners should be allowed a certain degree of control over the learning environment.

For language-learning purposes, a learner-controlled environment provides many advantages that are not available in a program-controlled environment. In this study, for example, students in the Learner Control groups could go through the lessons at their own pace and according to any desired sequence. In addition, they could freely explore the facilities embedded in those lessons. For instance, if they did not quite understand the viewpoints of the Palestinian historian toward the War of Independence and needed to watch the video segment again, they could always click on the Video button to replay the video segment. If they encountered unfamiliar word(s) while watching the video segment, they could always stop the video temporarily and click on the Glossary button to check out the meanings as well as the Chinese translations of the word(s). After they got the meanings of the unfamiliar word(s), they could continue the video segment simply by clicking on the Re-Start button. Although the Program Control and L earner Control versions of the program shared the same content, the Program Control version did not allow the students to seek just-in-time help and pursue learning according to their own needs and wishes. The Learner Control condition, on the other hand, put hypermedia resources at the learners' disposal.

Such a finding, however, does not contradict Jonassen's (1989) view that navigation through hypermedia is an obvious problem often encountered by beginning level students in a particular subject area. The problem of navigation may still exist among EFL students who have control of their learning environment, but the advantages the learner can get from Learner Control programs--such as flexibility, interactivity, and the network of multimedia sources--outweigh the potential disadvantages of navigational problems. Nonetheless, the problem of navigation in hypermedia-based learning environment might still exist and should not be overlooked. Therefore, efforts should continue to develop mechanisms to help learners cope with the problem of navigation in hypermedia in order to develop a really suitable learning environment for EEL students.

The results of the study showed that the effects of advance organizers on recall scores were significant. Students receiving the advance organizers had significantly higher recall scores than students not provided with the advance organizers. However, the interaction effects between the control condition and the advance organizer condition were not obvious. The advance organizer helped both Learner Control groups and Program Control groups. The students benefited from the advance organizer that functioned as a conceptual framework for students to relate those lesson concepts to their relevant concepts in their cognitive structures, and consequently enhanced their learning. The finding suggests that the advance organizer, if constructed appropriately, has effects on EEL learning from interactive hypermedia lessons, and works equally well for Learner Control and Program Control programs.

In addition, the results indicate that the advance organizer can significantly improve the comprehension of both High Ability English learning strategy users and Low Ability English learning strategy users. The finding that the advance organizer works equally well for High Ability and Low Ability strategy users as well in Learner Control and Program Control programs supports Ausubel's (1968) proposition that the advance organizer is an efficient way to introduce material for all learners.

Somewhat surprisingly there were no significant differences in recall scores between High Ability groups and Low Ability groups. In other words, interactive hypermedia lessons helped all of the students be more effective and efficient EEL learners. This finding does not imply that the students in High Ability groups did not learn effectively in those lessons. The importance of the finding is that an interactive hypermedia environment is able to help the students in Low Ability groups to become more effective learners. In a sense, interactive hypermedia provides an ideal environment to do language learning strategy training for students with lower levels of language learning strategy use. However, more research is needed to confirm this finding.

The results also revealed that there was an interaction effect between variations of control and students' English learning strategy use. For Low Ability groups, students in Learner Control groups scored higher on the recall test than did students in Program Control groups. For the High Ability group, the differences between Learner Control treatment and Program Control treatment were not significant. Compared with High Ability groups, those learning facilities provided in the Learner Control treatment had more significant effects for Low Ability groups. As discussed in the previous section, the Learner Control treatment provided more learning facilities than the Program Control treatment. If the learner makes use of those learning facilities, they may stimulate or substitute for the student's own learning strategies. Therefore, it is suggested that learner-controlled hypermedia lessons have particular value as training environments for students who are not effective in learning English in real situations.

With respect to students' attitudes toward learning English from interactive hypermedia lessons, these students showed very positive attitudes toward learning English with interactive multimedia, regardless of experimental treatment. Since none of the subjects had used interactive multimedia before, such a result is not surprising. This study suggests that similar research using the same subjects should be conducted in the future to see if there is an attitude change of these students toward learning English from interactive hypermedia lessons.

CONCLUSIONS

This study indicated that a learner control treatment and the use of advance organizer are beneficial for EFL students learning English from interactive hypermedia lessons. Learner control is especially beneficial for students with lower English learning strategy use, who became more effective EFL learners. Findings of this study suggest the following practical applications in EFL learning:

1. Because students learn English better in a learner control condition, teachers should provide a more interactive environment with choices for users to guide their own learning in their EFL curriculum. For software designers, EFL programs should consider the issues of learner control and interactivity in their programs.

2. The results of this study confirm the usefulness of advance organizers in interactive hypermedia lessons in EFL learning. The effects of advance organizers should encourage software designers to integrate them in the software as a conceptual framework for students.

3. The design and development of EFL programs should consider EEL students' characteristics. For students with lower levels of English learning strategy use, interactive multimedia provides a good environment to do the English learning strategy training. Future research needs to examine other learner characteristics in learning from interactive multimedia.

References

American Broadcasting Companies, Inc. (1989). In the holy land. [Videodisc]. St. Paul, MN: American Broadcasting Companies.

Apple Computer Inc. (1991). HyperCard [Computer program]. Cupertino, CA: Apple Computer.

Arnone, M.P., & Grabowski, B.L. (1991). Effects of variations in learner control on children 's curiosity and learning from interactive video. (ERIC Document Reproduction Service No. ED 334 972)

Ausubel, D.P. (1968). Educational psychology: A cognitive view. New York: Holt, Rinehart, & Winston.

Barnes, B.R., & Clawson, E.U. (1975). Do advance organizers facilitate learning? Recommendations for further research based on an analysis of 32 studies. Review of Educational Research, 45(4), 63 7-659.

Bricker, E.J. (1989). The effect of advance organizers in the teaching of science. Master thesis. NJ: Kean College. (ERIC Document Reproduction Service No. ED 313 682)

Carnes, E.R., Lindbeck, J.S., & Griffin, C.F. (1987). Effects of group size and advance organizers on learning parameters when using microcomputer tutorial in kinematics. Journal of Research in Science Teaching, 24(9), 78 1-789.

Clark, C.H., & Bean, T.W. (1982). Improving advance organizer research: Persistent problems and future directions. Reading World, October, 1982, 2-16.

Diptoadi, V.L. (1991). Effects of prereading strategies on ESL reading by Indonesian college students having different characteristics. (ERIC Document Reproduction Service No. ED 338 036)

Eastman, P.M. (1977). The use of advance organizers for facilitating learning and transfer from quadratic qualities. School Science and Mathematics, 77, 377-384.

Gay, G. (1986). Interaction of learner control and prior understanding in computer-assisted video instruction. Journal of Educational Psychology, 78(3), 225-227.

Gay, G., Trumbull, D., & Smith, J. (1988). Perceptions of control and use of control options in computer-assisted video instruction. Tech Trends October, 1988.

Hansen, D.W. (1982-83). User-controlled technology in the humanities. Journal of Educational Technology Systems, 11(2), 13 1-142.

Jonassen, D.H. (1989). Hypertext/hypermedia. Englewood Cliffs, NJ: Educational Technology.

Kenny, R.F. (1992). The generative effects of instructional organizers with computer-based interactive videodisc. (ERIC Document Reproduction Service No. ED 348 002)

Krashen, 5. (1982). Principles and practice in second language acquisition. New York: Pergamon.

Lehman, J.D. (1993). Lehman IVD V.2.3. [Computer program]. West Lafayette, IN: Purdue University.

Liu, M. (1992). Hypermedia-assisted instruction and second language learning: A semantic-network-based approach. (ERIC Document Reproduction Service No. ED 349 954)

Lopez, C.L., & Harper, M. (1989). The relationship between learner control of CAI and locus of control among Hispanic students. Educational Technology, Research and Development, 37(4), 19-28.

Luiten, J., Ames, W., & Ackerson, G. (1980). A meta-analysis of the effects of advance organizers on learning and retention. American Educational Research Journal, 17(2), 211-218.

Mayer, R.E. (1979). Can advance organizers influence meaningful learning? Review of Educational Research, 49(2), 371-383.

Meyer, B.F.J. (1975). The organization of prose and its effect on memory. Amsterdam: North Holland Publishing.

O'Banion, K., Goldman, S., Sharp, D., Vye, N., Bransford, J., Beaty, J., & Saul, E. (1993). Multimedia support for language comprehension skills in at-risk kindergarten students. Paper presented at the annual meeting of the American Educational Research Association, Atlanta, Georgia.

Oxford, R.L. (1990). Language learning strategies: What every teacher should know. New York: Newbury House/Harper & Row.

Oxford, R.L. (1993). Language learning strategies in a nutshell: Update and ESL suggestions. TESOL Journal, Winter, 1992/1993, 18-22.

Rivers, W.M. (1987). Interactive Teaching. New York: Cambridge University.

Ross, S.M., & Rakow, E.A. (1981). Learner control versus program control as adaptive strategies for selection of instructional support on math rules. Journal of Educational Psychology, 73(5), 745-753.

Shyu, H.Y., & Brown, S.W. (1993). Learner control: The effects during computer-based videodisc instruction. Paper presented at the annual meeting of the American Educational Research Association, Atlanta, Georgia.

Steinberg, E.R. (1977). Review of student control in computer-assisted instruction. Journal of Computer-Based Instruction, 3, 84-90.

Steinberg, E.R. (1989). Cognitive and learner control: A literature review, 1977-1988. Journal of Computer-Based Instruction, 16(4), 117-121.

Underwood, J. (1984). Linguistics, computers, and the language teacher: A communicative approach. Rowley, MA: Newbury House.

Williams, M.D. (1993). A comprehensive review of learner-control: The role of learner characteristics. Paper presented at the annual meeting of the Association for Educational Communications and Technology, New Orleans, LA.

Table 1

Sample Advance Organizer

There exists a cause-effect relationship between many historical events. Taking modern Chinese history for example, the weak Manchu dynasty resulted in the Chinese Revolution in 1911 and the establishment of a republic of 1912. In fact, the cause-effect relationship also can be seen in the history of the Middle East. The history of the Middle East shows that bitter political disputes led to war several times in the Middle East in the past. Those wars then caused some other political events, and have kept the area in constant turmoil and have threatened the peace of the world. We can find such a cause-effect relationship in many historical events happened in the past of the Middle East.

Table 2 ANOVA for S.I.L.L. Scores by Students' Strategy Group
Source                   DF     SS     MS  F Value
Student Strategy Group    1  12.71  12.71   217.33
Error                   109   6.37   0.06
Total                   110  19.08
Source                  Prob[greater than]F
Student Strategy Group                .0001
Error
Total
Table 3 GLM ANOVA for RecallScores
Source         df  Type III SS      MS  F Value  Pr [greater than] F
CTRL            1       702.52  702.52     8.95               0.0035
AO              1       488.74  488.74     6.23               0.0142
LEVEL           1         0.00    0.00     0.00               0.9947
CTRL x AO       1        32.68   32.68     0.42               0.5201
CTRL x LEVEL    1       391.04  391.04     4.88               0.0277
AO x LEVEL      1        37.92   37.92     0.48               0.4885
CTRL x AO x
LEVEL           1        36.51   36.51     0.47               0.4967
Error         103      8081.20   78.46
Total         110      9875.95
CTRL = Control
AO = Advance Organizer
LEVEL = Level of Strategy Use
Figure 2. Plot of the interaction effect between the variations of
control and students' level of language learning strategy use for
all groups Variations of Control
      LC     PC
High  42.86  41.06
Low   50.42  32.29
Note: Table made from line graph