Learning Strategies and Hypermedia Instruction.

Students are frequently being asked to learn from hypermedia. Some characteristics of hypermedia, including the use of multiple media and the increased control over sequence provided to the learner, may impede learning for the less strategic student. This study investigated the potential of incorporating learning strategy instruction into hypermedia learning materials. Using an intact high school computer class, a sixweek intervention was conducted where the experimental group participated in learning strategy training in conjunction with hypermedia computer networking lessons. Pre and post measures of metacognitive awareness and achievement are reported. The results indicated that student's regulation of their own cognition benefited from strategy training but their knowledge of cognition and achievement was commensurate with a control group. The discussion includes reasons why strategy instruction through hypermedia may not be successful and provides suggestions for future research.

Learners are being asked to use hypermedia instructional materials more and more in today's educational communities. This increase in use has outpaced understanding of how students can learn effectively in such environments. Several characteristics of hypermedia may serve to impede some learner's ability to comprehend the material presented. These characteristics include novelty, the multiple modes of presentation (e.g., excessive cognitive load), the nonlinear paths, and increased learner control. Most of these characteristics have been cited as reasons to be encouraged by the potential of hypermedia for instruction, but they also raise numerous questions about the learner's ability to use the environment to promote his or her understanding. It is likely that only the "strategic" learner will benefit from hypermedia instruction.

Learning strategies, often referred to in the literature as cognitive and metacognitive strategies, have been intensely researched in the past 20 years. Learning strategies such as summarizing text prior to reading and setting goals have been demonstrated to be effective in traditional environments (Pressley & Woloshyn, 1995). As in other learning situations, it is important that the student effectively use learning strategies in an environment that makes extensive use of hypermedia. Unfortunately, little is known about the use of strategies in this mode of learning that is quickly becoming commonplace (Alexander, Graham, & Harris, 1998). The goal of this study was to determine if learning strategies can be taught in a hypermedia learning environment and if the learning strategy instruction has a positive impact on achievement.

Hypermedia learning environments provide opportunities to incorporate diagrams, video clips, sound bites, and text into instruction. A common approach to designing hypermedia instructional materials is to have a main path, similar to sequential pages in a book that students should follow. In addition to this main path, other features provide alternative representations of the same concept or related information. Each of these additional features is intended to increase student comprehension by providing more information when the student needs or requests it. A common example is linking a recently introduced word to its definition. If the learner does not remember the term, they can select the word to see a definition. Other features that are commonly available include a links to objectives, site-maps, and self-check questions. However, these features are of little value if they are not used or used inappropriately.

Learning Strategies

Increasing comprehension when reading from text has been studied extensively. A great deal is known about how good readers learn from text (Baker & Brown, 1984) and about how text should be formatted to encourage comprehension (Keller & Burkman, 1993). Reading comprehension depends in part on the effective utilization of cognitive resources (Baker &Brown, 1984). This should have a dramatic impact on how the instructional use of hypermedia is approached. Learning strategies, such as summarizing text, can improve information processing by more efficiently using cognitive resources. Research concerning learning strategies is extensive. Studies by Weinstein (1978) in the early 1970s represent some of the first strategy instruction studies. Weinstein examined the effectiveness of elaboration strategies on recall. Training activities in these studies asked students to engage in tasks that resulted in deeper processing of the content being read. Weinstein also conducted interviews with Army recruits and college stu dents in hopes of identifying common learning strategies. The strategies reported from these interviews have been widely cited as starting points for research investigating the effectiveness of different learning strategies.

Why learning strategies are so effective can be described in terms of metacognition. Flavell is credited with the introduction of the term metacognition in the mid-1970s (Hartman, 1998). Metacognition literally means thinking about thinking. Metacognition has been subdivided into knowledge and regulation of cognition (Jacobs & Paris 1987). Knowledge of cognition refers to a learner's understanding of his or her own thought processes (Schraw, 1998). Knowledge of cognition has been further divided into declarative knowledge (about), procedural knowledge (how), and conditional knowledge (when) (Jacobs & Paris, 1987; Schraw, 1998). The use of a strategy is dependent on the student's awareness of the strategy (declarative), understanding of how the strategy works (procedural) and knowing when to use the strategy (conditional). Regulation of cognition is understood as those behaviors that demonstrate control of, and/or utility with, one's knowledge of cognition. Examples would include a student's ability to monito r, evaluate, and plan for their learning (Schraw, 1998).

Metacognition thus relies on one's knowledge of their abilities and strategies that may improve their own learning. Metacognition also depends on the student's ability to regulate their thinking in a way that takes advantage of their knowledge of cognition. For example, if a student knows that they have a difficult time remembering dates, they may develop a timeline to better facilitate memorization. The understanding of this difficulty with dates is an example of the student's knowledge of his or her own thinking. A timeline is an example of knowledge of cognition in that it represents a possible cognitive strategy. The decision to use the strategy is an example of regulation of cognition. Therefore, the effective use of strategies is dependent on several existing factors, not the least of which is the ultimate decision on the part of the learner whether or not to use a strategy. The learner's capacity in each of the above areas is vital to success in any learning environment--including hypermedia instructi on. The learners repertoire of reading strategies such as summarization (knowledge of cognition) and their willingness to invoke such strategies (regulation of cognition) when appropriate, have a dramatic effect on their understanding.

Research suggests that each of these skills can be taught and can subsequently improve academic achievement (Hattie, Biggs, & Purdie, 1996). Hattie et al. conducted an extensive meta-analysis of learning skill interventions. Fifty-one strategy instruction studies were reviewed in the paper. They found an average effect size of 0.45 in the studies. One of the major conclusions of the review was that strategy instruction should be contextual. In other words, the strategy instruction should be incorporated into the normal curriculum as opposed to a stand-alone topic. Rosenshine, Meister, and Chapman (1996) reviewed intervention studies that investigated the use of question generation as a cognitive strategy to improve comprehension. The strategies reviewed used several approaches to foster active processing in students. Procedural prompts were used in virtually all of the studies reviewed by Rosenshine et al. These prompts served as scaffolds in teaching specific cognitive strategies. An example may ask a stude nt to identify the main idea in a paragraph. The generation of questions requires active processing from the learner and encourages comprehension monitoring. They found an average effect size of 0.36 on standardized tests and 0.86 on experimenter-developed tests. The two reviews taken together indicate that learning strategy instruction is an effective way to improve comprehension.

Learning Strategies and Hypermedia

Using computers to deliver content instruction is common. However, using computers to deliver learning strategy instruction is not. Few studies have addressed the potential of hypermedia environments to teach learning strategies and encourage their use. Derry and Murphy (1986) reviewed learning strategy research and theory in hopes of better informing the design of systems that could promote learning ability. Their review revealed several pertinent findings. The first was that two distinct approaches to learning strategy training exist. One approach is to teach the strategies in a setting detached from the mainstream curricula. Another approach is to provide instruction that is embedded within the curricula. Deny and Murphy concluded that the most reasonable approach was a combination of the two that (a) provides initial detached instruction as a foundation and (b) provided opportunities to apply the strategy in a realistic setting.

Puntambekar (1995) developed a computer assisted instructional system called Metacognition in Studying from Texts (MIST) to train students to think about what they are reading. This system suggests relevant activities during three stages of the reading process; planning, reading, and post-reading. The computer-based tutor was designed to engage students in metacognitive activities that enhance comprehension. For example, students were instructed to interact with the tutor before reading a text through several planning activities. These activities provide suggestions for approaches that are appropriate for the type of reading to be done. In one study, Puntambekar and deBoulay (1997), attempted to learn how students' planning and monitoring activities changed as they used MIST. In this study students used the MIST system in five different reading sessions. They measured the planning and monitoring done by the students by measuring the time spent in these sections of the MIST system. They discovered that stude nts initially increased the amount of time working on the activities and as time passed the amount of time on activities decreased. Puntambekar and deBoulay claimed that the initial increase in time demonstrates an increase in planning activities and the subsequent decrease is due to developed automaticity of planning activities. Unfortunately, numerous explanations other than increased planning could be used to describe the differences in initial time on task (e.g., lack of planning). This study failed to control for the type of activity through the course of the study. The first two activities required students to read text passages to acquire the details, the next two activities required students to read for the gist, and the fifth and final activity was described as a "revision" activity. The only conclusion that can be drawn from this study is that high and low ability students differed in how quickly they learned the program. Controlled research using this type of system holds a good deal of potential for better informing the instructional design of strategy instruction.

Cardinale and Smith (1994) used a computer-based tutorial about microcomputer components to investigate the effectiveness of learning strategy instruction. They presented 60 adults with learning strategy training incorporated into a computer-based tutorial. They found that those receiving the learning strategy enhanced tutorial for review scored significantly better on measures of their knowledge of microcomputer components. Two types of learning strategy tutorials were used. One emphasized memory strategies while the other emphasized comprehension. The memory strategy tutorial had a greater effect on recall while the comprehension tutorial was more effective for conceptual questions. This approach can be distinguished from MIST in that it more closely incorporated the strategy instruction with the content.

Dehn (1997) conducted a study that introduced students to two types of strategy training and two types of computer presented text. The two types of strategy training were "informed" and "non-informed." The informed strategy training included instruction on the significance of the strategy use and used a more didactic instructional approach. The non-informed training was less explicit about the strategies and the rationale for their use. This form of training emphasized scenarios that were more likely to elicit strategy use. Like the current study, Dehn hypothesized that strategy instruction would improve metacognitive awareness and subsequently improve achievement. The computer-based environment described in the Dehn study is similar to the computer curriculum used in the current experiment. The Dehn curriculum afforded the opportunity to review definitions, main ideas, and pictures as well as decrease the text difficulty. Unlike this study, these options were not available in the control group version of th e text. Dehn hypothesized that the participants using the version with the assistance options would demonstrate improved comprehension. In Dehn's study, the "control" group also received a form of strategy training. Therefore, this was not a controlled observation of the effect of strategy training on achievement. The study merely compared to different approaches. The results of this study indicated that the informed strategy instruction group demonstrated a higher level of reading awareness and strategy knowledge than the non-informed strategy group. This is questionable because the measures simply indicate that informed strategy subjects are better able to identify the mechanisms used, not that there is any increase in use. Subjects who received non-informed strategy instruction received less information about terminology and instead focused on indirect approaches that would encourage strategy use. Dehn also found that students who received the more direct strategy instruction were more likely to use the co mputer assistance options such as definitions and main ideas. On measures of comprehension, no main effect was found for either type of strategy training or availability of assistance options. Contrary to these results, one would expect the increased use of assistance to positively impact comprehension.

Taken as a whole, this research indicated that instruction holds significant promise in the incorporation of strategy instruction into content learning. Studies with a true control group are needed to clearly identify changes in strategy use and subsequent achievement when strategy instruction is provided. Characteristics that are likely to intervene in the success of such instruction include the degree to which the instruction is integrated into the content, the target outcomes (recall vs. conceptual understanding) and the availability of online assistance (definitions, etc.).

Purpose of the Study

The purpose of this study was to determine if learning strategy instruction provided in conjunction with content instruction, that is hypermedia based, could improve strategy use and subsequent achievement. More specifically, two questions are addressed. Number one, can learning strategies can be taught effectively using a student-driven, hypermedia instructional materials? Previous research has failed to adequately demonstrate the effectiveness of this mode of instruction for learning strategies because it failed to adequately control for the inclusion of strategy instruction. Question number two, will this strategy instruction subsequently improve achievement? Previous research in strategy instruction has clearly indicated that the use of learning strategies will improve comprehension, however, early research with hypermedia has produced contradictory results (Dehn, 1997).

The first question, can learning strategies can be effectively taught in a hypermedia environment, is important for a number of reasons. Most descriptions of strategy training require that the training include an extensive treatment of each strategy to be learned. For example, Graham and Harris (1993) advocated a seven step process that includes preskill development, an initial conference, discussion, modeling, memorizing, practicing with peers, and independent performance. Pressley's Good Strategy User model advocated a process that incorporates strategies, metacognition, motivation, and knowledge into one instructional approach (Symons, Snyder, Cariglia-Bull, & Pressley, 1989). This study investigated the necessity of such extensive strategy instruction. This leap was predicated on a number of assumptions. Number one, it is assumed that this was not the first time students were exposed to these strategies and therefore the extensive treatment is unnecessary. A second assumption was that the reason secondar y students often resisted the use of strategies was not because they were unaware of them, but because they lacked the necessary desire to implement such an effortful approach. A third assumption was that by closely tying the strategy instruction to the content instruction, a less rigorous approach was necessary. In light of the purposes of this study and these assumptions, the following hypotheses were tested. The first hypothesis was that students receiving limited strategy instruction incorporated into hypermedia lessons will demonstrate improved awareness and use of learning strategies. The second hypothesis was that students receiving strategy instruction will exhibit superior performance on a posttest measuring content understanding. This hypothesis is congruent with current research on learning strategy instruction (Hattie et al., 1996).

Schraw and Dennison's (1994) Metacognitive Awareness Inventory (MAI) was used to test the first hypothesis by measuring the students' knowledge about learning strategies and their use prior to and following the intervention. This self-report measure asks students questions about the knowledge and use of learning strategies, both cognitive and metacognitive. The second hypothesis of this study predicted that the learning strategy instruction would subsequently improve achievement. A content exam addressing material introduced during the intervention was used to measure achievement after the intervention. It was predicted that students receiving learning strategy instruction in conjunction with their content instruction would demonstrate a significant gain in learning strategy use and understanding as measured by the MAI when compared with students who did not receive the learning strategy instruction. It was also predicted that the students receiving the learning strategy instruction would outperform the othe rs in a measure of achievement as measured by a multiple-choice content exam.

METHOD

Participants

Participants were enrolled in a large high school in a large suburban high school in the central United States. This progressive high school partnered with a local software company to provide a computer-networking course for 11th and 12th graders. Each of the participants in the study was enrolled in the course.

Materials

The materials used in this study included lessons for the computer-networking component of the course, strategy instruction lessons and task journals.

Computer networking lessons. The computer networking content was delivered exclusively by way of the hypermedia curriculum. The curriculum was developed by a networking hardware company and is highly technical. The hypermedia format allowed the students to easily navigate within and between lessons. Each lesson presented the student with several pages of text, which they could read at their own pace. Within the text were buttons the students could push to view relevant diagrams. In addition to the content, each lesson contained review questions, lesson goals, and a practice quiz. The review questions discussed material from the previous lesson. The lesson goals provided an overview of the lesson. The practice quiz was a series of 10 true and false questions concerning the material from that lesson. Students could respond to the question and then receive immediate feedback. Some words are hyperlinked to definitions.

Learning strategy instruction lessons. The learning strategy instruction was also presented to students by way of computer in a hypertext format. The strategy instruction lessons were each displayed on one page. Each strategy instruction page corresponded to a computer network lesson. The strategy instruction pages described a learning strategy, provided an example and assigned a related task for each particular lesson. A task journal template was designed for each week of lessons. Students were told to record the results of the assigned task in the appropriate area of the task journal. For example, during the second week of the intervention, students read strategy instruction pages that discussed setting goals. One day focused on short-term goals. The journal task for that day was to answer questions about short-term goals and to set several for the week.

Instruments

Student's strategy use and understanding was measured using the MAI (Schraw & Dennison, 1994) before and after the intervention. This 52 item self-report measure addressed the student's knowledge and regulation of cognition. Participants noted their agreement with each statement by responding to each item on a seven point Likert scale. A response of one indicated that the statement was not true of the individual. A response of seven corresponded to a statement that the individual felt was true of them. Seventeen knowledge of cognition items measured students understanding of strategies, when to use strategies and their own cognition. An example of a knowledge of cognition item is; "I understand my intellectual strengths and weaknesses." Thirty-five regulation of cognition items measured metacognitive constructs such as student's ability to monitor their own learning. An example of a regulation of cognition item is "I consider several alternatives to a problem before I answer."

Achievement was measured with a content exam that was given at the conclusion of the six-week study. Another content exam was given prior to the intervention. This exam measured the student's understanding of the computer-networking concepts presented up to that point and served as a pretest for the study.

Procedures

The study was conducted over a six-week period. The 13 participants were randomly assigned to one of two conditions: (a) the strategy instruction group (n=7) or (b) the control group (n=6). The students in the strategy instruction group received learning strategy instruction and completed a related journal task four days out of the week. These activities took 5-10 minutes of time each day. The control group did not receive the strategy instruction and their additional time was spent on activities for other course components.

All students completed the MAI and a computer knowledge test prior to the intervention. The first day of the intervention the students in the strategy group were introduced to the strategy instruction materials and given a brief explanation of what strategies were and how they may be helpful in learning.

Students in the treatment group began class each day by participating in a 5-10 minute independent learning activity. The learning activities introduced students to a learning strategy. Table 1 lists the topics that were addressed each week. The control group will simply begin the computer-networking lesson. The course is structured so that they can immediately switch to tasks related to another course upon completion of the day's lesson. As the lessons were self-paced, students finished at different times regardless of their placement in the experimental or control group.

A new skill was taught during the first day of each week and expanded upon during the next three days. The strategy lessons included the following components, which were adapted from Graham and Harris (1993):

1. initial presentation--discuss current strategies and benefits of new;

2. explain the use of strategy, how and when;

3. provide an example of the strategy being used;

4. memorizing the strategy; and

5. performing independently.

For example, during the week when students learned about setting goals as a learning strategy, an initial conference was conducted before the intervention. The next four components, although not explicitly directed, were addressed during the week's activities.

RESULTS

The resulting data were analyzed using a mixed-subjects repeated measures design. Time provided the within group variable (pre and posttest) while, the between group variable was instruction type (strategy instruction or control). An alpha level of .05 was used for all statistical tests. It is important to note that the formation of treatment and control groups by random assignment of the participants resulted in groups that differed significantly in terms of their pretest exam scores F(1,11) = 5.73, p[less than].05, (strategy instruction, M 60.86, SD = 15.05, no strategy instruction, M = 75.67, SD = 11.96). Evaluation of results from nonequivalent groups is troublesome. There is no guarantee that differences found in the study reflect differences resulting from instructional approach.

Hypothesis 1-Learning Strategy Use

Hypothesis 1 predicted an increase in the learning strategy awareness by students receiving learning strategy instruction. Therefore, the scores of the MAI were expected to be greater than the control group on the posttest observation. Coefficient a for the entire MAI was .95 on the pretest and .96 on the posttest. The scores are described in Table 2 and Figure 1. The MAI contains items that measure knowledge of cognition and regulation of cognition independently. By analyzing the scores separately, a clearer picture of the differences between groups can be seen. Table 3 and Figure 2 describe the knowledge of cognition scores for each group. Differences between groups were negligible on the knowledge of cognition items on the pre and posttests. No significant main effect was found for instructional approach on the knowledge of cognition scores. Table 4 and Figure 3 describe the regulation of cognition scores for each group. The differences between groups on the regulation of cognition posttest were large (st rategy instruction, M = 154.14, SD = 39.05, control, M= 137.17, SD = 17.14), but did not reach statistical significance. Statistical significance indicates the probability that the two different scores obtained from random samples reflect a true difference in the population. Statistical significance does not reflect the importance or magnitude of the differences between treatments and is very dependent on the sample size of an experiment (APA, 1994). Effect sizes are measures of magnitude and are particularly important considerations in interpreting results of studies with small sample sizes. An effect size was calculated using the difference between the pretest and posttest group means, divided by the pooled standard deviation. The calculated effect size for the differences between groups in regulation of cognition scores = 0.50 standard deviations. In terms of practical significance, this is an important difference.

Hypothesis 2--Improved Performance

Hypothesis two predicted the strategy instruction group would outperform the control group on a posttest measuring content understanding. A mixed-subjects, repeated measures design was also used to analyze performance on the content exams. Time (pre and posttest) provided the within group variable while the between group variable was with or without strategy instruction. Results for the content exams can be found in Table 5 and Figure 4. There was no main effect for instructional approach on the content exams.

Results Summary

Strategy instruction did not impact students knowledge of cognition as measured by the MAI. Although not statistically significant (arguably due to low power), strategy instruction did appear to have a substantial positive impact on the students' regulation of cognition when compared to the control group. This positive impact did not however, translate into improved performance on the content exam that followed the intervention.

There is some indication that the strategy lessons may have improved students' metacognitive awareness of regulatory strategies. Little evidence was found to indicate that general cognitive strategy use improved or that the increased awareness improved students' performance. The knowledge of cognition items is more indicative of cognitive strategy use such as summarization.

Why students' knowledge of cognition would remain the same and regulation of cognition would increase may be partially answered by indications that students' already possessed a substantial repertoire of strategies. Little gain would be expected on a measure of cognitive strategy use if use was already very high. Development issues may also be at play and will be addressed further in the summary and conclusions section.

DISCUSSION

Hypothesis 1--Learning Strategy Use

The first hypothesis in this study stated that learning strategies can be taught in a student driven computer-based instructional format with limited instruction. It is helpful to address this hypothesis in the context of knowledge of cognition and regulation of cognition. The strategy instruction appeared to have no positive impact on students' knowledge of cognition as reported by the MAI. The strategy instruction did appear to have a positive impact on the students' regulation of cognition as reported in the MAT. The differences on the regulation of cognition items on the MAI amounted to an effect size of 0.50. Although the difference was not statistically significant, it is reasonable to believe that a more sensitive experiment would reach significance.

Perhaps the most interesting result is the disparity between the gains in regulation of cognition items and the relatively unchanged knowledge of cognition items. The student's use of strategies ultimately depends upon the decision to use a strategy. Extensive knowledge of the when, how, and why one would use a strategy is of little value if one never decides to put it to use. Regulation of cognition may be the component that should be emphasized with students in this age group.

Hypothesis 2-Achievement

The second hypothesis was that students receiving strategy instruction would exhibit superior performance on a post-test measuring content understanding. The results did not support this hypothesis. Although some students did indicate an increased awareness of the importance of regulation of cognition, (from the MAI) classroom observations indicated that they were not puffing forth a great deal of effort to use the strategies that had been discussed. This finding is commensurate with the Dehn study, which also found that in spite of increased strategy training, student achievement did not improve.

It is possible that the student did not feel that this increased awareness was relevant to this course. The student needs to attribute the use of strategies to improved performance. Some students had indicated that they found the course easy, and therefore did not use any learning strategies. Other students felt the material was difficult, but for unknown reasons would not use a different approach.

CONCLUSIONS

As mentioned in the beginning of the article, several characteristics of hypermedia may actually impede understanding and thus learning strategy instruction takes on added import. These same characteristics may also mediate the success of a strategy instruction component. While encouraging learners to use learning strategies, the environment should be conducive to their use. In other words, simply having self check questions and objectives available does not guarantee their use. An important consideration concerning the success of any intervention designed to improve learning is the student's desire to succeed or in this case learn. Even if the student has the skills to succeed in this setting, they also need the will. Here three possible reasons for the lack of effective use of learning strategies in hypermedia are addressed: cognitive load, metacognitive development, and a Garner's theory of settings.

Cognitive Load

One potential roadblock to increased understanding in this particular intervention is the increase in cognitive load associated with implementing a successful learning strategy. Cognitive load refers to the mental requirements of a task (Sweller, 1989). The material covered in this course was very technical and required a good deal of concentration to incorporate new material with the old. It is possible that this hindered the students' ability to implement strategies that would impinge on vital working memory space.

Metacognitive Development

The finding of differences in regulation of cognition, but not knowledge of cognition was unexpected. This difference lends further credence to the suggestion that the distinction between the two contributors to metacognition (knowledge and regulation) is tenable (Schraw & Dennison, 1994; Brown, 1987; Jacobs & Paris 1987). The question then becomes: Why would one component increase and another remain the same? Knowledge of cognition includes knowledge of potential strategies, referred to as "declarative knowledge," and when to use the strategies, "conditional knowledge (Schraw, 1998)."

One possible explanation is that the developmental state of the participants was such that they were more likely to make gains in regulatory behaviors than an increase in knowledge of cognition. Kuhn (1999) has said that a critical change in development is the increasing control over the process of integrating existing understandings with what she calls "metaknowing" skills. These skills include metastrategic and metacognitive. An explanation based on inadequate development would not maintain that the computer-based instruction failed to teach strategies, but that the participants possessed a sufficient amount of declarative and conditional knowledge before the intervention began. The problem the students face is integrating their understanding of these skills with their existing "meta" knowledge. Thus, gains on knowledge of cognition items (students understanding of strategies) would be minimal. Pretest scores on the MAT lend support to this explanation. The average per item score on the knowledge of cognit ion components in the pretest MAI is 5.06 versus 4.34 on the regulation of cognition items. Also, some students indicated that they had previously been taught many of these strategies.

Theory of Settings

Presenting students with the skills and the means to engage in meaningful strategic processing does not ensue that they will use the prescribed approach. This study supports such a notion. Although students were given a wide range of opportunities to use strategies, it appeared that their meaningful use was limited. Determining the reasons why students did not use strategies in this particular study could be helpful for future interventions.

Garner (1990) proposed a theory of settings to help address the problem of why strategies are not used appropriately. Garner identified five reasons why strategies are not used. These include (a) poor monitoring, (b) primitive (ineffective) routines, (c) limited knowledge base, (d) attributions that are not conducive to strategy use, and (e) the minimal transfer of strategic activity. Monitoring refers to the metacognitive activity that constantly evaluates ones understanding of the material being read. Primitive routines include the habits that students have used in the past that do not facilitate understanding. The limited knowledge base refers to the understanding the student has about the activity or tasks related to the activity.

Implications for Future Research

Future studies should include a tighter integration of the strategy instruction with the content instruction. Although Deny and Murphy (1986) suggested detached strategy instruction, the instruction may have been too detached. The ideal integrated approach includes helpful tools that are easy to access, easy to use, and facilitate deeper processing while causing a minimal amount of deviation from the content (Norman, 1993; Park, 1995). This should alleviate some of the problems that appeared during this study, such as the sequence of instruction, and the perception that it is simply more work. Tighter integration with the curriculum could extend to examples that are taken directly from the current content. Analysis of the student's use of strategies could also be improved by monitoring selections made while reading the content lessons. For example, the student who regularly reviews the lesson objectives, or spends more time reviewing practice quizzes, is demonstrating that they are a "good strategy user."

It is widely accepted that strategy use will improve understanding. However, it is becoming apparent that teaching learning strategies and increasing student's use of them is more complex than originally believed. Several intervening components such as context and developmental stage of the learner need to be investigated. It is very probable that divergent approaches for students with divergent abilities will be more effective than a standard algorithm for strategy instruction. Hypermedia instruction affords opportunities to tailor instruction to the needs of individual learners. Hypermedia instruction also has the capacity to encourage strategic learning in explicit and subtle ways. The computer's ability to deliver customized instruction and encourage strategic behaviors is a powerful reason to be encouraged by the potential role of hypermedia instruction in education.

References

Alexander, P.A, Graham, S., & Harris, K.R. (1998). A perspective on strategy research: Progress and prospects. Educational Psychology Review, 10, 129-154.

Baker, L., & Brown, A.L. (1984). Cognitive monitoring in reading. In J. Flood (Ed.), Understanding reading comprhension (pp. 21-44). Newark, DE: International Reading Association.

Brown, A. (1987). Metacognition, executive control, self-regulation, and other mysterious mechanisms. In F.E. Weinert & R.H. Kluwe (Eds.), Metacognition, motivation, and understanding. Hillside, NJ: Erlbaum.

Cardinale, L.A., & Smith, C.M. (1994). The effects of computer-assisted learning-strategy training on the achievement of learning objectives. Journal of Educational Computing Research, 10, 153-60.

Dehn, M.J. (1997). The effects of informed strategy training and computer mediated text on comprehension monitoring and reading comprehension. (ERIC Document Reproduction Service No. ED 402 545)

Derry, S.J., & Murphy (1986). Designing systems that train learning ability: From theory to practice. Review of Education Research, 56, 1-39.

Garner, R. (1990). When children and adults do not use learning strategies: Toward a theory of settings. Review of Educational Research, 60, 517-529.

Graham, S., & Harris, K.R. (1993). Self-regulated strategy development: Helping students with learning problems develop as writers. The Elementary School Journal, 94, 169-181.

Hartman, H.J. (1998). Metacognition in teaching and learning: An introduction. Instructional Science, 26, 1-3.

Hattie, J., Biggs, J., & Purdie, N. (1996). Effects of learning skills interventions on student learning: A meta-analysis. Review of Educational Research, 66, 99-136.

Jacobs, J.E., & Paris, S.G. (1987). Children's metacognition about reading: Issues in definition, measurement, and instruction. Educational Psychologist, 22, 255-278.

Keller, J., & Burkman, E. (1993). Motivation principles. In M. Fleming & W.H. Levie (Eds.) Instructional message design: Principles from the behavioral and cognitive Sciences. (pp. 3-49). Englewood Cliffs, NJ: Educational Technology.

Kuhn, D. (1999). A developmental model of critical thinking. Educational Researcher, 28(2), 16-26.

Norman, D.A., (1993) Things that make us smart: Defending human attributes in the age of the machine. Reading MA: Addison Wesley.

Park, S. (1995). Implications of learning strategy research for designing computer-assisted instruction. Journal of Research on Computing in Education 27, 435-56.

Puntambekar, S. (1995). Helping students learn "how to learn" from texts: Towards and ITS for developing metacognition. Instructional Science, 23, 163-182.

Puntambekar, S., & duBoulay, B. (1997). Design and development of MIST: A system to help students develop metacognition. Journal of Educational Computing Research, 16, 1-35.

Pressley, M., & Woloshyn, V. (1995). Cognitive strategy instruction that really improves children's academic performance. Cambridge, MA: Brookline Books.

Rosenshine, B., Meister, C., & Chapman, S. (1996). Teaching students to generate questions: A review of the intervention studies. Review of Educational Research, 66. 181-221.

Schraw, G., & Dennison, R.S. (1994). Assessing metacognitive awareness. Contemporary Educational Psychology, 19, 143-154.

Schraw, G. (1998). Promoting general metacognitive awareness. Instructional Science, 26, 113-125.

Symons, S., Snyder, B.L., Cariglia-Bull, T., & Pressley, M. (1989). Why be optimistic about cognitive strategy instruction? In C.B. McCormick, G.E. Miller, & M. Pressley (Eds.), Cognitive strategy research: From basic research to educational applications (pp. 104-132). New York: Springer-Verlag.

Sweller, J. (1989). Cognitive technology: Some procedures for facilitating learning and problem solving in mathematics and science. Journal of Educational Psychology, 8, 457-466.

Weinstein, C.E. (1978). Teaching cognitive elaboration learning strategies. In H.F. O'Neal (Ed.), Learning strategies (pp. 31-55). New York: Academic Press.

[Graph omitted]

[Graph omitted]

[Graph omitted]

[Graph omitted]

Table 1
Learning Strategy Lesson
Titles
Week  Topic          Day 1           Day 2           Day 3
1     Summarization  Identifying     Summary         Organizing
                     Important       Statements      Information
                     Information
2     Goals          Why Set         Short Term      Long Term
                     Goals?          Goals           Goals
3     Objectives     Why pay         Using           Making it a
                     attention to    Objectives      habit
                     objectives?
4     Monitoring     How to monitor  What do you     Monitoring
      Learning       your learning   Want to learn?  Strategy Use
5     Concept        Description     Example         Practice
      Mapping
6     Questions      Using           Self            Comp.
                     Questions       Generated       Questions
                                     Questions
Week  Day 4
1     Generating
      Topic
      Sentences
2     Strategy
      Goals
3     From the
      Beginning
4     Monitoring
      Your
      Progress
5     Practice
6     Strategy
      Questions
Note. Lesson topics
adapted from Pressley,
M. & Woloshyn, V. (1995).
Cognitive Strategy
Instruction That Really
Improves Children's
Academic Performance.
Cambridge, MA: Brookline
Books.
Table 2
Means and Standard Deviations for
Metacognitive Awareness Inventory
Group                       Pretest  Posttest
Strategy Instruction (n=7)
M                           238.57   242.14
SD                           58.27    59.87
Control (n=6)
M                           238.67   227.17
SD                           14.11    24.62
Table 3
Means and Standard Deviations for Knowledge
of Cognition Items
Group                       Prestest  Posttest
Strategy Instruction (n=7)
M                             90.4      88.0
SD                            15.7      22.0
Control (n=6)
M                             92.0      90.0
SD                             5.9       9.9
Table 4
Means and Standard Deviations for
Regulation of Cognition Items
Group                       Pretest  Posttest
Strategy Instruction (n=7)
M                           148.14    154.14
SD                           43.61     39.05
Control (n=6)
M                           146.67    137.17
SD                           14.29     17.14
Table 5
Means and Standard Deviations
for Content Exams
Group                 Pretest  Posttest
Strategy Instruction
M                      60.86    63.37
SD                     15.05    17.44
Control
M                      75.67    82.22
SD                     11.96     6.10