Processing Information By Writing

Our personal experiences as writiers probably encourage the conclusion that we must learn something when we write. We typically plan, ponder, and edit when we write. Sometimes the product we generate even surprises us - perhaps we are surprised we knew the things that appear in our final product. Writing is difficult work so we must be learning something.

Knowledge Telling Is Not Always Knowledge Transforming

Some conclude the optimistic assumptions regarding the benefits of writing must be tempered. In considering how a writer generates a product, Bereiter and Scardamalia (1987) propose that writers may be involved in knowledge telling and/or knowledge transforming. The first approach results in a dump of ideas from long term memory triggered by the assignment and information that has already been recalled. In other words, a product can be generated without changing what is stored in memory or how this content is organized. Most instructors probably have a feeling for how this works. They have asked students to reply to an essay question and received at least some responses that seem to be everything students knew about the topic rather than a specific answer to the question. It can be hard to know if the student thinks he has answered the question or if this is just an often productive ploy. It is the first situation that shares characteristics with the idea of knowledge telling.

In contrast, writing experiences classified as knowledge transforming require that previously stored information be reorganized, reinterpreted, etc. in order to complete the writing task. Bereiter and Scardamalia contend that it is not always possible to tell what has happened by examining the product. The arguments in what appear to be a deeply thoughtful essay could have already existed in the writer's memory as a consequence of reading or the storage of comments offered by a teacher. Some demonstration of change in knowledge or some method for evaluating the processes a writing task generates would offer assurance that more than recall was involved. Note that writing can contribute to learning in different ways and knowledge telling only argues that writing may not immediately influence understanding as a consequence of composing.

Some Research on Writing to Learn

In reviewing research on "writing to learn", authors have reached a conclusion similar to the conclusion reached in summarizing the research on many other applied educational practices. This nearly universal conclusion might be paraphrased as "sometimes it works and sometimes it does not".

There are many reasons educational reviews come to what might appear to be an inconclusive summary. Sometimes, some of the studies suffer from serious methodological weaknesses and the optimistic conclusions of the most positive studies end up being misleading. Often, an educational practice ends up being interpreted differently by different researchers and the classroom experiences of students supposedly experiencing a practice with the same name can be quite different. Sometimes the content to be learned, the age or sophistication level of students, the skill level of educators in implementing a practice, or what is used as a measure of learning can made a big difference. What good reviews, often done as "meta-analyses", attempt to do is sort out factors that may be associated with positive, neutral and negative outcomes. The question, says Klein (1999), should be "When writing contributes to learning, how does it do so"?

Our comments here will focus on two recent reviews of writing to learn. While these reviews do acknowledge the typical mixed results, they also offer insight regarding important moderating variables and offer some suggestions for practice.

Klein (1999) offers a review organized around the writing processes hypothesized to be responsible for learning.

1. Point of utterance (no revision and limited planning). This explanation assumes that learning occurs in the attempt to generate comments on the topic (no revision or planning is expected). This category might be described as spontaneous writing - the learner is asked or personally commits to writing on a topic. An activity fitting within this category would be the five minute writing tasks some college lecturers assign at the end of a class. Note that this type of writing seems very similar to what Bereiter and Scardamalia (1987) described as knowledge telling.


2. Forward search - This explanation is centered on the benefit of having an external product available. This external product (what has been written) allows the writer to overcome the limits of working memory and offers an opportunity to review, reorganize, and develop new ideas.


3. Genre-related - This explanation focuses on the benefits of transforming ideas to the structure demanded of a specific genre. For example, in a "position paper" a writer is expected to take on of several possible positions are generate an argument supporting this selection. In selecting, organizing, and connecting knowledge to fit the demands of a formal writing task, the learner creates an understanding that would not exist without the imposed task


4. Backward search - This explanation assumes that skilled writers formulate complex goals (characteristics of desired product, needs of audience, etc.) and then rework existing knowledge in terms of these goals. This seems a more generalized version of the "genre-related" explanation requiring more sophisticated and complex problem solving.

While these explanations may center on different types of writing tasks, note that several of the explanations could be involved in more sophisticated writing tasks. For example, in a position paper, a writer generates text, has the opportunity to review and evaluate an externalized product, and must satisfy genre-specific requirements.

Klein noted that tasks that might be associated with the different explanations differed on several dimensions. As listed, the tasks range from spontaneous to deliberate, from activities requiring limited writing skill to tasks requiring sophisticated writing skill, and from tasks involving little self monitoring to tasks assuming sophisticated self monitoring.

Based on these observations and the mixed findings from researchers, Klein (1999) suggested that practitioners look to forward search and genre-related activities. Activities associated with these explanations for how students may benefit from writing are within the range of capability of most students and require some level of critical thinking.

Klein (1999) offers one other observation on student behavior. When students have experienced well structured content (read a book, listened to a lecture), there is a tendency for students to use ideas as presented rather than to restructure ideas when responding to writing task. Some situations, such as a science experiment or the consideration of historical documents, presents students "raw" information increasing the likelihood that writing will encourage higher order thinking. Wiley and Voss (1999) note that comparisons of of students asked to write and those in control conditions were more likely to differ when students were required to learn from multiple sources or when writers were asked to argue for a position. Both situations require reorganization of information.

Bangert-Drowns, et al. (2004) - A Recent Meta-Analysis of Writing-To-Learn

Possibly the most comprehensive recent analysis of K-12 classroom-based writing to learn was completed by Bangert-Drowns and colleagues (Bangert-Drowns, Hurley, & Wilkinson, 2004). Less based in theories of the potential mechanisms by which writing may influence learning than the analysis completed by Klein (1999), the summary of findings from many research studies allows consideration of some different task and learner characteristics.

First, though, it is worth considering a caveat identified by Bangert-Drowns and colleagues. These authors note that the dependent variables in most research tended to rely on "conventional measures of achievement" and may not have been sensitive to the development of sophisticated reasoning or a more elaborate conceptual framework. Recently, a similar concern seems to surface in many educational contexts. The idea is that evaluation practices that solely assess student ability to retrieve basic factual information may miss other important accomplishments and may fail to appropriately credit educational experiences contributing to other changes in student capability.

Findings from the Bangert-Drowns and colleagues meta-analysis:

1. There appear to be grade-level differences in the benefit of writing-to-learn. Middle school students appear to benefit LESS from such tasks. The authors offer no explanation for this finding and note that it may be an artifact.


2. Interventions of longer durations (more weeks) were associated with larger effect sizes. The authors speculate that the benefits of writing-to-learn may accumulate over time possibly as writing experiences feedback to influence learning strategies.


3. The average time on specific writing tasks was negatively related to achievement - shorter duration tasks were better. It is important in thinking about this finding to note that these writing tasks were mostly completed during class time (an hour or less). Involvement in writing competed for other uses of class time, e.g., the presentation of additional information. The authors also speculated that long writing tasks may reduce the motivation of some students.


4. Writing assignments that included a metacognitive task were most beneficial. In such assignments, students are asked to evaluate current understanding and comment on their confusions. This purposeful approach to activating metacognition is a little different than an approach that the task of writing per se generates metacognitive experiences.


5. Whether or not students received external feedback on their writing, did not have a large impact on learning.

Hypercomposition

For those of us interested in student multimedia authoring, it is frustrating that the research base associated with this topic is so sparse.

One of the few positive outcomes associated with student multimedia authoring we are aware of was provided by Lehrer (1993). Lehrer was evaluating the consequences of hyperauthoring in a design project conducted with 8th graders. The students were creating multimedia products on the topic of the Civil War. One year after completing the project an interviewer not associated with the original asked hyperauthors and control participants about 5 topics associated with the Civil War (e.g., Civil War causes, role of slaves, connection to present). Responses to the queries were scored for evidence of elaboration (i.e. reporting more than surface details). One year later those students who participated in the design projects demonstrated much greater frequency of elaborations.

Lehrer summarizes:

"... students' long-term recall of the Civil War suggested that the design approach lead to knowledge that was richer, better connected, and more applicable to subsequent learning and events." (p. 221)

There are potentially many issues associated with such findings. For example, was the time expended worth any increase in knowledge or understanding? Perhaps the control group learned more about other things. It is curious that examining the benefits of experiecnes over extended periods of time is such a rare research methodology. Experiences resulting in deeper understanding, better integration of ideas, etc. would seem to be more visible as surface level understanding fades.

References

Bangert-Drowns, R. L., Hurley, M.M., & Wilkinson, B. (2004). The effects of school-based writing-to-learn interventions on academic achievement: A meta-analysis. Review of Educational Research, 74, 29-58.

Bereiter, C. & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Erlbaum.

Klein, P. D. (1999). Reopening inquiry into cognitive processes in writing-to-learn. Educational Psychology Review, 11, 203-270.

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: Erlbaum.

Wiley, J. & Voss, J.F. (1999). Constructing arguments from multiple sources: Tasks that promote understanding and not just memory for text. Journal of Educational Psychology, 91, 301-311.