For Understanding, Some Writing Tasks Are Better than Others

According to a study by Jennifer Wiley and James Voss of Washington State University and the University of Pittsburgh, respectively, the title above seems to be unequivocal.

These researchers conducted a series of experiments where students of history were given four different kinds of writing tasks based on the same reading material.

One quarter of the students were given the following prompt: “Your task is to take the role of historian and develop a narrative about what produced the significant changes in Ireland’s population between 1846 and 1850.” (Wiley and Voss, 1999: 303.)

For the other groups, the word “narrative” was replaced by “summary,” “explanation,” or “argument.” One further variation was that half of the students in each condition were given multiple sources to read as a basis for their writing, and the other half received a single textbook chapter as the source.

The research was based on the idea that tasks requiring a student to construct a mental model of a situation are more likely to promote deeper understanding than tasks that can be performed with a more superficial representation of the information.

As such, the researchers were interested in two questions: (1) Which tasks are more likely to promote the construction of mental models? (2) Do these tasks promote deeper understanding?

In order to answer the first research question, student writing for each of the four task types was analyzed to determine:

a) The percentage of borrowed sentences, either taken directly or paraphrased from the source material.

b) The percentage of transformed sentences, combining two or more pieces of information from the source material—or combined information from the source material with a new claim or fact that was not contained in the source material.

c) The number of connections, including inferences, causal connections, temporal connections, correlations etc.

d) The number of causal connections.

In order to answer the second research question, students were given a variety of tests that assessed recall of surface features, as well as tasks that required a deeper understanding through the ability to (a) make inferences and (b) recognize underlying principles.

What were the results of the study?

Research Question 1 - Which tasks are more likely to promote the construction of mental models?

• Students who wrote from the textbook chapter used a greater proportion of borrowed sentences; used a lesser proportion of transformed sentences; and used less connectives as well as less causal connectives.

• Students who wrote arguments used a significantly greater proportion of transformed sentences; used a significantly lesser proportion of borrowed sentences; and used more connectives as well as more causal connectives.

• Summaries did not differ greatly from narratives in terms of the analyses conducted.

• Explanations prompted more causal connections than narratives or summaries, but did not approach the level of mental model construction demonstrated by the arguments.

• The combination of argument writing and using multiple sources was particularly powerful in promoting the construction of mental models, as measured by the analyses conducted.

Research Question 2 - Do these tasks promote deeper understanding?

• Students who read the textbook chapter and wrote narratives or summaries recognized presented information better than all other conditions.

• Students who read multiple sources were able to recognize appropriate inferences relating to the material better than those who read the textbook chapter.

• Only students who read multiple sources and used these to construct an argument were able to consistently identify underlying principles and root causes.

What does this mean for our classrooms?

The results of this study suggest a reasonably clear distinction between writing tasks that promote recall of surface knowledge, and those that promote the construction of deeper understanding. It seems that reading a single source and using it as the basis for writing a narrative or a summary can promote recall of information, at least in the short term.

This scenario does not seem support students in constructing the kind of mental models that lead to deeper understanding, however. For this to happen, students need to be given opportunities to combine information from multiple sources into an argument.

Why might this be the case? And why did the explanation task, which in theory also requires students to consider connection and causation, not have a greater impact on student understanding?

These questions cannot be answered with certainty from this study, but the researchers have proposed some possible explanations. It may be that arguments are more personal than explanations or other writing tasks, thus requiring the writer to consider their own opinion and construct a representation of it for others.

A second possible explanation, and one that probably works in conjunction with the more personal nature of writing an argument, is that opinions in an argument need to be supported and justified with evidence. This need for justification would oblige the writer to carefully consider the information contained in the sources in order to bring it together to defend the position they have decided to argue. This kind of thinking would necessarily also involve the comparison and evaluation of evidence from multiple sources, if these are available.

More generally, the researchers suggest that we should consider the idea that the conditions that promote recall of information may not be the same as those which promote deeper understanding. They suggest that multiple-choice tests and short-answer quizzes that emphasize recall of facts may actually be biased against students who engage in deeper understanding, given the result that students who read textbook chapters and wrote narratives were more easily able to recognize that information again on assessments.

Obviously, both purposes have their place, but we need to be clear about what our purpose is —and the relative value we place on each.

Reference

Wiley, J., and Voss. J. (1999) “Constructing arguments from multiple sources: Tasks that promote understanding and not just memory for text.” Journal of Educational Psychology 91 2, pp. 301-311.

---

Please fill out the form below if you would like to post a comment on this article:

Nickname

(this will appear with your comments)

Email

Comments

---

Comments

There are currently no comments posted. Please post one via the form above.

MORE FROM

GORDON ELDRIDGE: LESSONS IN LEARNING