Case for Discussion: Improving Student Understanding

Handout from faculty seminar on Using Writing to Develop Student Understanding

Take a few minutes to read the case, “Improving Students’ Understanding” (reverse side of the page). Jot down answers to the questions. After you have a chance to think about the questions on your own, discuss your ideas with people at your table (or around you). I will interrupt the discussion shortly and ask volunteers to offer their explanations:

1. Make predictions about the performance of the three groups of students. How well—relative to one another—do you think the groups performed on the test of understanding?
2. Explain the reasons for your predictions.

Improving Students’ Understanding

Dan Schwartz and John Bransford at Vanderbilt University conducted a series of studies to examine how different learning experiences facilitate college students’ understanding of subject matter. Their experiments were integrated into classes so that students learned actual course material as they participated in the experiments. One of the experiments took place in a sophomore level psychology class during a unit on human memory theories and research. Students were assigned to one of three conditions in which they studied specially prepared course material in class and a week later took a test to determine their understanding of the subject matter. The three groups did the following:

Group 1: Read + Summarize + Lecture.

• In class, students read a specially prepared chapter on memory and wrote a summary of the material.
• Next class period they heard a lecture that integrated and explained the material from the text.

  Note: The experience of Group 1 is a common instructional sequence—students prepare for class by reading and writing about the material, and then listen to a lecture about the subject matter in class.

Group 2: Analyze + Lecture.

• Students did NOT read the specially prepared chapter on memory.
• Students analyzed specially prepared “data sets” of memory studies. Students were asked to find and graph what they thought were revealing patterns for each study.
• Next class period students heard the same lecture as Group 1, which integrated and explained the material about memory theories.

  Additional information about the data sets. Each data set briefly described an actual memory study and reported how research subjects performed in it. The data sets formed “contrasting cases” in that they depicted results from different studies. The sets contained no explanatory information about the studies or memory theories.

Group 3: Analyze + Analyze.

• Students did NOT read the specially prepared chapter on memory.
• In class, students were given twice the amount of time to analyze the same data sets as Group 2.
• They did NOT listen to the lecture.

Testing students’ understanding. The groups were tested one week after they had studied the material. In the test they read about a new memory experiment (one they had never seen), and then had to predict the likely results of the experiment based on their knowledge of memory theories. This kind of prediction task is a rigorous test of understanding. The researchers reasoned that if students had gained a deep, expert-like understanding of the target concepts, then they should make relevant predictions given a new problem.

How Did the Groups Do?

As the graph indicates Group 2 (Analyze + Lecture) did significantly better than Group 1 (Read + Summarize + Lecture) and Group 3 (Analyze + Analyze). Group 2 had a much better grasp of the concepts, and Groups 1 and 3 were equally poor at making predictions.

Note. In previous versions of this study, researchers also used a true/false test to determine whether students recalled the basic facts of the memory theories and studies. On this test, students who studied by Read + Summary + Lecture remembered just as much as those who had studied by Analyze + Lecture. In addition, the experimenters analyzed students’ summaries in Group 1 and found that key concepts from the studies appeared in their summaries. In other words, both groups had the facts, but the Analyze + Lecture group understood them well enough to make predictions.

Talking Points: Improving Students’ Understanding

The sequence of contrasting case analysis and lecture was a potent combination that enabled students to make three times more predictions than the other learning methods. The sequence of reading, summarizing and lecture did not foster understanding of the material. In fact, it was no better than the double analysis method in which the students neither read the material nor heard the lecture.

So, what’s going on here? What accounts for the superior performance of Analysis + Lecture? Why was Analysis + Lecture more effective than both the Read + Summarize + Lecture and the Double Analysis methods?

Analysis of contrasting cases versus reading and summarizing. Students’ summaries contained the same factual information as that contained in the chapter, indicating that the students had picked up on key ideas. The act of summarizing is a way to make sense of new material by deciding about important vs. unimportant ideas, condensing information and translating ideas into ones own words. But summarizing did not prepare students to take advantage of the explanatory lecture. These students could describe the studies and their results but not why they turned out that way.

Analysis of contrasting cases entails a different mental process than summarizing. As students analyzed the data sets, students found distinctive features and patterns in the data even though they did not know the significance or “meaning” of the patterns. By becoming aware of similarities and differences, they developed what the researchers refer to as “differentiated knowledge” of the data sets. These students were well attuned to differences in the data even though they did not know what the patterns represented. On the other hand, students who summarized the material most likely developed general connections among the concepts and theories.

The lecture. The lecture provided an explanatory framework for the material—it explained why the results turned out as they did. Students who had analyzed the data were “ready” to make sense of the explanation. They had discerned distinctive features of the material and then with the aid of the lecture were able to render the patterns meaningful. But students who read and summarized the material and then heard the lecture did not develop a grasp of the concepts. This group could recall the material, but they could not use the instructor’s explanation to advance their understanding of the concepts. Most likely the lecture served to introduce new ideas and to reinforce connections among the facts. Deep understanding required both differentiated knowledge (as developed by discerning the contrast among cases) and explanatory knowledge (as developed through the lecture).

Instructors typically assume that a good explanation in itself is sufficient to produce student understanding; that in some sense we can transfuse our understanding into the minds of our students. This study illustrates the constructivist nature of learning. The lecture had little effect on students’ understanding even when they had studied the material in advance. But students who had analyzed the data were able to use their differentiated knowledge to construct meaningful representations of the lecture material.

©2001, Bill Cerbin and Terry Beck