New York City--In a surprising finding, a researcher reported here last week that the vast majority of American students tested in a new evaluation of writing skill did well.
The preliminary data, from the first international comparison of precollegiate students’ prose-writing ability, differed sharply from results of recent U.S. assessments. The study also correlated success in prose writing with certain emphases in teaching--finding, for example, that an instructional focus on reading and on essay planning enhanced achievement.
“This was a big shock,” said Eva L. Baker, director of the center for the study of evaluation at the University of California at Los Angeles. “U.S. people don’t always get a chance to say this.”
“We are too used to hearing about all the flaws our students demonstrate, from not knowing simple historical facts to lacking competitive strength in mathematics,” she said.
The study found that 86 percent of American 6th, 10th, and college-bound 12th-grade students wrote papers judged “competent” or better. While older students performed better than younger students, 6th graders performed well on many tasks.
The results differ markedly from those of a 1984 study by the National Assessment of Educational Progress, which concluded that most students “are unable to write adequately, except in response to the simplest of tasks.” (See Education Week, Dec. 10, 1986.)
Ms. Baker suggested that students in her study performed better than those in the naep assessment because they were given one hour to write their responses, rather than the 15 minutes provided students in the naep study.
“Testing student proficiency with such tiny time periods for writing is like educational fast food--it provides only short-term satisfaction,” she said.
By contrast, she added, “giving students a reasonable amount of time is critical: time to think about what to say, and then, time to write.”
But Ina V.S. Mullis, deputy director of naep, said it was unlikely that the additional 45 minutes would improve student performance so markedly. She noted that a recent preliminary study, in which students were given an hour to write their papers, found no significant improvement in student scores.
The difference in scores should probably not be attributed to test conditions, according to Ms. Mullis, but to variations in the populations of test takers.
She noted, for example, that the college-bound 12th graders in Ms. Baker’s study would be likely to achieve at higher levels than the nationally representative sample of 17-year-olds that naep uses. “I would look first into who the kids were, and where they came from,” she said.
Teaching Practices
Ms. Baker’s study was the first test of writing composition conducted by the International Association for the Evaluation of Educational Achievement. Some 13 nations participated in the study, but the researchers have not yet been able to analyze the results for comparative purposes.
In the United States, the researchers randomly selected and tested a total of 1,500 6th graders, 1,500 10th graders, and 1,100 college-bound 12th graders in about 400 public and private schools nationwide.
The students were asked to perform 3 of 10 possible writing tasks, ranging from informal functional tasks, such as writing a note to explain a missed meeting, to longer narratives, such as describing meeting a new friend. The tasks were divided into three genres: descriptive, persuasive, and reflective. Students in all countries were given the same tasks.
Papers were judged according to the students’ use of language to achieve the goal of the paper. The papers were graded on a 5-point scale; those rated 3 were considered “competent,” without major problems.
The study found:
In narrative writing, 69 percent of 6th graders, and 89 percent of 10th graders, scored 3 or above. In addition, it found, 23 percent of 6th graders, and 46 percent of 10th graders, scored 4 or above.
In persuasive writing, 24 percent of 6th graders, 72 percent of 10th graders, and 86 percent of 12th graders scored 3 or above, while 3 percent of 6th graders, 26 percent of 10th graders, and 48 percent of 12th graders scored 4 or above.
In reflective writing, which the researchers considered the most difficult task, nearly 79 percent of 10th graders and 89 percent of 12th graders performed competently.
The researchers also found that certain teaching practices correlated with student performance. For example, Ms. Baker noted, for students at all grade levels, teachers’ emphasis on reading correlated positively with writing performance.
“Reading obviously provides both content--what to say--and schema--how to say it,” she said.
In addition, she noted, teachers who stressed essay planning and group work tended to produce students who performed well on the assessment.
On the other hand, she said, an emphasis on spelling and grammar appeared to lower students’ scores. In 8 of 10 tasks, she said, emphasis on parts of speech was significantly and negatively correlated with writing performance.
No Easy Tasks
Ms. Baker rejected the possibility that the results of her study were skewed upward by lenient judges or easy tasks. The researchers established an objective scoring scale, and trained judges in applying the scale, she said. The researchers also checked periodically to ensure that the judges’ scores matched the established scale, she added.
The fact that younger students never performed better than older students, and that private-school students performed at higher levels than public-school students, except the college-bound seniors, indicates that the tasks were valid, Ms. Baker said.
In addition, the 6th graders did relatively well on narrative writing, which is a frequent assignment in elementary school, and less well on persuasive writing, which is seldom assigned at that level.
