ATLANTA -- An analysis of research papers written in first-year composition courses at 15 colleges reveals that many students simply copy chunks of text from the sources they cite without truly grasping the underlying argument, quality or context.

“The findings are not happy news for how writing is taught,” Rebecca Moore Howard, an associate professor of writing and rhetoric at Syracuse University, said here Thursday at the annual meeting of the Conference on College Composition and Communication. “[Students] are not selecting authoritative, meaningful sources and not reading them carefully. They are not, in a word, engaging.”

Howard's presentation -- with her co-principal researcher, Sandra Jamieson, professor of English, director of composition and department chair at Drew University -- of the initial findings of the Citation Project could carry broad implications for how writing is taught at the college level, not just in composition courses, but across disciplines.

“We were often stunned by what we found,” Jamieson told the standing-room crowd Thursday, adding that the biggest shock often came when researchers reviewed papers from their own institutions. “What we can say with confidence is that the trends cross school types and student types.”

Howard and Jamieson led a team of 20 researchers in analyzing 164 student research papers produced in first-year composition classes. They studied a range of institutions in 12 states from diverse regions of the country, culling work from community colleges and four-year public universities, private colleges and universities, and religiously affiliated and Ivy league institutions.

The researchers analyzed the students' 1,832 research citations and assigned each of them to one of four categories:

• Exact copying -- a verbatim cut-and-paste, either with or without quotation marks.
• "Patchwriting" -- the copying of the original language with minimal alteration and with synonyms substituting for several original words (patchwriting is often a failed attempt to paraphrase, they said).
• Paraphrasing -- a restatement of a source's argument with mostly fresh language, and with some of the original language intact; it reflects comprehension of a small portion, perhaps a sentence, of the source material.
• Summary -- the desired form of citation because it demonstrates true understanding of a large portion, if not the entirety, of the original text; summarizing was identified by the researchers when student writers restated in their own terms the source material and compressed by at least 50 percent the main points of at least three consecutive sentences.

Only 9 percent of the citations were categorized as summary. “That's the stunning part, I think: 91 percent are citations to material that isn't composing,” said Jamieson. “They don't digest the ideas in the material cited and put it in their own words.”

The researchers acknowledged that the act of classifying the citations is inherently subjective, and said that, if anything, they tended to err on the side of the most generous interpretation when assigning the citation to a category. They are also still refining their classifications and coding the data from one more institution. Still, Howard said, such adjustments are unlikely to alter their basic findings.

While many inside and outside academe might see the Citation Project's initial results as further proof that most students cut and paste from the Internet (indeed, the researchers found that 44 percent of citations fit that description) or that students plagiarize rampantly, Howard cautioned against falling back on such simple conclusions.

“We don't think we have lazy, fraudulent students,” said Howard, who has written extensively about plagiarism, arguing that teachers should not just adopt a tough posture, but should instead be far more nuanced in instructing their students in how to avoid it. “We think we have students working for efficiency and doing efficient writing.”

The researchers also aimed to do more than identify proper citation (or come down hard on apparent plagiarism). They wanted to tease out a subtler dynamic: how the students use, relate to and understand their sources. The research team looked at how the students demonstrated their comprehension of their sources' underlying arguments, as well as what kinds of materials they used and from what parts of the text they quoted.

Most sources were cited only once, they found. In addition, more than half of the 1,832 citations were to source material that was five pages or shorter, and more than three-quarters of the citations referred to information that appeared on the first three pages of the original material, which suggests that students tend to rely on short sources and may fail to read to the end. That so few citations were classified as summaries -- 164 out of the 1,832 -- also indicates that many students are alighting on several different sources without spending much time reading them, then cobbling them together into what Howard called “an incomprehensible pastiche.”

There were exceptions to the rule of one citation per source, though these weren't exemplars either. In an example that drew gasps from some in the audience, about 90 percent of one page in a student's research paper was either a direct quote or patchwriting -- and 9 of its 17 citations referred to the same page of the same source: a single entry on WebMD. While the paper was an extreme and atypical example, it also demonstrated a common trend: students tended to rely heavily on their sources -- so heavily, in fact, that students rarely seem to fully own the material and marshal it to form a novel argument, the researchers said.

“The compelling, unnerving issue is that the student has nothing to say,” said Howard of the piece that drew so heavily on WebMD. “How could she, since she's writing a research document from reference materials?”

Several audience members, and the researchers themselves, grappled with the pedagogical implications of the findings. Jamieson wondered whether the long-held standards for research papers, which were articulated in 1959 in the Writer's Guide and Index to English by Porter Perrin and Karl W. Dykema -- that students should read and digest material so they can talk about the subject before writing about it (they called it “part of the morality of writing”) -- still apply in today's world.

The idea that summarizing should be the goal of most citations may have held in a different era, said Jamieson, “before the Internet culture of Tweeting and sampling, in which all the rules have changed -- aside from ours.”

“Whatever else the Internet has done," Jamieson continued, "it has made it easier to find sources and harder to tell what's junk.”

Some in the audience said the findings point to the need to place greater emphasis on teaching students how to select proper sources. “It's probably not far off to say that their sources are the first hits on Google,” one audience member observed.

Another commenter was not prepared to give up on the 20th-century expectations of student research and citation. “There's some value to reminding students about the authority on certain subjects that are not in a digital archive,” she said. “What we've forgotten is that libraries were the repositories where people made judicious claims about what sources are worth reading.”