21 October 2016
I’m teaching four sections of first-year English composition this semester, so this subject is near and dear to my heart. Two Stanford researchers, Andrea A. Lunsford and Karen J. Lunsford, have conducted a longitudinal study of college freshman writing, comparing the results from students in 2006 with earlier studies from 1917, 1930, and 1986, and the results are quite surprising.
First, twenty-first century students are making errors at the same rate as their earlier counterparts. The error rate in all four samples varied between 2.11 and 2.45 errors per 100 words.
Their writing has not, as many would suppose, gotten worse. The types of errors, however, have shifted. In the earlier samples, spelling errors were the most common type, but they had dropped to fifth in the 2006 group, almost certainly the result of automated spell checking. (Interestingly, spelling errors were the second most likely type to marked as wrong by the graders, indicating that teachers are especially prone to noticing them.) Instead, the most common error in 2006 was a “wrong word” error, again, probably the result of automated spell checking.
But even more surprising is the amount of writing students do today. In 1917 the average essay was a mere 162 words in length. It grew to 231 words in 1930 and 422 in 1986. But the 2006 students wrote essays that averaged 1,038 words. The article on the study published by JSTOR Daily (linked above) chalks up the longer length to word-processing technology, which was in its infancy in 1986 but almost universally used by college students by 2006.
The types of essays have changed as well. In the earlier years, the majority of the essays were personal narratives. But of the 2006 sample, only 9% fit that category. Instead, over 70% of the essays were research papers (the most common type assigned), argumentative papers without sources, and close readings and analyses. There is far more emphasis on research and intellectual rigor than in the past. The shift to research papers has also changed the types of errors being made. Source and attribution errors accounted for three of the top twenty error types in the 2006 study, but are not included in the earlier studies, which relied largely on personal narratives.
Texting and instant-messaging terms, like imho and lol, were not to be found in the essays, reconfirming what linguists have known for years, that texting is not corrupting the language skills of the young. Also, very few of the papers incorporated images, hyperlinks, multi-colored fonts, or other graphical innovations that are easily done with a computer. The students overwhelmingly stuck to plain text. (Also, the graders primarily used pen and pencil, not electronic grading and mark-up programs. And those that did use electronic marking, primarily relied on the mark-up tools in MS Word, rather than specialized programs.)
So the conclusion is that students today are, in fact, not worse writers than in the past. And, they are writing far more and in more rigorous genres than their earlier counterparts.
[Correction: I had originally said the researchers attributed the longer length of the papers to word processing technology. The study does not mention this. It is the JSTOR Daily article on the study (linked above) that makes the connection.]
