Article Citation:
Isakson, Carol, and Jan Spyridakis. 1999. “The Influence of Semantics and Syntax on What Readers Remember.” Technical Communication 46 (3): 366-381.
Summary of article:
In this article, Carol Isakson and Jan Spyridakis wanted to determine whether decisions made about syntax (word choice and arrangement) and semantics (meaning of words) have an effect on the ability of a reader to comprehend information presented to them. Based on past research, Isakson and Spyridakis expected to see an improved recall of topics presented in the beginning or end of a paragraph or document, in active rather than passive voice, in an independent clause rather than a dependent clause, and in a clause rather than a phrase (368).
To test this idea, they conducted a study with 39 native English-speaking engineering students who were divided into two groups. Each group read one of two possible science texts that were selected with similar difficulty, paragraph and sentence structure, readability statistics, and number of idea units (IU’s). The researchers defined idea units as “the smallest information units that readers may logically pause at to emphasize, to enhance meaning, or to take a breath.” (369-370). Students read a copy of the text and then completed a math problem before taking the comprehension test. The purpose of the math problem was to have an interruption between reading the text and taking the comprehension test. Students also recorded the time it took to read the text and to finish the comprehension test.
The researchers found that for the two texts, readers recalled a higher percentage of new information that was presented at the end of a paragraph. On the other hand, students remembered a similar level of IU’s for the text that repeated facts at the end of a paragraph. In addition, readers more accurately recalled topics that were included in an independent clause or in the first or last paragraph of the text (377). Based on these results, Isakson and Spyridakis recommend that writers put emphasized content in:
• Clauses
• Independent clauses
• Relative clauses at the end of a sentence (if using relative clauses)
• The first paragraph of a document
• The last paragraph of a document and the last sentence of a paragraph (377).
Analysis/Evaluation of article:
Dr. Jan H. Spyridakis is a professor of technical communication at University of Washington, Seattle whose research focuses on methods of technical communication, user-centered design and international technical communication. Carol S. Isakson is also listed as being affiliated with University of Washington, Seattle; however, I was unable to locate information on her background or qualifications in my online search.
The article was published in 1999, so it is somewhat outdated; however, the information is still relevant to technical communicators today. The researchers did a thorough literature review which covered studies related to how a reader’s perception of a text (how familiar they are with the subject matter), how syntax of a text (passive vs. active voice), and how semantics of a text (where information is presented) influence a reader’s comprehension and recall. The studies that they included in their literature review supported the hypotheses that they presented for their study.
As far as their study is concerned, I thought that they used methods that were appropriate, with the exception of their post-test activity. They had participants write down everything that they recalled from the text and record how much time they took to write the recalled information. Then, they scored the recall writings using two raters. Although they had a 93% inter-rater reliability, they mentioned in the results that the analysis related to recall of information based on location of sentences or location of paragraphs in the text may not have been accurate based on their scoring method. I think it would have been more useful to use a post-test with questions instead of having participants just write down everything they recalled.
In addition, they used two different texts with different topics, which despite similar text criteria and difficulty level caused very different results. I think it would have been beneficial to use texts with the same topic, or use one text and rewrite it for the different treatments.
Lastly, the sample size that was used was somewhat small. The researchers had 39 participants (20 read one text and 19 read the second text) who were students. It would be interesting to see what the results would be if the sample size was larger.
Reflection:
Despite some of the issues with the study, the results were sound and showed that the syntax and semantics of a document can influence what readers recall and comprehend. The researchers also suggest several options for further research that may be interesting to pursue for my own research project. For these reasons, this article will be appropriate for my research project.
No comments:
Post a Comment