http://www.uri.edu/artsci/com/Logan/teaching/html/wrt333/assign_2011/a3_style.htm

WRT333

Style and Readability (2 parts)
Given Sep. 26 | Due Oct. 5

Syllabus | Assignments | Table of Pages

Preparation: Review lecture notes from week 4:

This is a two part assignment worth, in total, 15% of your semester grade.

Part 1: Style and Readability Exercises

These exercises are based on class handouts.

Assignment: Complete the following exercises, writing your answers in pencil or pen on a copy of the handout or on a separate sheet of paper clearly labeled to identify the exercise. Include your name on all papers submitted.

How will Part 1 be graded? This part is worth 10% of your final grade. There are 30 sentences to be revised. Each correct revision is worth 1/3th%.

Part 2: The Homework Machine

(Here is an MSWord copy of this part of the assignment)

Recognizing and Correcting Writing Flaws*

Situation: Professor Bullfinch has asked you to review the Results section from his latest scientific journal article. Bullfinch and his associate, Dr. D. Dunn, have invented a machine which writes essays, solves arithmetic problems, and even does book reports. Dunn and Bullfinch have tested the machine against the performance of students in a writing class.

In their methods, Dunn and Bullfinch describe how 10 students were recruited out of 20 in the class. The 10 agreed to turn in homework generated by the machine. Neither the instructor nor the other 10 students were told of the experiment. Professor Bullfinch compared homework generated by the machine to homework written by students by comparing grades and by evaluating the individual grammatical and stylistic errors noted by the writing instructor. To show off their invention, Professors Dunn and Bullfinch have programmed the homework machine to generate the following Results section to summarize findings. Wisely, they have also discretely asked you to check the machine's writing—carefully.

The Dunn-Bullfinch homework machine has been programmed to mimic human writing patterns. The errors which you find in the text will help to improve the machine. Edit carefully. If the homework machine can be perfected, you will have contributed greatly toward the reduction of suffering among generations of future scholars (and their professors)!

Assignment: Editing is often more fun than writing. If you know what to look for, editing becomes a game, a version of "What's Wrong With This Picture?" The following Results text contains a number of flaws, including flaws of content (Methods or Discussion instead of Results) and flaws of readability (noun clusters, passive voice constructions, nominalizations, jargon, excessive left-branch length, split infinitives, errors of voice or tense, subject and verb agreement errors, sentence fragments or incomplete thoughts, and dangling modifiers). You are asked to discover these flaws and to correct them.

At least 20 flaws can be found (including passives like this one). You are asked to underline these and to mark each one with a superscript. On a separate sheet, write the number and a short description telling what has caused the flaw (e.g., "passive construction").

When you have completed marking the document and describing its flaws, rewrite it, making appropriate changes in word, phrase, or sentence. You may rewrite by hand but be certain that your writing is large and clear enough for your ancient instructor to decipher. You may also rewrite by machine, but please attempt to align your correct text with the flawed text it replaces (cut and paste if this will help).

Make two copies of the flawed text. Use one as a scratch version, appropriate for initial marking and note taking. Edit this first, then transfer appropriate underlines and comments to a neatly edited final copy for submission. Bring to our next class:


*With apologies to Jay Williams and Raymond Abrashkin, authors of Danny Dunn and the Homework Machine. McGraw-Hill Book Company.

Results

The difference between the homeworks written by the machine and those written by the students were evaluated using the SAS (Statistical Analysis System) analysis of variance (ANOVA) procedure. Differences were measured by taking the average of grades for each of the two student groups.

Average grades for homework generated by the machine were not significantly different from grades for student generated homework. For four written assignments, the average numerical evaluation score for machine homeworks were 2.45. Comparing this to the students, their average score is 2.41, which is not significantly different (F=1.98, α = .05).

Enumerating instructor correction indication marks under six categories. Comparison of the machine to the students was based on 1) nominalizations and noun clusters, 2) agreement between pronouns and there antecedents, 3) agreement in number between subject and verb, 4) use of passive voice constructions, 5) correct use of tense, and 6) use of jargon or euphemisms.

Incidence levels of noun group arrays were determined to be not different between the machine and the students (F=1.49, α = .05). These clusters, including 3-noun clusters, of which the machine constructed fewer than the students, who tended, however, to, in general, write more 4-noun cluster phrases, were never not in any of the papers, but seem to occur with a frequency of about one per paragraph.*

Generation of errors was consistently of lesser frequency by the homework machine when caused by disagreement between pronoun and antecedent or by disagreeing between subject and verb. A considerable number of the students (83% of student papers) have subject and verb agreement problems, most frequently caused by disagreement in number (e.g., using a singular verb with a plural subject). This compared to only 7% of the machine homeworks.

*Exception, an individual machine error excess string-length noun cluster generation tendency problem.

Although lowering of student grades occurred because of flaws in subject / verb agreements, a causal compensating factor resulting in the lack of differences in grades was the machine's inability to suppress jargon. Although the machine, which clearly and obviously possesses a markedly larger vocabulary than the students, did not overextend itself by using obscure words, it nevertheless could not entirely accomplish elimination of a tendency toward somewhat overblown and pretentious phraseology.

From the analysis, it must be concluded that there appear to be lesser differences between the machine and the students in their written word utilization abilities. It should be clear that if homework machines can learn to suppress their loquacious propensities and if they can significantly reduce tendencies toward turgidity, they no doubt will eventually prove superior to the students, although this may be it's own un-doing as instructors will be able to detect the difference in homeworks.

How will this part be graded? This part is worth 5% of your final grade. You will be given 1% for each example you copy and explain (but only if it succeeds in illustrating the grammatical point listed), and if you correctly rewrite it. Make certain that your rewritten sentences each correct the problem that you have said obscures meaning for general readers.