Writing Development Through Three Feedback Methods

Abstract

This study aimed to examine and classify features of errors in paragraph writing and to
investigate the effects of different methods of feedback upon accuracy and fluency in paragraph
writing. Data were collected from four paragraphs written by 105 participating students in three
groups, each group receiving one of three different methods of feedback on three progress
writings. Group A (N=32) received direct feedback, i.e., corrections were made for all errors on
students’ paper. Group B (N=38) received comment at the end feedback, i.e., all errors were
underlined and comments were written at the end of students’ paper pointing out the types of
errors without teacher’s correction. Group C (N=35) received coded feedback, i.e., all the errors
were underlined and coded with symbols representing kinds of errors without teacher’s correction.
The data were analyzed and tested by using Wilcoxon Signed Ranks Test. Analyses on all four
writings found 27 features of errors. The three most frequent features of errors were wrong
“sentence structure”, wrong use of “noun” and wrong or inappropriate “word choice”. Statistical
tests found significant decrease of total error rate in the final writing in every group. Meanwhile,
the amount of words significantly increased in group A and group C.