Developing Writing Diagnostic test on Thai curriculum of Matthayomsueksa 6 under R0i-Et Primary Educational Service Area 3

Abstract

Learning activities need to measure and evaluate students’ achievement
in the same time. The defect diagnostic test in writing skill on Thai curriculum of
Matthayomsuksa 6 aimed to discover what parts of the content that students
not understood and helped teacher to manage class effectively. Writing skill was
the important basic skill of language learning both basic and advance learning.
This research aimed to construct diagnostic writing skill test on Thai curriculum
for Matthayomsuksa 6 under the R0i-Et Primary Educational Service Area 3, to
determine the quality of the writing diagnostic constructed test, including
determining the difficulty power, validity power, discrimination power, and
reliability power. The samples used were 380 Matthayomsuksa 6 students from
21 schools under the R0i-Et Primary Educational Service Area 3, semester 2,
2010 academic year, through multi-stage random sampling. Instruments used
were 2 tests, including test 1 : explore test with cloze test, divided into 4 parts
120 items, tested by 110 people from samples, and test 2 : diagnostic test
with multiple choice test designed with 4 alternatives which constructed from
explored defect test. The test was tested for three times. First, diagnostic test
with 80 items, tested by 95 samples in order to determine the item difficulty
and item discrimination powers, improved and classified the ineligible test off.
Second, the diagnostic test with 72 items, tested by 80 samples for
determining the item difficulty and item discrimination powers, chose 60 items
for eligible test. Third, the diagnostic test included 60 items which tested by 95
samples for determining the item difficulty and item discrimination powers, and
test reliability. The collected data were analyzed by percentage, mean and
standard deviation.

The research results appeared as follow :
1. The first test, it was found that the diagnostic test with 4 parts
80 items had the item difficulty powers ranged from 0.17-0.91, the item
discrimination powers ranged from 0.02-0.88, then, improved and 72 items were
chosen to test in the second test.
2. The second test, it revealed that the diagnostic test with 4 parts
72 items, chosen 60 items of eligible test, the item difficulty power ranged
from 0.30- 0.89, the item discrimination power ranged from 0.21 - 0.91 and then
took to the third test.
3. And the third test, it revealed that the diagnostic test with 4
parts 60 items, the item difficulty powers ranged from 0.21-0.86, item discrimination
powers ranged from 0.20 - 0.79, and test reliability was 0.77, mean score was
10.82, standard deviation scores were 1.74, 1.74, 2.01, 1.32, score point was 9,
and mean scores were 10.82, 10.93, 5.05, 10.72.
4. The results of determining the defect of the test from the third
test revealed that students had the defects, namely spelling which equally to
50.78 percent , using words, which equally to 28.33 percent, and composing the
sentence which equally to 20.89 percent. So, it was regarded as the defects of
student from the diagnostic test could be analyzed the defects in Thai language
learning.