52
3.5.4.1 Validity and Reliability
The researcher wanted to know whether the instrument was good or not, so some criterias were considered. The criterias of a good instrument were validity
content validity and construct validity and reliability.
3.5.4.1.1 Content Validity
The content validity of the questionnaire was based on the objective of the research. The objective of the questionnaire was
to measure the students’ perception after the researcher gave the treatment using KWL Know, Want to
know, Learned and Jigsaw techniques. Then, it was clear that the questioannire was really needed to measure the perception of students on the implementation of
KWL Know, Want to know, Learned and Jigsaw techniques.
3.5.4.1.2 Construct Validity
The construct validity of the questionnaire was achieved by looking at the relationship among the indicators. The indicators measured the same aspect the
students’ perception, so it had positive association. Meanwhile, negative association was if the indicators measure different aspect.
3.5.4.1.3 Reliability
The reliability refers to the consistency of the test, and how the test can measure the same subject in different time but it has the same results Setiyadi, 2006:16.
In other words, the result of data are consistent in its score and give us an
53
indication of how accurate the test score are. To measure the reliability of the questionnaire, the researcher used Crombach Alpha. The researcher took the
questionnaire from expert, but it was important to measure the reliability one more time. It was caused by different research and the subject had different result
of reliability. The criterias of reliability using Crombach Alpha are:
0.80 up to 1.00 is very high 0.60 up to 0.79 is high
0.40 up to 0.59 is average 0.20 up to 0.39 is low
3.6 Scoring System
In scoring the result of the stu dents’ test, the researcher used Percentage Correct.
The percentage correct score was used in reporting the result of classroom achievement tests and the ideal highest score was 100. The researcher calculated
the average of the pre-test and the post-test by using the formula below:
� = 100
� �
Notes : P : Percentage of correct score
F : Number of right answers
N : Total number of items on test
Hatch and Farhady, 1982:46