3.8 Method of Data Analysis
3.8.1 Normality Test of Pre-test and Post-test Data
This test is used to know whether the data that will be analyzed have normal distribution or not. Here, the writer used chi-square formula.
å
=
- =
k i
i i
i
E E
O
1 2
c Where:
2
c = chi square
i
O = observation frequency
i
E = expected frequency k = the numbers of interval class
i = 1,2,3,....,k Sudjana, 2005: 273
Criterion
For α= 5 and df= k-1, if
2
c
value
2
c
table,
the data is normally distributed.
3.8.2 Test of Significance
To see whether the means of pre-test and post-test was statistically significant, the t-value should be obtained and consulted with the critical value in
t-table. The writer used the t-test formula:
In which, t
= t-value M
d
= mean from different score between pre-test and post-test x
d
= deviation of each subject d-M
d
∑x
2
d = total deviation square N
= number of subject df
= N-1 Arikunto 2006, 306
Criterion
There is a significant difference between pre-test and post-test if t t
table
CHAPTER IV RESULT OF STUDY
This chapter presents the result of the research, including the result of try- out, pre-test, and post-test, normality of pre-test and post-test data, the difference
between two means of pre-test and post-test, test of significance, and the discussion of the result.
4.1 The Result of Try Out
The result of try out was analyzed based on item validity, level of difficulty, discriminating power, and reliability. Those aspects influenced the
number of items that were used in the pre-test and post-test. Therefore, those aspects were very important to calculate.
4.1.1 Item Validity
Concerning with validity, Heaton 1979: 152 proposes that the validity of a test is the extent to which it measures what it supposed to measure. In this
research, the writer used the item validity and calculated them by using Point Biserial Correlation.
For example, the validity index of number 1 is 0.40. Then the writer consulted the table of r with N = 40 and significance level 5 is 0.312. Since the
result of the computation is higher than r in the table, the index validity of item number 1 is considered to be valid. See Appendix 4
36
Kategori