Instruments and Data Gathering Technique
25
Table 3.1 The Data Analysis Table No
Problem Data Required
Instrument
1. How guessing meaning from
context procedure is implemented in vocabulary
learning in class 8A of SMP Negeri 15 Yogyakarta
The students‟ works and the
result of observation
1. Student
worksheet 2.
Observation sheet
2. What are the results of the
implementation of guessing meaning from context procedure
in vocabulary learning in class 8A of SMP Negeri 15
Yogyakarta The students‟
pretest and posttest results, the
students‟ responses toward
the questionnaire 1.
Pretest and posttest
2. Questionnaire
In order to find the answer to whether or not the guessing meaning procedure could help the students in class 8A of SMP Negeri 15 Yogyakarta, the
researcher had to compare the mean of the pretest result of the research subjects to the mean of the posttest using dependent t-test. For determining whether or not
there had been a significant, positive difference between the mean of the pretest and posttest score, the researcher used dependent t-test. Figure 3.1 shows the
formula for the dependent t-test.
Figure 3.1 The Dependent t-Test Formula
26 Where:
t : t ratio
D : average difference
∑D
2
: different scores squared, then summed ∑D
2
: different scores summed then squared N
: number of pairs The result of the t-test, known as t
, would then determine the researcher‟s decision to either retain or reject the null hypothesis.
Ary, Jacobs, and Razavieh et al. 1990 explain “the null hypothesis is a statement that there is no actual relationship between the variables and that any
observed relationship is only a function of chance” p.162. A null hypothesis must always be tested in a negative sentence and can either be retained or rejected
by a researcher depending on the result of a statistical test done previously. In order for a researcher to retain or reject the null hypothesis, the t-test must either
exceed or be less than t-values required for significance at a certain level. By retaining the null hypothesis, a researcher states that he did not find any or less
than enough evidence to prove that his experimentation had led to a change he had expected, or that any change observed after the experimentation could not be
attributed to the variables he had modified. By doing the opposite, researcher state that the change observed after the experimentation was likely to be the result of
their experimentation; that the variables that had been modified caused the PLAGIAT MERUPAKAN TINDAKAN TIDAK TERPUJI
27 change. This statistical research theory would be used by the researcher to answer
question number one. The researcher formed an assumption of the null hypothesis in this research
as mentioned below: Null Hypothesis H
o
There is no significant difference between the mean of the pretest scores and the mean of the posttest scores.
Alternative Hypothesis H
1
The mean of the posttest scores is significantly higher than the mean of the posttest scores.
The theory of level of significance is an important part of the null hypothesis. Ary et al. 1990 define the level of significance as the “predetermined
level at which a null hypothesis would be rejected” p.165. The researcher must determine the level of significance before she carries out her research. The level
of significance one chooses may vary, but the most used level of significance is the .05 level, as was the level of significance set by the researcher before
experiment. If a researcher decides that she uses the .05 level significance for her research and based on the test of significance she rejects the null hypothesis, she
implies that she does not believe that “the null hypothesis is true because the chance is only 5 out of 100 .05” Gay, 1992, p.431.
Apart from the theory of level of significance, another inseparable part of any type of tests of significance is degrees of freedom. According to Gay 1992,
degrees of freedom constitute “a function of such factors and the number of PLAGIAT MERUPAKAN TINDAKAN TIDAK TERPUJI
28 subjects and the number of gro
ups” p. 434. Different formulas to determine degrees of freedom apply depending on the test of significance a researcher would
like to carry out. This research for instance, used the t-test for dependent samples. The formula used to determine the degrees of freedom is df = N-1, with N being
the number of research subjects. There are 33 students in the 8A class but only 30 students who are involved in this study, so the degrees of freedom of the study are
29.