4. Findings and discussion
4.1. Findings from the pre-test and post-test
information gap activities, questions, and answers, role play, games, telling stories, think-pair-share.
3. Methodology
Action research has been carried out to analyze a particular problem or procedure in a teacher’s field, where the teacher in charge would like to find a solution. It was performed with students in a class by the teacher-researcher. The researcher claims, for these reasons, that action research designed by Kemmis and McTaggart (1998) is acceptable.
This study is conducted on first-year ADA at ADAFA. A series of lectures are currently being used in the GE program of the subjects of the study. Cadets can practice all skills, particularly by speaking with their current content, as a GE-level for ADA officials and staff. To comply with the GE program, the cadets have to attend 12 lessons.
The research subject is the first-year cadets in the class of ADA at ADAFA, which consists of 33 male cadets. Because every year, there are more than 200 cadets enrolled in the Academy, and none of them are female. The class’s members were chosen by taking the entrance examination of group A1 (Mathematics, Physic, and English). All 33 cadets were in the academic year of 2020-2021, and they are from 18 to 23 years of age. Almost all of them feel unfamiliar with the military environment, which is very strict and challenging because they were high-school students from mountainous to suburban areas in Vietnam. Besides, their discrepancies in locations and forms of entry exam somehow contributed to dramatically different levels of their English. Their English ability level in the Common European Reference System is equivalent to A2 - B1 when they finish the GE program. There are various characteristics of cadets in military institutes as a whole and ADAFA. Twice a day, the cadets attend the ADAFA class. They have six periods each morning, depending on their schedule, for two or three different subjects. The afternoon provides four self-study periods in classrooms under the guidance of a platoon commander, a newly graduated officer. After that, they have to do other military duties as planned.
Four instruments, such as speaking tests, questionnaires, interviews, and observation, were deployed to address the problems.
Table 4.1. Cadets’ scores in pre-test and post-test
No. Cadets Pre-test Post-test
1 Cadet 01 5.2 5.6
2 Cadet 02 3.6 4.4
3 Cadet 03 6.4 6.0
4 Cadet 04 6.0 6.8
5 Cadet 05 4.8 5.2
6 Cadet 06 4.0 4.4
7 Cadet 07 4.4 4.4
8 Cadet 08 8.0 8.8
9 Cadet 09 6.8 7.6
10 Cadet 10 8.0 9.2
11 Cadet 11 5.2 5.2
12 Cadet 12 6.0 6.8
13 Cadet 13 4.0 4.4
14 Cadet 14 8.4 9.2
15 Cadet 15 7.2 8.0
16 Cadet 16 9.2 9.6
17 Cadet 17 4.8 5.6
18 Cadet 18 5.6 5.2
19 Cadet 19 7.6 8.4
20 Cadet 20 6.0 7.2
21 Cadet 21 3.2 4.0
22 Cadet 22 3.2 4.4
23 Cadet 23 5.2 5.2
24 Cadet 24 6.0 6.8
25 Cadet 25 6.8 7.2
26 Cadet 26 4.4 5.2
27 Cadet 27 4.0 4.4
28 Cadet 28 5.2 5.6
29 Cadet 29 6.4 7.2
30 Cadet 30 9.2 9.2
31 Cadet 31 5.6 6.4
32 Cadet 32 4.0 4.0
33 Cadet 33 3.6 4.0
*(5.0 is the criterion minimum of completeness)
From the data calculated above, the researcher found that the percentages of cadets who passed the criterion minimum of completeness in the post-test from the class ADA had improved compared to the pre-test.
Descriptive statistics
The results from both tests have been applied to and evaluated for the SPSS program.
The descriptive data primarily include statistics such as minimum, maximum, and mean scores. The results and descriptive data from the pre-test and post-test can be seen in Table 4.2.
Table 4.2. Descriptive Statistics of pre-test and post-test
N Minimum Maximum Mean Std. Deviation
Pre-test 33 3.2 9.2 5.69 1.68531986
Post-test 33 4.0 9.6 6.23 1.76360512
In order to understand the progress of cadets who passed the criterion minimum of completeness, the researcher also measured the percentage of cadets’ speaking scores. In the pre-test, it was observed that 63.63% of cadets passed the criterion minimum of completeness. It means that there were twenty-one cadets who graduated and twelve cadets who had a ranking under the criterion minimum of completeness.
Next, the researcher did a post-test to know if cadets’ speaking ability was improved.
The average post-test score is 6.23, and the cadets’ score from pre-test to post-test is improved to 0.54.
Figure 4.1. Classification of Test Results in Pre-test and Post-test
According to figure 4.1, the statistics showed that the cadets’ scores improved between the test and post-test. Fail and pass scores in the post-test were lower than it in the
pre-39.39%). Besides, there were upward trends in results on excellent and fair scores by two times in post-test compared to the pre-test. Both of these scores increased from 6.06% to 12.12%. Meanwhile, the percentage of cadets who got good marks on both tests remained the same. This indicated that there are differences in cadets’ scores for the use of communicative activities.
Paired samples test statistics
In detail, cadets’ scores in terms of five criteria in post-test increased significantly compared to the pre-test. The researcher also used the SPSS program to analyze these factors.
Table 4.3. Paired samples test of pronunciation criteria
Criteria Mean score N Sig. (2-tailed)
P1 1.248 33
0.000
P2 1.406 33
Table 4.3 expressed that the mean score of pronunciation criteria in the pre-test (P1), which constituted 1.248, was considerably less than in post-test (P2) 1.406. The sig. 2-tailed showed a difference in pronunciation which was at 0.000 (< 0.05). In other words, the cadets’ pronunciation ability had been increased between the two tests.
Table 4.4. Paired samples test of grammar criteria
Criteria Mean score N Sig. (2-tailed)
G1 1.127 33
0.002
G2 1.248 33
As shown in table 4.4, the mean score of grammar criteria in the pre-test (G1) was 1.127; meanwhile, the post-test (G2) was 1.248. The sig. 2-tailed was 0.002 (< 0.05);
therefore, it could be concluded that the grammar criteria of cadets were improved in the post-test.
Table 4.5. Paired samples test of vocabulary criteria
Criteria Mean score N Sig. (2-tailed)
V1 1.224 33
0.032
V2 1.297 33
Table 4.5 showed that the mean score of vocabulary criteria in the pre-test (V1), which constituted 1.224, was significantly lower than the mean score of vocabulary criteria in the post-test (V2), accounting for 1.297. Besides, the sig. 2-tailed was 0.032 (< 0.05); hence, communicative activities for cadets had been effective regarding vocabulary criteria.
Table 4.6. Paired samples test of fluency criteria
Criteria Mean score N Sig. (2-tailed)
F1 1.079 33
0.012
F2 1.152 33
Table 4.6 indicated that the mean score of fluency criteria in post-test (F2) was 1.152, bigger by 0.073 compared to fluency criteria mean score in the pre-test (F1), which was 1.079. In addition, the sig. 2-tailed was 0.012 (< 0.05). It means there is a change in the cadet’s speaking fluency after the implementation of action research.
Table 4.7. Paired samples test of comprehension criteria
Criteria Mean score N Sig. (2-tailed)
C1 1.018 33
0.003
C2 1.115 33
In table 4.7, it could be seen that the comprehension ability of cadets was developed between pre-test and post-test as evidenced by the mean score of comprehension criteria in the pre-test (C1) being 1.018, while in the post-test (C2) was 1.115. The sig. 2-tailed showed an improvement in comprehension criteria between the two tests, which was at 0.003 (<
0.05).
From the above findings, it could be summarized that all five criteria were increased in which pronunciation and grammar abilities of the first-year ADA cadets were developed significantly through using communicative activities.
Test statistics from the Paired sample test have been used to analyze whether or not the average variations in before and after tests were meaningful. SPSS used paired samples technique for data processing for statistics. Table 4.8 shows the results of the SPSS data analysis tool.
Table 4.8. Paired samples test of pre-test and post-test
Firstly, the Sig. (2-tailed) was 0.000 (< 0.05), with a 95% confidence interval, it has been inferred those discrepancies were made between 2 variable groups: pre-test and post-test results. In addition, this value demonstrates the trust and importance of the paired samples test. Second, the average pre-test scoring value was 5.69, and the average post-test result was 6.23. As shown in Table 4.8, the first column, namely Mean, showed that the data was correlated between pre-test and post-test.
In summary, the results from both experiments demonstrated that the scores of the two tests were statistically meaningful.