This section provides an insight into the effects of CCQs on students‟
vocabulary acquisition. According to the vocabulary knowledge in the literature review, there are four types of vocabulary which are passive recognition, passive recall, active recognition, and active recall. This section represents the results of the independent t-test on these four types of vocabulary, then compares the mean scores
48
of the full test on vocabulary knowledge. Lastly, a Pearson correlation is shown to indicate the strength of the relationship between the use of CCQs and students‟
vocabulary acquisition. The tables in each part of this section are shortened to highlight the relevant statistics. Full results run by SPSS are shown in the appendix (appendix I, J, K, L).
4.5.1. Receptive vocabulary knowledge - Passive recognition
Table 4.5. Scores of Passive Recognition Vocabulary in the posttests of the control group and experimental group
Table 4.5 provides the statistics for the scores of task 1 about the recognition of passive vocabulary. As can be seen from the table, an independent t-test was run on the data with a 95% confidence interval for the mean difference. The result reveals that the significance level is 0,000 which is less than 0,05, providing enough strong evidence that the null hypothesis is rejected. Therefore, the two-population means are not equal. Closer inspection shows that the control groups achieved higher mean scores than those of the experimental ones. However, the gap between
Group Statistics
STUDENTS N Mean Std Deviation
POINTS control
experimental
51 51
2.441 2.167
.1627 .5447 Independent Sample Test
Levene‟s Test for Equality of
Variances
t-test for Equality of Means
Sig. 95% Confidence Interval of the Difference
Lower Upper
Points Equal variances assumed
.000 .1166 .4324
Equal variances not assumed
.1152 .4338
49
the two groups is not dramatically significant which is 0,274 since 0,5 are given for one correct answer. Hence, students studying with or without CCQs are able to have from four to five correct answers. Overall, the result indicates that the use of CCQs has no positive impact on the students‟ performance in terms of passive recognition.
4.5.2. Receptive vocabulary knowledge - Passive recall
Table 4.6. Scores of Passive Recall Vocabulary in the posttests of the control group and experimental group
Table 4.6 compares the statistics for the scores of task 2 about the recall of passive vocabulary. Similarly, an independent t-test was run on the data with a 95%
confidence interval for the mean difference. Undeniably, the null hypothesis is rejected because the significance level (0,000) is less than 0,05, leading to the fact that the control and experimental groups‟ mean scores are not equal. What stands out is the experimental groups achieve remarkably higher scores than the control ones (2,206 > 1,304). The gap between the two scores, which is 0,902, implies that the students learning with CCQs have about two more correct answers than the ones without CCQs. Therefore, it can be concluded that CCQs have positive effects on A2 students‟ passive recall vocabulary knowledge.
Group Statistics
STUDENTS N Mean Std Deviation
POINTS control
experimental
51 51
1.304 2.206
.8607 .4144 Independent Sample Test
Levene‟s Test for Equality of
Variances
t-test for Equality of Means
Sig. 95% Confidence Interval of the Difference
Lower Upper
Points Equal variances assumed
.000 -1.1673 -.6366
Equal variances not assumed
-1.1686 -.6353
50
4.5.3. Productive vocabulary knowledge - Active recognition
Table 4.7. Scores of Active Recognition Vocabulary in the posttests of the control group and experimental groups
In table 4.7, the result of the scores of task 3 about the recognition of active vocabulary knowledge is presented. In order to compare the mean variances, an independent t-test was run on the data with a 95% confidence interval. Looking into the table, it can be seen that the significance level is 0,000, much less than 0,05, which means the mean scores of the two groups are not equal. The experimental groups performed better with 2,392 mean scores while the control counterparts only got 1,980. Achieving 2,392 over 2,5 indicates that the majority of students in the experimental groups did not make any mistakes in task 3. In contrast, those in the control ones got approximately one incorrect answer. Generally, the result provides important insights into the positive effects of CCQs on A2 students‟ active recognition vocabulary knowledge.
Group Statistics
STUDENTS N Mean Std Deviation
POINTS control
experimental
51 51
1.980 2.392
.7481 .2305 Independent Sample Test
Levene‟s Test for Equality of
Variances
t-test for Equality of Means
Sig. 95% Confidence Interval of the Difference
Lower Upper
Points Equal variances assumed
.000 -.6292 -.1943
Equal variances not assumed
-.6311 -.1925
51
4.5.4. Productive vocabulary knowledge - Active recall
Table 4.8. Scores of Active Recall Vocabulary in the posttests of the control group and experimental group
Table 4.8 compares the statistics for the active recall vocabulary knowledge of the control and experimental groups. An independent t-test was run on the data with a 95% confidence interval. The significance level calculated is 0,000, less than 0,05, resulting in the fact that the two groups‟ mean scores are not equal. The experimental groups attain a much better mean score which is 2,206, outweighing that of their counterpart. While the majority of the control group failed to complete task 4, the experimental one had about four or five correct answers.
The considerable gap (1,324) indicates the incredible positive impacts of the use of CCQs on students‟ productive recall vocabulary knowledge.
Group Statistics
STUDENTS N Mean Std Deviation
POINTS control
experimental
51 51
.882 2.206
.9033 .4379 Independent Sample Test
Levene‟s Test for Equality of
Variances
t-test for Equality of Means
Sig. 95% Confidence Interval of the Difference
Lower Upper
Points Equal variances assumed
.000 -1.6024 -1.0447
Equal variances not assumed
-1.6037 -1.0433
52
4.5.5. Vocabulary knowledge test
Table 4.9. Scores of vocabulary knowledge in the posttests of the control group and experimental group
Table 4.9 gives information about the statistics for the scores of the full test about students‟ vocabulary knowledge. An independent t-test was run again on the data with a 95% confidence interval for the mean difference. The result shows that the significance level is 0,000 which is less than 0,05, making the null hypothesis be rejected. As a result, the two-population means are not equal. The mean scores of the experimental group are much higher than that of the control one (9,235>6,343), which can be interpreted that the experimental group performing better in the test.
Furthermore, the standard deviation of the control group is also much higher than that of the experimental one (2,4114 > 0,8268), leading to the fact that the gap in the vocabulary knowledge in the experimental groups is more equal to their counterparts‟. The aforementioned evidence proves that CCQs have a positive effect on A2 students‟ vocabulary acquisition.
Group Statistics
STUDENTS N Mean Std Deviation
POINTS control
experimental
51 51
6.343 9.235
2.4114 .8268 Independent Sample Test
Levene‟s Test for Equality of
Variances
t-test for Equality of Means
Sig. 95% Confidence Interval of the Difference
Lower Upper
Points Equal variances assumed
.000 -3.6004 -2.1840
Equal variances not assumed
-3.6058 -2.1785
53
4.5.6. Correlation between asking CCQs and students’ performance in vocabulary acquisition
Table 4.10. The correlation of the students‟ learning with CCQs and their points in the test
STUDENTS POINTS
STUDENTS Pearson Correlation Sig. (2-tailed) N
1 102
.630**
.000 102 POINTS Pearson Correlation
Sig. (2-tailed) N
.630**
.000 102
1 102
**.Correlation is significant at the 0.01 level (2-tailed)
Table 4.10 presents the correlation coefficient of the students learning with CCQs and their points. In this case, the figure of 0,0630 shows a relatively strong positive correlation between the two variables, which means the students studying with CCQs tend to have better points on the vocabulary test, and the effect is quite noticeable. Another figure to be explained is the 2-tailed significance value, which in this table is 0,000 < 0,05, providing strong evidence that the correlation above is highly significant. Therefore, there is enough statistical data to conclude that the null hypothesis is accepted.
4.5.7. Discussion of the independent t-test and Pearson
To summarize, in all of the tests, there is a significant difference between the means of the two groups. To be specific, most of the statistics of the experimental group outweigh those of the control ones apart from the passive recognition vocabulary knowledge. As a result, the null hypothesis is rejected; students studying with CCQs perform better than their counterparts studying without CCQs.
The correlation coefficient also provides more evidence for the above conclusion. The positive correlation proves that CCQs have a positive effect on
54
students‟ vocabulary acquisition, causing the null hypothesis to be accepted. These results are in line with what is expected in the literature review. However, what is outstanding is that CCQs do not have a positive effect on the passive recognition knowledge; students in the control groups even obtain higher mean scores than the experimental ones. This result is out of the expectation when CCQs do not have any positive effect on students‟ passive vocabulary recognition. It may result from the different English competence of students so that the control groups performed better than the experimental groups. Another point taken into consideration is that recognition vocabulary is easier than recall vocabulary since students completed multiple-choice exercises and did not need to remember the spelling.
Comparing the mean scores of the two groups, the gaps are most significant in recall vocabulary knowledge, which are 0,9 for passive recall vocabulary and 1,4 for active recall vocabulary. This result indicates that CCQs affect the recall vocabulary knowledge the most.