Results of the autonomy scale

The autonomy scale was comprised of four main sections, responsibilities, abilities, motivation, and activities; furthermore, both responsibilities and activities included two subsections. Responsibilities consist of the participants’ perceptions of their teacher’s responsibilities and their own responsibilities for learning English, and

activities contained students’ autonomous behavior shown in the in-class and extracurricular activities. The results of the autonomy scale were further discussed based on the six categories of autonomy respectively.

The autonomy scale was used as the pre-test and the post-test to investigate the differences between the experimental and control group on their learner autonomy after the eight-week online extensive reading program. Means and standard deviations of the two groups on the pre-test and post-test observations are shown in Table 4.

Table 4. Descriptive Statistics of Category of Autonomy on Pre-test and Post-test

As Table 4 revealed, in general, when comparing with the pre-test, the experimental group obtained higher mean on the post-test, but the control group did not (See Table 4). Instead, the control group had a slightly decrease in their total mean on the post-test. Furthermore, the experimental group demonstrated higher total mean on the post-test than the control group. When it comes to the analysis of each category separately, the experimental group also obtained higher means than the control group in terms of all categories by comparing group means. In order to determine whether

Pre-test Post-test

Control Experimental Control Experimental

Mean SD Mean SD Mean SD Mean SD

Ss’ responsibilities 43.65 11.59 45.04 9.76 41.35 12.74 47.19 7.89 T’s responsibilities 28.89 9.39 31.89 9.37 30.37 11.24 31.89 7.80

abilities 35.60 9.97 37.45 8.18 35.24 11.48 40.12 8.92

motivation 3.12 1.09 2.99 .90 2.82 1.15 3.06 .97

out-of-class activities 40.77 13.10 40.78 10.22 40.65 12.94 43.77 10.40 in-class activities 10.65 3.78 10.88 3.24 11.30 3.95 11.42 3.62 Total 162.69 31.67 169.02 23.52 161.71 31.82 177.46 23.80

these differences were statistically different or not, an independent-samples t-test was conducted to examine if the differences between the two groups reached a significant level .05 on the pre-test as well as on the post-test.

Table 5. Independent T-test of Learner Autonomy on Pre-test and Post-test Pre-test independent-samples t-test indicated that there were no significant differences between the experimental and control group on the pre-test on their autonomy, p=.144 (See Table 5); in other words, the two groups did not differ from each other on their learner autonomy before the intervention. After the treatment of the eight-week reading program, an independent-samples t-test was conducted again to explore if there were some statistically significant differences between the two groups on the post-test on their learner autonomy. The results unveiled that some significant differences were found between the two groups on the post-test on their learner autonomy, p=.000. The magnitude of the differences in the means was medium (d=.58) (See Table 5); therefore, it indicated that the experimental group significantly outperformed the control group on their learner autonomy on the post-test.

Furthermore, the results of the autonomy scale could also be analyzed based on the six categories of learner autonomy between the two groups. The results indicated

that there were some statistically significant differences between the experimental and control group on their perception of teacher’s responsibilities in English learning on the pre-test, p=.040 (See Table 5), but no significant differences showed between the two groups on the other five categories (for students’ responsibilities, p=.406; abilities, p=.192; motivation, p=.399; out-of-class activities, p=.996; in-class activities, p=.680).

However, after the intervention of the eight-week reading program, the results revealed that the differences of students’ responsibilities (p=.000) and abilities (p=.003) between the two groups attained a statistically significant level .05 on the post-test (See Table 5). Cohen’s d statistic indicated a medium effect size on students’

responsibilities (d=.61) and abilities (d=.48) as well, which denoted that the experimental group significantly increased their own responsibilities and abilities on autonomous learning than the control group after the eight-week online extensive reading program. Also, it is noted that the differences of their perception of teacher’s responsibilities in English learning between the experimental and control group reached the significant level .05 on the pre-test; nevertheless, the differences between the two groups failed to reach a significant level .05 on the post-test. An inspection of the mean scores suggested that the control group obtained higher scores (M=30.37) on the category of teacher’s responsibilities on the post-test while the experimental group obtained the identical scores (M=31.89) on the pre-test and the post-test.

In addition to a comparison between the group means, a pair-samples t-test was performed to recognize the differences within each group between the pre-test and post-test. For the control group, no significant differences were found on their overall autonomous learning behavior between the two tests (p=.737), and the magnitude of the differences in the means were very small (d=.04) (See Table 6). However, when comparing the means based on the six categories of learner autonomy, there was a statistically significant decrease in their motivation mean scores from Time 1 (M=3.12)

to Time 2 (M=2.82), p=.011. The Cohen’s d (.29) indicated a small effect size (See Table 6). The possible reason may be the different input of English. Since the content of English textbook became more difficult than previous ones, no matter on magnitude of the differences in the means were close to medium (d=.47), as shown in Table 6. Further analysis of the means of each category indicated that the increase of students’ responsibilities (p=.029), abilities (p=.003), their autonomous behaviors displayed in the out-of-class activities (p=.000) as well as in the in-class activities (p=.037) attained a statistically significant level .05 between the pre-test and the post-test. Cohen’s d ranged from .23 to .44, revealing a small to medium effect size.

Table 6. Pair-samples T-test of Learner Autonomy on Each Group Control group

In a word, the results of the autonomy scale showed that on the pre-test, there were no significant differences between the control and experimental group generally;

however, the differences between the two groups attained a significant level on the post-test. It’s clear that the experimental group did outperform the control group on the post-test after the eight-week intervention of the online extensive reading program, in terms of their perception of their own responsibilities and autonomous abilities in English learning in particular. When it comes to the differences within each group based on time of measurement, for the experimental group, there were some statistically significant differences between the pre-test and the post-test in general while the differences between the two tests did not reach a significant level for the control group.