Noticing Results

48

49

Table 4.1 Mean and SD of first and second time underlining (noticing) scores of each group

first time second time

Mean (SD) Mean (SD)

Non-output (n=23) 3.26 (2.44) 2.78 (2.13) Cloze (n=28) 2.89 (2.25) 5.04 (3.80) Editing (n=28) 2.36 (1.91) 4.36 (1.95)

Figure 4.1 Mean scores of first and second time underlining of three groups

As the table shows there is an increase between first and second time underlining scores for both output groups. Non-output group, on the other hand, has failed to achieve the same. This indicates that the participants who were involved in two output tasks underlined more items related to passive voice in the subsequent input activity. The results were submitted to one -way ANOVA analysis to examine whether there was any significant difference between the groups.

The results revealed that there was no sig nificant difference between scores on first underlining (F (2, 76) = 1.10, p = .34) which means before any of the tasks, the noticing scores were same.

However, significance was found on second underlining scores (F (2, 76) = 4.25, p = .02) which took place after each group carried out the required task respectively. This suggests that those who were engaged in doing output tasks managed to outperform those who were engaged in

0 2 4 6 8 10 12

First time second time

Non-output Cloze Editing

50

non-output task, in this circumstance, a true or false task.

Furthermore, a post hoc comparison using the Fisher LSD test revealed that there was significant difference between non -output group and cloze group (p<.01 ), non-output group and editing group (p<.05 ), while no significant difference was found between cloze and editing g roup.

LSD was chosen because it was suggested to be the most powerful post hoc, especially when there were only three groups in comparison (Larson -Hall 2010). Though both output groups improved on their second underlining scores, and the results of the pai red-samples t test showed that their improvement from first to second time underlining scores were significant (t (27) = -2.99, p <.01 for cloze group; t (27) = -4.67, p <.01 for editing group), the between group comparison through one-way ANOVA analysis revealed no significant difference (F (1, 54) = 0.71, p = .40).

Therefore, though output groups outperformed non -output group on triggering noticing, which output task is more effective in doing so was not testified. Thus it can be concluded from these findings that output tasks in this study triggered the noticing of the target form compared to the non-output task and consequently lent suppo rt to the noticing function of output hypothesis along with many previous studies (Song & Suh 2008; Uggen 2012).

However, there was no evidence found on the differential effect of task type. To further explore this issue, the numbers of attempts participants produced on the output task sheet were compared, which is shown in following section.

4.1.2 Number of attempts in task sheets of two output groups

The table below summarizes the mean scores along with standard deviation of numbers of attempts in first and second output for the two output groups and the following figure illustrates the increase of the numbers from first to second output task of each group.

51

Table 4.2 Mean and SD of numbers of attempts of two output groups

first time second time

Mean (SD) Mean (SD)

Cloze (n=28) 3.00 (2.45 ) 7.14 (3.08) Editing (n=28) 5.00 (2.67) 7.61 (2.96)

Figure 4.2 Mean numbers of attempts of two output groups

As the table shows, the editing group outperformed the cloze group in the numbers of attempts manifested during their first output. Though the editing group also showed more numbers of attempts on second output than the cloze group, the numbers were very close. However, both groups managed to manifest an increase in the numbers of attempts from first to second output.

The results of the one-way ANOVA revealed that there was a significant difference between numbers of attempts in first output (F (1, 54) = 8.54, p =.005). This indicates that the editing task managed to direct participants’ attention to the target forms more than the cloze task did.

Therefore, at least, it shows an advantage of editing task over cloze task on directing learners ’ attention to target form.

However, no significant difference was found between the numbers of attempts during their second output (F (1, 54) =.33, p = .57). A paired-samples t test comparison was conducted to see

0 2 4 6 8 10 12

first time second time

Cloze Editing

52

whether there was any statistically significant difference between the numbers of first and second time attempts for each group. The results revealed that the increase in numbers were significant for both groups (t (27) = -8.86, p <.001 for cloze group; t (27) = -6.94, p <.001 for editing group).

This increase in numbers indicates that both output tasks managed to draw learners’ attention to target form though it is unclear which task is more bene ficial since though editing task appeared to draw more attention at first output, that was just after the first input phase, it did not show similar advantage in the second output, which was completed immediately after the second input phase.

The reason attributed to this might be the different characters of the tasks. Editing task was designed in an error correction form which would direct learner attention to the erroneous parts regardless of how much engaged the learners in the previous input activity, while the cloze task which requires learners to fill in the missing words or phrases was more closely related to the previous input activity. This also explains why in second output the cloze group managed to improve numbers of attempts significantly and editing task, despite the advantage in first output failed to outperform cloze group. The same can be observed from the figure, the increase in numbers of attempts from first to second time for cloze group is sharper than that of editing group.

Apart from looking into the underlining scores of all groups or the number of attempts made by two output groups, the gain in the respective task of each group should also be examined in order to provide a further evidence to demonstrate the facilitating effect o f noticing on the task performance.