IV Conclusions and suggestionsResul

IV Conclusions and suggestions
Results of the experiment described above provide the evidence necessary to maintain that students produce better language after observing subtitled similar task videos performed by native speakers in an online TBLL, in terms of fluency, as measured by the mean number of words per minute, and in terms of accuracy, as measured by the ratio of error-free T-units to total T-units. The effect of videos on fluency has also been confirmed by Borras and Lafayette (1994) and Weyers (1999) through post-test language production tests.
In addition, because observing subtitled similar task videos was administered as a pre-task activity in this study, the results suggest that pre-task activities improve fluency and accuracy of language produced during communicative task completion activities. This is congruent with the existing research in planning and prior-knowledge pre-task activities on increasing language fluency (Good & Butterworth, 1980; Foster & Skehan, 1996; Wendel, 1997; Wigglesworth, 1997; Mehnert, 1998; Skehan & Foster, 1999; Foster, 2001; Ellis & Yuan, 2004) and accuracy (Foster & Skehan, 1996; Mehnert, 1998). This result is also aligned with the previous research studies on the effect of rehearsing the same task as the assigned one on fluency (Bygate, 1996; Bygate, 2001; Lynch & Maclean, 2001) and accuracy (Bygate, 1996; Gass et al., 1999; Lynch & Maclean, 2001). However, this result is contradictory with the previous research studies on the effect of rehearsing similar tasks on accuracy (Gass et al., 1999) and fluency (Bygate, 2001). These contradictory results indicate that rehearsing similar tasks does not enhance fluency and accuracy of language production, but observing, rather than rehearsing, similar tasks increases language production in terms of fluency and accuracy. Therefore, it would be logical to assume that transfer of linguistic knowledge in the form of language production can be achieved by the suggestions of the input comprehension hypothesis rather than comprehensible output hypothesis in TBLL, through exposing students to input comprehension instead of asking them to practise language before practising a similar language. However, this claim needs to be examined in detail in a future study.
Also, the results support the study by Skehan (1998), who suggested that ‘repetition’ eases students’ cognitive ‘processing load’, which leads to allocation of more cognitive resources during task completion. This suggestion is observed in this study by the improvement in fluency and accuracy of language production during the task completion phase after repeatedly observing similar task videos performed by native speakers.
While statistical significance for syntactic complexity was not found in this study, partly because of the small number of participants, previous research studies showed that pre-task activities including observing subtitled similar task videos enhances syntactic complexity of language (Foster & Skehan, 1996; Skehan & Foster, 1997; Ortega, 1999). More research studies are needed before any conclusive remarks may be made on this issue.
Although the result regarding lexical complexity was not statistically significant, partly due to the small sample size, students who watched subtitled similar task videos produced lexically more complex language than students who did not watch the videos. This result is aligned with the previous study by Garza (1991), which reported that students who watched videos with subtitles scored higher on the use of the original lexicon of the video segments. In addition, the result also supports some of the previous pre-task
Downloaded from ltr.sagepub.com by guest on September 2, 2013
392 Language Teaching Research 14(4)
research on lexical complexity, which reported that the lexical complexity of language increased when students were provided with pre-task activities (Crookes, 1989; Yuan & Ellis, 2003). Some previous research, on the other hand, reported that pre-task activities do not improve lexical complexity, such as Gass et al. (1999), who found providing similar tasks reduced lexical diversity of language.
Future research studies should be conducted to examine the effects of subtitled similar task videos on language production in an online TBLL environment as compared to a face-to-face TBLL environment. Furthermore, the proficiency level of students could be tested to see its influence on language production in an online TBLL environment with subtitled similar task videos