This chapter will first summarize the major findings vis-à-vis the research questions proposed in Chapter Two. Then, discussions on the results revealed in this study will be presented, followed by the pedagogical implications, the limitations of the study, and finally the suggestions for future research.
Summary of the Major Findings
This current study aimed to examine the effects of captioning on EFL middle school students’ incidental vocabulary acquisition and investigate the relationship between the students’ English language competence and vocabulary acquisition through video viewing. The major findings will be summarized as follows in accordance with the three research questions in this study.
RQ 1: Does caption availability enhance EFL middle school students’ aural form recognition?
With respect to the first research question, the results indicated that the
participants in the captioned group, on the whole, could recognize significantly more aural word forms than those in the non-captioned group. This finding suggested that the availability of captioning could facilitate EFL middle school students’ aural form recognition from video viewing.
RQ2: Does caption availability improve EFL middle school students’ vocabulary acquisition?
Concerning the second research question, results suggested that the participants in the captioned group scored significantly better than those in the non-captioned group on overall vocabulary acquisition (i.e. form-meaning mapping). This result proved the positive effects of captioning on incidental vocabulary acquisition of EFL middle school students at a deeper level.
RQ3: Are the gains of aural form recognition and vocabulary acquisition modulated by the varying linguistic competence of EFL middle school students?
56
To answer the third research question, the participants’ average scores on the achievement tests from the previous three semesters were used to determine their English language competence. The results revealed that for both form recognition and vocabulary acquisition, scores obtained by the students at high level of linguistic competence differed significantly from the students at intermediate and low levels of linguistic competence. Such outcome suggested that the incidental learning of vocabulary knowledge through watching authentic cartoon videos was significantly influenced by students’ linguistic competence. In addition, no interaction effect
between the caption condition and students’ linguistic competence was found on either overall form recognition or vocabulary acquisition.
In conclusion, the present study indicated that the availability of captioning did markedly increase EFL middle school students’ spoken word gains in terms of form recognition and form-meaning mapping through watching video material. Moreover, the results also suggested that the students’ linguistic competence appeared to be a significant predictor of the amount of vocabulary gains through incidental learning in multimedia context.
Discussion
In this section, discussion on the findings indicated in Chapter Four is first presented, followed by the qualitative results based on the teacher’s classroom observation and the information collected from the follow-up questions.
Captioning and Aural Form Recognition
With regard to the first research question, this study provides strong evidence that the availability of captioning has beneficial effects on aural form recognition of EFL middle school students. To be more specific, the sixty Taiwanese middle school students who were exposed to the video material with captioning tended to recognize considerably more phonological word forms which were encountered previously in the video material. In contrast, the fifty-eight students in the non-captioned group tended to make substantially fewer word gains in terms of aural form recognition.
As mentioned earlier, the major concern about captioning is that for foreign language learners, the availability of captioning may be a distractor which may impede L2 development because viewers are likely to focus on reading the visual words
57
without paying attention to the oral input when captioning is available (Borrás &
Lafayette, 1994). To further clarify the role of captioning on L2 learners’ language acquisition, some studies (Markham, 1999; Sydorenko, 2010) examined its effects on phonological word recognition (i.e., knowing what the word sounds like) from viewing captioned video. However, the results were inconclusive and the issue remained unresolved. As previous studies revealed mixed outcomes, the findings in this present study are of importance to shed more light on the effects of captioning on phonological word recognition.
Furthermore, it should be noted that the present study used 118 EFL Taiwanese young learners (i.e., eighth graders) in middle school as participants and indicated a significantly positive effect of captioning on spoken vocabulary acquisition. Such positive finding is consistent with Markham’s (1999) research, which indicates the marked effect of captioning on listening word recognition of advanced, university-level ESL students. Based on the results of the study, Markham suggests that rather than ignoring the auditory input when captioning was available, the advanced ESL young adult learners seemed to be able to transfer the visually-presented textual input to subsequent listening-only tasks. In other words, to the 118 ESL students, captioning may serve as an aid for gains of second language.
In accordance with the results of Markham’s study, this current study extends the value of captioned video on spoken vocabulary acquisition to young learners in the EFL educational context. But simultaneously, this study also yielded clear evidence that although all the participants benefited from the presence of captioning, the amount of benefit varied with learner’ linguistic competence levels. In other words, this study demonstrated that by adding the visual text to video material, the EFL young learners’
recognition of spoken vocabulary tended to make significant improvement; those of high competence, however, seemed to be the largest beneficiary of dual input (auditory and visual) in terms of listening word gains of foreign language. In sum, based on previous caption-based research, this study established the validity of captioning on EFL learners’ spoken vocabulary acquisition, but at the same time pointed out the significant influence of the learner’s linguistic level on gains of foreign language from video viewing.
Captioning and Vocabulary Acquisition
58
The second research question aims to investigate vocabulary acquisition at a deeper level; that is, whether the availability of captioning could enhance EFL middle school students’ performance on form and meaning associations. The results suggested that the participants under the captioned condition scored significantly better than those under the non-captioned condition. In other words, the availability of captioning tended to facilitate connections of phonological word form and meaning to a significant extent.
In addition to the quantitative data, the positive effects of captioning on
vocabulary acquisition were also revealed from the qualitative data. According to the participants’ feedback collected from the follow-up question session, with the presence of captioning, most participants, both high and low achievers, reported that they felt relaxed during the viewing sessions and seemed to be able to have better
comprehension or make better inferences from video context clues. Additionally, some of the high achievers further indicated that the visual text helped to make the authentic speech and some unclear articulation easier to follow. These comments suggest that by adding captioning to video material with high audio-visual correlation, language in the video tends to become more accessible to EFL young learners at different levels, which in turn, may reduce the burden on them to process the language input, and finally makes the mapping of aural form and meaning easier.
Moreover, it should also be noted that this favorable outcome was generated from a two-time repeated exposure to each of the chosen video episodes without explicit instruction. Such positive finding seems to reveal the advisable number of times of video viewing for improved gains of foreign language through incidental learning.
Webb’s (2008) study examines the context effect on L2 incidental vocabulary acquisition and suggests that L2 word knowledge can be acquired incidentally if learners meet the word in context enough times. The number of times required for word gains, however, depends on the contexts in which words are met. According to the results of Webb’s research, there are two qualities of context crucial for incidental vocabulary learning—one is appropriate text coverage, and the other is rich contextual clues. In other words, it could be expected that the target words encountered repeatedly in the difficulty-appropriate text which provides the participants with sufficient
information to infer the meaning of a word are going to be acquired better than those that are met in difficult and less informative text. This study provides evidence that when FL words are encountered in the informative context (i.e., video with high
59
audio-visual correlation) with appropriate difficulty (at least 90% of known words), two times of exposure to the video material is adequate for the occurrence of incidental vocabulary acquisition of EFL middle school students.
Taken together, the use of multiple input modalities (imagery, auditory, and textual) appears to not only improve recognition of the surface forms of language but also facilitate increased depth of processing in terms of vocabulary acquisition (i.e., form and meaning mapping). Nevertheless, as mentioned earlier, in spite of the consistent benefits derived from captioning on both vocabulary tests, the mean scores on overall form recognition and vocabulary acquisition are different in an unexpected manner.
The underlying reasons are provided in the following subsection.
The Unexpected Differences between the Scores of the Two Types of Vocabulary Tests
At first, the participants were expected to score higher on form recognition tests than on form-meaning mapping tests since form recognition, the initial step of
vocabulary learning, is often regarded as word knowledge at superficial level while the connection of word form and meaning requires relatively deeper cognitive processing.
This study, however, yielded a contrary outcome under both treatment conditions. The results of mean scores indicated that the students tended to obtain lower scores on form recognition than form-meaning mapping under either captioned (M = 3.733 for form recognition; M = 4.233 for form-meaning mapping) or non-captioned conditions (M = 3.017 for form recognition; M = 3.707 for form-meaning mapping). These unexpected differences could be interpreted in light of the design of the two vocabulary tasks and students’ perception of them.
In fact, the fabric of the two vocabulary tests was identical. First, both tests were in the multiple-choice test format with a total of seven items in a test. Additionally, each test item was composed of four options for the participants to choose from (i.e., one key and three distractors). However, to measure word gains at different levels, the design of the two types of tests was not exactly the same. The first vocabulary task, aural form recognition, was a listening-only multiple-choice test. In each test item, the participants had to aurally identify the target word that appeared in the video previously after hearing four phonologically-similar words. On the other hand, the second vocabulary task, form-meaning mapping, was not a listening-only test. In each test item, while the
60
participants had to listen to four words in the first test, in the second one, the participants only listened to one target word spoken and then tried to identify its meaning from the four visually-presented options in L1.
From a theoretical perspective, Buck (2001) indicates that listening tasks which require processing more information tend to be more difficult than those requiring processing less information. By comparing the number of words presented aurally in one item of form recognition test (N = 4) and vocabulary acquisition test (N = 1) in this current study, the information required processing aurally in form recognition tests was four times the information required processing aurally in vocabulary acquisition tests.
In addition, results of Wu’s (1998) study demonstrated that for Chinese EFL learners at university level, viewing the printed options in multiple-choice listening task enabled the learners to anticipate the incoming input and equipped them with the focus for listening. Moreover, his study also suggested that the visually-presented options might allow large amount of uninformed guessing, which sometimes even resulted in
learners’ selection of the correct answer for the wrong reason. Accordingly, the vocabulary acquisition test adopted in this current study, which allowed the
participants to read the printed options on the test paper before the target word spoken out, was assumed to be less difficult than the form recognition test.
Qualitative data collected from the follow-up question session also indicated that the participants noticed the aforementioned difference and perceived that the two types of tests represented different difficulty-level. Specifically, for many students, the form recognition test was a little demanding for them. For example, some of them reported that they felt anxious when taking form recognition tests because they were not familiar with such listening-only test design. Additionally, some indicated that listening to four options made them nervous because it was not easy for them to memorize the
pronunciation of all the four words in one item. Moreover, it seemed like a great challenge for some participants to distinguish the target word from the other
phonologically-similar distractors. In contrast, the test of form-meaning mapping was relatively easy for them since in each item, they only had to listen carefully to one spoken target word and then try to select the correct answer from the four printed Chinese translations.
Based on the information demonstrated in previous studies and revealed from the researcher’s observation and participants’ feedback, it appears that the unanticipated
61
outcome of scores on the two vocabulary tests may be likely resulted from the different test design to some extent. Nevertheless, the results of this study indicated that the effects of captioning remained consistently significant on both vocabulary tests.
Therefore, although the difficulty-level of the two vocabulary tests in this study was not perfectly balanced, this study still revealed that the presence of captioning could profoundly improve EFL learners’ gains of both form recognition and vocabulary acquisition.
Linguistic Competence and Word Gains from Video Viewing
To determine the effects of captioned video materials on spoken word gains of learners at different levels of linguistic competence, this section is divided into (1) the caption effects between groups, and (2) the caption effects within groups.
Between–group Effects
Concerning the third research question, the discussion could be subdivided into two subsections—(1) whether the effect of captioning on spoken vocabulary
acquisition was influenced by learners’ linguistic competence level, and (2) whether the word gains of overall form recognition and vocabulary acquisition were modulated by the level of linguistic competence. First, the results indicated that the effects of
captioning did not depend on EFL young learners’ linguistic competence according to the insignificant interaction between caption condition and linguistic competence demonstrated by two-way ANOVA. In other words, captioning exerted a facilitative effect on all participants’ vocabulary acquisition regardless of their linguistic levels.
To address the second sub-question, the participants’ performance on overall form recognition and vocabulary acquisition was analyzed solely on linguistic competence variable. The results suggested that the participants who differed in linguistic
competence level benefited from captioning to different extents in overall form recognition and overall vocabulary acquisition. Specifically, in both vocabulary tests, the students who were of high linguistic competence tended to outperform those at the intermediate level; those at the intermediate level of linguistic competence tended to outscore those at the low linguistic competence level. Such finding is in line with Lwo and Lin’s (2012) study and suggests that captioning which used to be considered as optimal listening support for low-level learners actually benefits high-level learners
62
more. In Neuman and Koskinen’s terms, vocabulary acquisition through incidental learning tends to follow the rule of Matthew Effect—“rich get richer.”
This result could be explained by the requirement for incidental vocabulary acquisition from context (i.e., massive exposure to comprehensible input). With regard to vocabulary learning, inference skill is often considered as a crucial part in word gains from context. Nagy (1995) reported that there are three categories of knowledge that contribute to contextual inferencing—(1) linguistic knowledge such as syntactic and vocabulary knowledge, (2) world knowledge, and (3) strategic knowledge. As for spoken text, Buck (2001) also indicated that the subskills involved in inferring word meaning from spoken text consist of lexical knowledge, parsing skills, and
inference-making ability. Accordingly, when video materials are employed for
language teaching and learning, learners with better linguistic competence in the target language are more likely to benefit from them owing to great vocabulary knowledge and sophisticated strategy use, the abilities they originally possess. By contrast, leaners with limited linguistic competence may be less effective at using video context for word growth, until they achieve a higher level of linguistic competence of the target language.
In sum, this study yielded the results which were in contrast to the belief of most EFL teachers with regard to the usefulness of captioning for students at different linguistic levels and demonstrated that learners’ linguistic competence appears to be a crucial factor affecting the efficacy of using authentic captioned video material for L2 acquisition. From a pedagogical point of view, this implies that video-based instruction which includes incidental vocabulary learning as one of the learning objectives must provide learners at lower level of linguistic competence with other learning support for improved learning outcomes.
Within-group Effects
The discussion in this section is based on the results of paired samples t-test. By comparing the means in the first test, as the baseline, with the average scores in the follow-up tests (from test 2 to test 8), the quantitative results in Tables 12 and 13 demonstrated that over the follow-up seven-week viewing sessions, the availability of captioning tended to be more beneficial to high-level EFL middle school students in fostering the ability of acquiring phonological vocabulary knowledge. In addition, the
63
results of the first and follow-up tests comparison suggested that low-level EFL students seemed to improve slightly after receiving the weekly viewing sessions under the non-captioned condition.
These results revealed that after undergoing the weekly video-viewing learning sessions without explicit vocabulary instruction for eight weeks, the high-level EFL middle school students improved in terms of spoken word gains through watching videos with captioning, while the low-level students made slight improvement in vocabulary acquisition through watching videos without captioning. Such outcomes, once again, were in opposite to the traditional assumption, which regarded captioning as a listening support which was more beneficial for students of low linguistic
competence while some high-level learners might consider it as an impediment to information processing. In sum, based on the significant main effect of captioning and the results revealed from paired samples t-test mentioned above, this study indicated a strong demand for more learning support for lower-level learners in multimedia learning environment.
Qualitative Results
In the previous section, discussion about the results presented in Chapter Four was reported. However, there are still some inherent noteworthy features that could not be
In the previous section, discussion about the results presented in Chapter Four was reported. However, there are still some inherent noteworthy features that could not be