major findings are presented in section 5.1. The impact of the results on practical

CHAPTER FIVE CONCLUSION

This chapter is comprised of three major parts. The summary of results and

major findings are presented in section 5.1. The impact of the results on practical

teaching environment is presented in section 5.2. Finally, the limitations of the study

and suggestions for future research are discussed in section 5.3.

5.1 Conclusion

The study was conducted to investigate the effects of different modes of glosses

on incidental vocabulary learning, especially the differences between animation+ text

glosses and graphics+ text glosses. The subjects were all beginning EFL learners

studying in junior high school in Taiwan. The subjects read the reading passages via

the Internet in the computer lab at school.

The two research questions are: (1) Is the effectiveness of graphics+ text glosses

and animation+ text glosses different from that of text-only glosses on incidental

vocabulary learning? (2) Do graphics cues and animation cues differ in their

effectiveness on incidental vocabulary learning?

Below are the major findings of the study.

1. Regardless of the gloss types, students’ vocabulary mean scores all improved

significantly immediately after the treatment. This means that all the gloss types

(graphics+ text, animation+ text, and text-only glosses) used in this study help

incidental vocabulary learning.

2. As to the immediate posttest mean scores, the overall test results show the

superiority of graphics group over text-only group. The production test results show

the superiority of graphics and animation groups over text-only group. This means

that glosses with visual cues (animation or pictures) better help incidental vocabulary

learning than those without. Besides, there is an advantage of graphics+ text glosses

over text-only glosses, while the differences between graphics+ text and animation+

text glosses didn’t reach significance. This indicates that graphics cues exercise their

power in a way different from animation cues. This finding is consistent with

Grabowski and Zhu(2006). They found that participants in the static graphics group

performed equally well as those in the animation strategies groups. In either way, the

finding provides one example supporting Paivio’s (1986) dual coding theory.

3. As for vocabulary gain, the results are kind of complex. Though the means scores

of animation group were the highest among the three, animation and graphics groups

both performed significantly better than text-only group only on the part of production

test. As for the overall vocabulary gain, no group differences were found. That is,

animation+ text glosses didn’t show any advantage over the other two gloss types, or

even showed inferiority to graphics+ text glosses.

4. In this study, all the recognition tests are not sensitive to different gloss types. On

the other hand, participants in the three groups performed significantly different on

the production tests.

5. From the results of delayed posttest, we found that the vocabulary retention is

low and the group differences disappeared compared to the results of immediate

posttest. This means that students forgot the target words very quickly. Two weeks

after the treatment, we can’t see the different effects of different gloss types.

5.2 Pedagogical Implications

The present study aimed to explore the effects of animation+ text glosses and

graphics+ text glosses on incidental vocabulary learning. The results of this study

offer several implications for vocabulary teaching and learning.

First, in this study, students using glosses all learned new vocabulary items (in

average 26.11% of new words) and the differences between pretest and immediate

posttest all reached significance. This suggests that junior high school students can

pick up new words while reading as long as relevant glosses are provided. Besides, if

students want to improve their receptive knowledge of vocabulary words, L1

definition can serve the purpose as well as text+ graphics or even text+ animation

glosses.

Second, if the purpose of vocabulary learning is productive knowledge, the

results in this study suggest that textual glosses with visual cues can better help

students learn than text-only glosses. Therefore, if visual cues are available, it is

worthwhile to put visual cues beside textual information. The study shows that

glosses are better remembered when they are dually-coded. Both graphics and

animation can help students to develop more concrete ideas about what the words

mean, help learners to build the form- meaning connection and thus to process the

vocabulary items deeper.

Third, the overall vocabulary immediate posttest results reveal an advantage of

graphics+ text glosses over animation+ text or text-only glosses. This suggests

graphics can better help learners map the form-meaning connection and facilitate

incidental vocabulary learning than animation or texts do. Teachers and courseware

designers don’t need to spend much time or money on the making of animation when

the purpose is incidental vocabulary learning because graphics can do a very good job

already.

5.3 Limitations of the Present Study and Suggestions for Future Research

The study has several limitations that readers or other researchers should pay

attention to. I will address the issues by category; namely, the participants, the word

type of target words, the way of assessment and the materials.

The first limitation concerns the small number of participants. Because of the

capacity of the computer lab, no more than 30 students can take the program at the

same time. Moreover, some students were absent on one day or two because of other

activities. At first, 90 students participated in the experiment, but only 77 students

completed all the tests (pretest, immediate posttest, and delayed posttest). If possible,

future studies can recruit a larger number of students, or can collect data from

different schools. In this way, the result could be more representative, and could be

more readily applied to realistic teaching environment.

Second, the target words in this study included only verbs. Thus, the

generalization of the results to other word types is limited. Since the graphics+ text

glosses reveal an advantage over text-only glosses as for the immediate posttest

results, it is worthwhile for future studies to further investigate whether the incidental

learning of other word types, such as nouns, adjectives and adverbs, also benefits

from graphics cues more than animation cues.

The third limitation of the study concerns the format of the assessment. In this

study, the production test, which included the first and final letter of a word, seemed

to become clues for students to guess the answers of recognition test. It is suggested

that more than four alternatives should be provided in the recognition test in order to

decrease the possibility of guessing. Besides, more distracting words could be added

in the tests to prevent students from getting alert to the target words after the pretest.

Finally, we should still interpret the results with caution because of the difficulty

of the materials. Though the two passages used were chosen carefully to fit students’

language competence, the length of the materials was a little longer than what they

were used to reading and there were still some new vocabulary items other than the

target words. Therefore, the two passages are a little more difficult for about one third

of the participants. The decision is made because, in this way, there would be enough

relevant target words to be observed. In future studies, the researcher can try to make

the materials easier and fit Nation’s (2001) criteria for efficient incidental vocabulary

learning that there are fewer than 5% of new words, or at best, 2% of new words in a

passage.