|
|
|
||||||||||||
|
||||||||||||
|
||||||||||||
|
||||||||||||
|
Integrating Computerized Kanji Learning
Materials in Classroom: Setsue Shibata, Ph.D., California State University, Fullerton Introduction The Japanese language is considered to be one of the most difficult languages for Western people to learn. Kanji, in particular, is one of the most challenging parts in the process of learning Japanese for those who have alphabetic-language background (Takebe, 1989, p.7; Tollini, 1994, p.107). There are more than 50,000 kanji characters; 3,000 characters are used regularly. Students are overwhelmed from the beginning not only by the number of kanji which they must master, but also by the complicated phonological and semantic systems and the large number of vocabulary words using each kanji characters. A survey conducted by Shibata (1999) revealed that one of the major reasons for the high attrition rate from lower to higher level Japanese classes is a negative impression and attitude toward learning kanji. However, since Japanese script is a combination of three types of writing systems, i.e., hiragana, katakana and kanji, kanji is essential to mastering reading and writing skills in Japanese and cannot be avoided. The Japanese Language Proficiency Exam (Nihongo Nooryoku Shiken), which was developed by the Japan Foundation to assess proficiency levels of Japanese as a second language, requires about 100 kanji characters for level 4, 300 for level 3, 700 for level 2 and 1500 for level 1, where level 4 is the Novice and level 1 is the Advanced level (the Japan Foundation, 1997). As for the teachers, teaching kanji is tremendously time-consuming within the limited time frame of classroom instruction. If computerized kanji learning materials are as effective as the traditional workbook-and-pencil instruction, then the instructors can use the classroom period for more interactive activities rather than dealing with kanji writing practice. It is a challenge to teach kanji effectively without discouraging students' interest toward kanji, all the while using classroom time more efficiently. Using computers to teach kanji is becoming very popular among Japanese teachers and many computerized kanji learning materials have been developed. Many previous studies have reported that computer-assisted instruction can substantially improve student achievement when compared with traditional instruction (Avent, 1993; Bhatia, A. 1991; Garrett, R.L., 1995; Hong, 1997; Johnson, M. & Osguthorpe, R., 1986; Nieves, 1994; Stenson et al, 1992). Few studies have been conducted on the Japanese language, especially with regard to learning kanji. Van Aacken (1996) studied the efficacy of computer use for learning kanji using a cohort design. She also looked at the results for high-level and low-level groups. She found no significant difference in scores between traditional instructional groups and computer-use groups, but found that the low-level group in computer-use groups enjoyed using the computer for learning kanji the most, and had the highest increase in scores over the course of the class. Additional study is needed using different approaches to explore the possibility of learning kanji with computers. In this study, three groups were compared instead of two: a group with a traditional kanji workbook, a group with computerized kanji leaning materials (CKLM) and a group with both workbook and CKLM focusing on immediate effect of computer-use. There are three purposes in this study: 1) Is there any difference among the three groups in the immediate recall of kanji recognition tasks? 2) Is there any difference among the three groups in the immediate recall of kanji production tasks? 3) Is there any difference among the three groups in the change of attitudes toward learning kanji? METHOD Subjects Voluntary participation was requested in the first semester Introductory Japanese class at a public university in California. Any student who had previous experience in learning Japanese or Chinese was excluded. Forty-two students agreed to participate, and were randomly assigned to one of the three groups: A group that used traditional workbook-and-pencil methods to study kanji (Group 1) and the two computer-use groups, i.e., a group that used CKLM only (Group 2) and a group that studied kanji using both workbook and CKLM (Group 3). The distribution of students' native languages was: twelve English, one each of Tagalog and Arabic respectively in Group 1; eleven English, one each of Tagalog, Spanish and Portuguese respectively in Group 2; and thirteen English, and one Indonesian in Group 3. As for gender, there were ten male and four female students in Group 1; six male and eight female students in Group 2; and six male and eight female students in Group 3. Students had no previous experience in learning kanji, but had been learning Japanese for about two months when the study was began. Selection of Kanji Characters The ten kanji characters to be learned in this study were selected from a vocabulary list of verbs and nouns describing daily activities, which the students were learning at the time of the study. Therefore, the students were familiar with the vocabulary and understood the meaning of the sentences when the kanji characters were written in hiragana or read aloud.(2) Kanji characters which were comprised of simple strokes and/or characters whose meaning could easily guessed from their shape were avoided. Computerized Kanji Learning Materials (CKLM) The CKLM used in this study were prepared by the author using the template that JPNet of Massachusetts Institute of Technology developed.(3) The CKLM contained the same information as the workbook, i.e., the shape of the kanji character, its meaning, reading, number of strokes, and mnemonic information, and the vocabularies which use the kanji character. Additional feature in CKLM is a video clip of calligraphy of the character, which shows the process of writing with a traditional Japanese pen (fude) and ink (sumi).(4) It appears when the viewer clicks the character in a box. Kanji Workbook A traditional kanji workbook, which accompanied the textbook, was used for Group 1 and Group 3. The workbook contained the same information as CKLM, with the exception of the video clips of calligraphy of the kanji characters. Survey on Attitudes toward Learning Kanji A questionnaire asking about feelings/attitudes toward learning kanji was developed by the author to measure the students' feelings and attitudes toward learning kanji. The survey was administered to all groups twice, i.e., before and after the studying of kanji. The survey is five-scaled, from scale 1, in which the students disagree with the statement most strongly, to scale 5, which represents the agreement to the statement most strongly. In other words, scale 1 indicates that the students have the most negative feelings/attitudes toward each statement about kanji, and scale 5 the most positive. There are four statements: "Memorizing kanji characters is not difficult," "Writing kanji characters is not difficult," "I do not feel overwhelmed by the number of kanji characters that I have to learn by the end of the second year," and " I am interested in learning kanji." Kanji Tests Two kanji tests were administered, i.e., a kanji recognition test and a kanji production test. On the kanji recognition test, students were asked to write the meaning of each given kanji character in English. On the kanji production test, students were asked to write each kanji character accurately in a blank square that matched its meaning in English. The kanji production test was administered only after the kanji recognition test was collected. Each test has 10 questions, and the maximum score is 100. For the kanji recognition test, the evaluation is either correct or incorrect, and each correct answer received 10 points. For the kanji production test, each accurate production received full 10 points, and one that had a stroke extra or missing received 5 points. Each kanji production was evaluated by two Japanese instructors to obtain the highest reliability. Procedure The subjects were divided into three groups randomly and were asked to fill out the questionnaire regarding their feelings/attitudes toward learning kanji. Then each group was randomly assigned as Group 1 (workbook-and-pencil method), Group 2 (CKLM-only) and Group 3 (workbook and CKLM). All three groups studied the same kanji characters for an equal length of time (totally 70 minutes). Group 1 received traditional kanji instruction using a workbook, i.e., the reading and meaning of the character, and an example of usage in a sentence. The instructor demonstrated the writing of each character on a blackboard. After this 30-minute introduction by the instructor, students studied and practiced writing kanji characters individually using a workbook for 40 minutes.(5) The survey on attitudes and the kanji tests were administered the following day. The kanji production test was administered after the kanji recognition test was collected. Each test was administered for 20 minutes. Each student in Group 2 was provided a computer in the Computer Lab and received an orientation on how to use CKLM for 15 minutes. After that, students studied kanji characters using CKLM for 70 minutes on their own. A lab assistant was available whenever the students had computer problems. The following day, the survey and the kanji tests were administered in the same manner as Group 1. Group 3 received the same 30-minute introduction of the kanji characters by the same instructor as Group 1 using the workbook. They then moved to Computer Lab and received CKLM orientation for 15 minutes, and then conducted self-study of kanji for 40 minutes for themselves. The survey and the kanji tests were administered the following day as they were to Group 1 and Group 2. RESULTS Performance on Kanji Tests The performance of each group on the kanji tests after the instruction are reported in Table 1. The mean scores on the kanji recognition test and kanji production test were analyzed and compared among groups by one-way analysis of variance. The Tukey method was applied when a significant difference was found. Table 1: Means and Standard Deviation for Recognition and Reproduction Tests by Group, With Results of ANOVA F tests, Tukey Post Hoc Testing
Significant Tukey pairs: a=Group 1 and 2, b=Group 1 and 3, c=Group 2 and 3 There were no statistically significant differences among the three groups on the kanji recognition test (F= 2.98; p > .05). Of a possible score of 100, Group 1 had a mean score of 69.29, Group 2 had a mean score of 67.86, and Group 3 had a mean score of 77.86. For the kanji production test, it was found that there was a significant difference among groups (F= 6.96; p < .01). Group 1 had a mean score of 42.86, Group 2 had a mean score of 40.71 and Group 3 had a mean score of 56.43. Tukey's post hoc test indicated that the mean score of Group 3 was significantly higher than the mean scores of Group 1 and Group 2. There was no significant difference between Group 1 and Group 2. In other words, the group with both traditional instruction and CKLM performed significantly higher than the other two groups, who had a workbook-only or CKLM-only. Attitudes toward Learning Kanji Analysis of variance was used to determine if there was a significant difference between the groups on their mean scores from the survey of attitudes toward learning kanji before and after studying kanji. There was no significant difference between groups on any of four statements in attitudes survey in the pre-test. Mean scores, standard deviation and ANOVA results are shown in Table 2. Table 2: Means and Standard Deviation for Pre-test on Attitude-Scales by Group, With Results of ANOVA F Tests, Tukey Post Hoc Testing
Pre-test showed that all three groups had similar attitudes toward learning kanji before studying: the scales on all four statements were somewhere between 3 and 2, i.e., the students tended to believe that learning kanji "seems to be rather difficult." The results of analysis of variance show that there were significant differences between groups on three of four statements in the post-attitude survey as shown in Table 3. Table 3: Means and Standard Deviation for Post-test on Attitude-Scales by Group, with Results of ANOVA F Tests, Tukey Post Hoc Testing
On the first statement, "Memorizing kanji is not difficult," a significant difference in mean scores was found (F= 5.866, p < .01). Tukey's method revealed that there was a significant difference between Group 3 and Group 1, and between Group 3 and Group 2, i.e., the group used both workbook and CKLM had significantly higher mean than the other two groups on the attitude toward memorizing kanji. On the second statement, "Writing kanji is not difficult," a significant difference was found between groups (F= 5.828; p < .01). Tukey's method indicates that there was a significant difference between Group 3 and Group 1, i.e., the group that studied kanji using both workbook and CKLM had higher means on their attitudes toward writing Kanji than the workbook-only group. On the third statement, "I don't feel overwhelmed by the number of kanji characters to learn," no significant difference was found between groups (F= 0.636; p > .05). On the forth statement, "I am interested in learning Kanji," a significant difference was found between groups (F= 3.510; p < .05). Tukey's method indicates that Group 3 had a significant higher mean score than Group 1, i.e., the group used both workbook and CKLM had a significantly higher mean than the group used only workbook. Table 4: Means and Paired T-tests on Attitude-Scales
Paired samples t-tests show that all three groups scored higher on their post-test than pre-test on Memorizing; t-tests were 3.309 (p< .01) for Group 1, 2.386 (p < .05) for Group 2, and 8.33 (p < .01) for Group 3. They also showed that there was no significant difference between pre-test and post-test on writing kanji in Group 1 (t=1.794, p > .05), but the post-tests were significantly higher than pre-tests for Group 2 and Group 3 (t=6.624, p < .01 for Group 2; t=7.071, p < .01 for Group 3). A significant difference between pre-test and post-test was found in all three groups on the attitudes toward number of kanji, i.e., all three groups obtained higher means on post-test than pre-test (t=2.511, p < .05 for Group 1; t= 5.090, p < .01 for Group 2; t= 6.511, p < .01 for Group 3). The results also indicate that all three groups scored higher on post-test than pre-test on their interest(t= 2.223, p < .05 for Group 1; t= 6.450, p < .01 for Group 2; t= 7.389, p < .01 for Group 3). DISCUSSION Performance on Kanji Tests The study indicates that there is no significant difference in kanji recognition tasks based on the type of instruction to be used; workbook, CKLM, or both. However, the study revealed that the group with both traditional instruction and CKLM achieved the highest on the kanji production test. It may be explained that while kanji recognition tasks require only a partial memory of the character, production tasks require more attention to accuracy of the character during memorization. Production of kanji requires students to write a specific number of particular strokes in a precise sequence. In Ke's study (1996), students performed better on kanji recognition tasks than kanji production tasks. As many kanji educators point out, kanji is memorized by writing, i.e., the ability of production of kanji is associated with graphomotor memory. Video clips of calligraphy in CKLM provides visual information regarding how to write a kanji character, but unless the students try to write it by hand, they do not really "master" the character. In this regard, the students who studied kanji with CKLM-only spent most of their time watching the monitor and had a shorter amount of time practicing writing than the other two groups. Visual effect of video clips of calligraphy in CKLM may be equally as effective as the workbook in helping to memorize the shape of the character, but kanji production proficiency cannot be acquired satisfactory without practicing writing repeatedly. On the other hand, workbook-and-pencil requires students to practice writing by hand. As the results show, instruction integrating computer with traditional workbook (i.e., combination of visual effects and hand-writing practice) seems to be more effective for kanji production tasks than instruction of either single method-only. In other words, as many previous studies indicate, simply replacing computers with traditional instruction, i.e., instruction of kanji with blackboard and workbook, cannot be expected to improve students' learning outcome of kanji production. Students seem to learn more effectively when the computer is integrated in classroom to be used as a supplemental learning tool. Jesky and Berry (1991) pointed out that more realistic cues are more effective to memory. Computerized video clips of calligraphy in CKLM provide an authentic image of writing, and it helps students to memorize the visual image of writing. It also shows students the process of writing, i.e., not only the sequence of strokes but also each stroke's feature, such as whether the end of the line should be "tomeru" (intentional stop on the position) or "nagasu" (stop the motion gradually), etc. Attitudes toward Learning Kanji This study supports many previous studies that found that using computers in language learning improves and promotes the students' motivation and interest in learning the subject (Davis & Lyman-Hager, 1997; Garrett, 1995; Johnston, 1987; Kubota, 1995; Markel, 1994; Miech et al., 1996; Van Acken, 1996). The results of this study also indicate that there is no clear relationship between the students' attitudes toward using a computer and their achievement while using it. The students in the CKLM-only group enjoyed using computers, and their attitudes toward learning kanji improved after using them. However, their level of performance did not exceed the students who did not use a computer. The group that used both workbook and CKLM improved their attitudes toward learning kanji the most, and they also performed best on the kanji recognition and production tests. Limitation of the Study and Suggestions for Further Research There are several limitations in this study. First, the number of subjects was small, and therefore other variables, such as students' gender and learning style could not be considered. Secondly, since this study focused on the first-year Japanese learners with no previous experience in Japanese or Chinese, further studies should be conducted focusing on Intermediate or Advanced kanji learners. Thirdly, this study only focused on recognition and production tasks, further research is needed with more in-depth pedagogical considerations of kanji, such as "production of kanji with accurate okurigana," "writing furigana on a Kanji character in a sentence," "translating the sentence which includes the kanji character", etc. Conclusion In conclusion, CKLM helps students' learning, promotes interest and motivation to learn, and assists instructors in using time more efficiently, especially when it is used as supplementary materials in class. The traditional workbook is still an excellent learning materials for the practice of writing, and to encourage graphomotor memorization, integrating CKLM in class seems to enhance the students immediate recall on both kanji recognition and reproduction outcomes. The real value of CKLM lies in its use by the instructor. Notes (1) This study was funded by Scholarly and Creative Activity Grant, Office of Grant at the California State University, Fullerton. I am grateful to Mrs. Masako Verdi and Ms. Marie Uzawa for their assistance. (2) For example, students understood the meaning of the sentences such as "asagohan o tabemasu." (3) The author is very grateful to Ms. Ann Lavin of JPNet, MIT for her technical supports and to Mrs. Tomoko Graham, former Japanese lecturer at MIT, for her valuable suggestions. (4) The original video clips of calligraphy were created by Prof. Saeko Komori of Chubu University in Japan. The author thanks to Prof. Komori for her generous idea to share these clips. The original video clips are shown in her homepage, http://www.intl.chubu.ac.jp/komori/ (5) From the author's experience, it takes at least 3 minutes for the brief introduction of each kanji character, i.e., its reading, meaning, and writing. The number of ten characters was considered to be reasonable to introduce in 30 minutes. References Avent, J. (1993). A study of language learning achievement differences between students using the traditional language laboratory and students using computer-assisted language learning courseware. Ph.D. diss., University of Georgia. Bhatia, A. (1991). Kanji retrieval by recursive location of elements using Hypercard. CALICO Journal 9(2), 4-25. Davis, J. N. & Lyman-Hager, M. A. (1997). Computers and L2 reading: student performance, student attitudes. Foreign Language Annals, 30, 58-72. Garrett, R. L. (1995). computer-assisted instruction in 2-year colleges: Technology for innovative teaching. Community College Journal of Research and Practice 19, 529-536. Hong, Wei. (1997). Multimedia computer-assisted reading in business chinese. Foreign Language Annals, 30, 335-344. The Japan Foundation. (1997). The Japanese Language Proficiency Test in the United States. [Brochure]. Santa Monica, CA: Author. Jesky, R. R. & Berry, L. H. (1991). The effects of pictorial complexity and cognitive style on visual recall memory. [EDRS: ED 334 987.] Johnson, M. V. & Osguthorpe, R. T. (1986). Computer-based foreign language instruction: Improving student attitudes. Educational Technology, 26, 24-8. Johnston, M. V. (1987). Attitudes toward microcomputers in learning. Educational Research, 29, 47-55. Ke, C. (1996). An empirical study on the relationship between Chinese character recognition and production. The Modern Language Journal, 80, 340-349. Kubota, R. (1999). Word processing and WWW projects in a college Japanese language class. Foreign Language Annals, 32, 205-217. Markel, M. (1994). Behaviors, attitudes, and outcomes: A study of word processing and writing quality among experienced word-processing students. Computers and Composition, 11, 49-58. Miech, E. J., Nave, B., and Mosteller, F. (1996). On call: A review of computer-assisted language learning in U.S. colleges and universities. [EDRS: ED 394 525.] Nieves, K. (1994). The development of a technology-based class in beginning Spanish: Experiences with using EXITO. Ph.D. diss. George Mason University. Shibata, S. (1999). [Survey on factors of high attrition rate from lower to higher division Japanese courses among college students]. Unpublished raw data. Stenson, N., Downing, B., Smith, J. & Smith, K. (1992). The effectiveness of computer assisted pronunciation training. CALICO Journal (Summer): 5-19. Takebe, Y. (1989). Kanji no Oshiekata. Tokyo: Alc. Tollini, A. (1994). The importance of form in the teaching of Kanji. Sekai no nihongo Kyooiku, 4, 107-116. Van Aacken, S. (1996). The efficacy of CALL in Kanji learning. On-Call, 10(2), 2-14. Appendix Appendix A: Computerized Kanji Learning Materials (CKLM) These materials can be observed at: http://www-japan.mit/fullerton Appendix B: Survey on Attitudes toward Learning Kanji For each of the following statement, circle the number which presents the most your current feeling/attitude toward learning kanji. 1. Memorizing kanji characters seems to be not difficult. (disagree) 1 2 3 4 5 (agree) 2. Writing kanji characters seems to be not difficult. (disagree) 1 2 3 4 5 (agree) 3. You are supposed to learn about 300 kanji characters by the end of the forth semester of Japanese. Do you feel overwhelmed by the number? (disagree) 1 2 3 4 5 (agree) 4. I am interested in learning kanji characters. (disagree) 1 2 3 4 5 (agree)
BIOGRAPHICAL NOTES Setsue Shibata received her Ph.D. from University of South Carolina, Columbia. She is currently an assistant professor of Japanese in the Department of Foreign Languages and Literatures at California State University, Fullerton. She specializes in Japanese pedagogy, second language acquisition and the issues of loss and maintenance of a heritage language. Address for correspondence: Setsue Shibata, Ph.D. e-mail: sshibata@fullerton.edu
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Interpersonal Computing and Technology: An Electronic Journal for the 21st Century © 2002 The Association for Educational Communications and Technology. Copyright of individual articles in this publication is retained by the individual authors. Copyright of the compilation as a whole is held by AECT. We ask that any re-publication of this article state that the article was first published in IPCT-J. Contributions to IPCT-J can be submitted by electronic mail in APA style to: |