The present paper describes various aspects of the work involved and results of step - by - step thesis instruction performed via computer network and by hand. Two computer networked courses were required to select topics, prepare an outline and bibliography, write a first draft, receive content editing and language editing comments and implement them into their reports via network. Compared with the work turned in on paper by the control courses, the outlines tended to be of a higher quality, but the resulting papers tended to be shorter. Many time delays due to host computer malfunction occurred, which may have contributed to the lack of length of the papers. In spite of these delays, both groups completed their tasks in the same time frame, due to the fact that the final step of reinputting language errors was not necessary for the networked courses.
The orthodox pattern of composition for most university courses is that of requiring report after report, each graded separately and provided with comments which rarely give the student any direct clues as to how to improve their basic skills in researching and organizing ideas.
This is especially true in the case of students in Japan, who rarely receive formal instruction in prewriting, writing and postwriting processes in English courses. This lack of instruction shows up especially when students participate in study abroad programs at foreign universities, which usually assume such basic writing skills as prerequisite. Moreover, it has been found (Santos, 1988 p. 85) that papers of foreign students are usually judged more harshly in terms of their content and structure rather than their correctness in terms of English, with the conclusion that "nonnative-speaking students need to improve their skills in the areas that most directly affect content, such as organizing, developing and supporting their ideas and argument." This would seem to emphasize the necessity of greater concentration on presentation and organization skills (and less concentration on grammar and sentence-level composition) than is presently the case in many English composition courses in Japan.
2. Process writing vs. contrastive rhetoric in the present study
Santos continues by suggesting that a process approach would best suit the needs of these students. The process approach to writing (Zamel, 1982) is a prevalent TESL composition theory which places emphasis not on the product itself, but rather on imparting the processes involved in achieving the product in order to improve results. Specifically, it advocates instruction in processes related to pre-writing such as brainstorming, listing, and organizing, as well as those related to post-writing and the actual writing process itself.
Indeed, the ideals of this theory would be greatly suited for teaching academic-style composition in Japan, but for two difficulties: l) process writing is designed more for freestyle composition rather than academic treatises, and therefore provides no guidelines for teaching either research and note-taking skills or formal structure, and 2) the outline, which would seem to be the most valuable tool for Japanese students to learn how to structure academic style reports, is downplayed in process writing. In fact, outlines are actually rejected in favor of imitation of the actual writing styles of skilled native speakers, who tend to begin writing without formalizing their thoughts, and come up with their structure cyclically (through a series of brainstorming-organizing-writing cycles) rather than linearly. As Zamel (1982, p. 197) puts it, "requiring students to formulate their ideas beforehand, to elaborate upon them by using some prescribed rhetorical framework and to submit these written products for grading purposes seems to ignore every-thing we have learned about the process." Because of the two difficulties mentioned above, a process theory type pattern of instruction was not used in this course program. On the other hand, the traditional composition pattern described in the introduction does not seem to do the trick for Japanese students either . Then, what type of instruction program would best fit the needs of these students?
Contrastive rhetoric studies (Kaplan, 1966, 1983, Leki 1991) have suggested that different cultures show different rhetorical styles. Aspects of the Japanese composition style in particular have been documented by Hinds. For example, regarding the difference in emphasis on unity, he writes (1987, p. 146):
“For English readers, unity is important because readers expect and require landmarks along the way.... In Japanese, on the other hand, the landmarks may be absent or attenuated since it is the reader's responsibility to determine the relationship between any one part of the essay and the essay as a whole."
In view of such differences in English and Japanese composition style and structure, “ESL teachers have a responsibility to teach the expectations of the English audience to L2 writers ... our students' texts will be easier for their professors to read if the writers show the kind of audience awareness that comes from knowing what the rhetorical expectations of the readers are." (Leki, p. 138)
The processes used in the course program described below were designed with this "responsibility to teach the expectations of the English audience to L2 writers" in mind. Although they are not directly related to those in vogue in the process theory, some are likewise intended to bridge gaps in students' prewriting task abilities, especially with respect to judging appropriate topics for research papers. On the other hand, more traditional methods are employed to teach students how to make use of library resources efficiently, and to write with the clear and explicit structure which will be expected by most American instructors.
3. Motivation for computer networking
In a fanciful introduction to his bibliography of computer-related works, Susser (1987) relates a story of Jiro Yamada, a Japanese university student who does all of the processes involved in composing a report for his composition class using a computer atop his kotatsu. His professor receives, corrects, and resends the report fully by computer as well. Although a fantasy, as pointed out by Susser, this story could be reality even today using existing technology. And as it may well become a primary modus operandi of education in the future, introduction of students to technology of this nature can only be beneficial in the long run.
On a realtime scientific level, a wide-scale meta-analysis of examination results reported in 99 comparative studies of college-level CBE (computer based-education courses) and conventional ones was performed by Kulik and Kulik (1986). There, the following results were described (p. 88):
"In 77 of the 99 studies, the students in the CBE class had the higher examination average ... The difference in examination performance of CBE and control students was reported to be significant in 22 studies. In 21 of the 22 cases, the significant difference favored the CBE class, whereas only one study favored conventional teaching. Overall, these box-score results favor CBE."
The results were found to vary depending on the course content, in that positive effects of using computer-based instruction was significantly higher in humanity-based CBE courses than in scientific ones. Based on these results, it was hoped that the implementation of a computer-based style of teaching English composition would produce some positive effects in comparison with non-CBE courses.
Coinciding neatly with the planning stages of the course program to be described was the establishment of the MEROS university network, using a UNIX-based host computer, at Kumamoto University of Commerce in 1991. All first year students were required to buy personal computers and were taught the basics of networking as well as word processing in their first semester. This provided a group of students with a working background in the technological side of the experiment, so that valuable class time would not be required for computer-related explanations or demonstrations.
Furthermore, development of an NEC computer version of “Family Feud” by the author (see Yoneoka 1993, in Japanese) allowed for the use of “Family Feud" as a basis for selection of research paper topics. This technique has been successful both as a vehicle for imparting cultural differences between the US and Japan, and for focusing students' concentration on specific differences providing themes appropriate for research. This is because the use of a particular question as a springboard was found to encourage the students to brainstorm about the ways in which cultural differences were reflected through its answers, and to actively wonder about the reasons behind those differences (corresponding to asking oneself “why" about a topic, as in process-writing style invention exercises).
4. The project
Students in 5 university courses (2 freshman general English, I freshman English conversation, 2 sophomore English expression courses) and 2 junior-college freshmen English conversation courses were required to produce one academic style paper over the course of one year. (Keep in mind that the main content of the courses was NOT writing but conversation-oriented). Composition techniques and guidelines for producing the paper were taught in sections as shown in Figure 1, and each will be discussed separately below.
The 2 freshman general English courses (hereafter referred to as CBE courses) were expected to use computer networking exclusively for the completion of their projects, whereas the control classes did all of their work by hand, with the exception of the final draft of the report, for which word processing was required. There was a total of 130 students in the general English courses, as compared with an approximate total of 150 students in the other 5 courses combined. Although the overall number of students in the two groups was similar, the relative course sizes (65 in the 1st group compared with between 15-50 in the second) is a factor which could have influenced the results and should be kept in mind.
The purpose of the project was to investigate the following questions:
What effects would the requirement of working through a computer network have on English composition education and on the quantity/quality of the reports?
What are the advantages/disadvantages of using a computer network for English composition?
4.1. Theme/Topic selection.
According to Raimes (1991), a process-centered approach requires students to choose their own topics, as opposed to a traditional form-oriented approach in which topics are assigned by the teacher. In the present project, a compromise between the two approaches was used. A total of 100 questions from the game show “Family Feud" (see Yoneoka 1993, in Japanese) were introduced to the students via computer game. These served to suggest specific Japanese, US or contrastive culture-oriented topics which could be easily recognized and researched. After one of the questions was selected by a student (in a first-come, first-serve manner) it was examined in terms of the cultural differences and/or similarities it represented. Often, one question could lead to several different topics. For example, the question “What is the first thing you would get for a trip to Europe?" (responses: 1. passport, 2. tickets, 3. luggage, 4, camera, 5. shots) prompted essays on topics as varied as comparative immigration procedures, travel expenses in various countries, dangers involved in immunizations, and even the gun control issue in America, due to a misunderstanding of the word “shot."
4.2. Research and outlining.
Next, a sample outline and bibliography was distributed and students were instructed to complete an outline of their own based on their selected themes. Students were given minimal instruction on acceptable types of references, where to find them and how to look for them, as well as pointers on format of the bibliography. The outlines were graded by section using a 10-point scale as follows:
Introduction - 2 points
Body - 4 points = 10 points
Conclusion - 2 points
Bibliography - 2 points
Students were requested to include 5 references in order to receive the full 2-point credit for bibliography.
The outline and bibliography were considered to reflect the result of work involved in the prewriting stages. Since this stage is often lacking in Japanese composition instruction (and also the most important in terms of learning to write in an academic-style format) it was required that the students “pass" this stage before moving on to the actual writing of the paper. To do this, they had to receive a total of 8 or more of the 10 points possible; otherwise, they were required to rewrite and resubmit their work.
One problem common to many students observed was the difficulty of finding "information to fit the theme," pointing to the probability that they considered the outline (content) to come first in a linear sense, and the research to come second. This shows a general naiviete towards references, i, e. that any kind of information can be found easily. In fact, several students included no bibliography at all in their outline, and when requested to rewrite their outlines to include the bibliography, were not easily able to revise their original thoughts to correlate with the data available.
4. 3. Drafting.
Another problem which seems to be endemic among Japanese students is the lack of confidence regarding writing directly in English. No matter how often told not to write a Japanese draft first and then translate it, many students just cannot help putting their thoughts down first in Japanese. This may lead to direct transfer of Japanese writing style, which comparative rhetoricists such as Kaplan have found to differ from the linear, logic-oriented style usually used in English.
To make Japanese drafting as unattractive a choice as possible, students in the computer courses were advised to do their drafting directly “0n top of" their outlines, which were already in their word processors; constructing paragraphs directly by “expanding" each item to include an introductory and concluding sentence, as well as supporting their point in the body of the paragraph. This provided the students with a more or less mechanical method for developing their work in English, so that they did not have to resort to their native language to come up with a structure.
4. 4. Content revision.
After first drafts were submitted either by E-mail (CBE courses) or by hand (control courses), they were reviewed for content and organization. This consisted not only of addition/deletion of information and structural corrections, but also of suggestions for refinement of the flow of ideas in the paper. Also, at this stage many students had still not yet come up with appropriate titles for their work, and thus were instructed to do so. For the computer based courses, all of these comments were written IN JAPANESE and or set off by carats (> >) in order to clearly distinguish the comments from the text (as would be done for example by using a red pen on paper). In the control courses, comments were also written in Japanese as much as possible for purposes of comparison, but there were times when English was used as well.
Upon receiving these comments, students were required to revise their work in accordance with the comments and resubmit it. If they received their projects back with further comments, this meant that they had not yet “passed" the content revision stage and needed more work. If not, the papers were then (and only then) subjected to a language revision.
4. 5. English revision.
Originally, it was planned that all students would have a hand in English revision through the use of available spelling/grammar checkers. However this turned out to be impossible, due to the fact that there was no such software at the university which could be used with the students' personal computers. It would have been possible to transfer students' files to either a Macintosh format or MS-DOS format for NEC to make use various spelling/ grammar checkers available for those machines, but lack of time (due to the fact that the English revision draft due date was extended from 11/30 to 12/31 because of a series of computer breakdowns from 10127 to 12/2) made this plan unfeasible.
Therefore, all English-related corrections were done by the instructor directly on the text submitted by the students. For the CBE courses, this yielded the final work (unless there were parts which could still not be understood or which were ambiguous and needed clarification), so there was no need to submit a third draft. On the other hand, for the control courses, a third draft (at this point word-processed) was required. Due to this fact, the two sets of courses completed their work within the same general time frame, in spite of the deadline setbacks due to computer breakdown for the network-based classes.
5. Comparison of composing via computer network vs. paper
5.1. Time related observations.
5. 1. 1. Computer literacy effect on meeting deadlines.
Of the two CBE classes, class 1-4 tended to lag far behind class 1-7 in general computer capability, both according to their data processing course instructors and as judged subjectively by the author. This led to a remarkable difference in ability to meet deadlines through the duration of the course. For example, as can be seen in Fig. 2, only 56% of class 1-4 succeeded in meeting the 5/31 deadline for topic selection, compared with 80% of class 1-7. The latter class was able to meet the deadline almost as well as most of the control courses (80% -90%), and indeed excelled the control classes in choosing topics AHEAD of the deadline (22% of the class chose their topics a week or more before 5/31 compared with 6% for the “fastest" control class). As the topics were to be “reserved" on a first-come, first-serve basis, this points to the fact that for the computer-1iterate class, reservation by computer was more easily accessible than finding the teacher and reserving outside of class time for the control classes.
5. 1. 2. Computer failure and deadline extension.
In addition to failure to meet deadlines as reported above, mechanical failure compounded the problem for computer-non-1iterate students. Several times during the course, the network host computer suffered breakdowns, resulting in forced extensions of deadlines for various steps in the composition process. Figure 3 below shows the dates of host computer breakdown and the various extensions implemented.
In addition, according to a show of hands at the end of the course, over 1/3 of the students' personal computers had trouble at some time during the year due to either mechanical or human operator error. These difficulties both with the host computer and with their own personal computers led to consider-able frustration on the part of many of the students and resulted in even greater time delays in many cases.
5. 1. 3. Instructor time requirements.
Although no exact measurements were taken of instructor time required to process student work via computer vs. by hand, the following subjective impressions can be mentioned.
1) As the system upon which the Meros network is based does not allow for continuous downloading or uploading of separate files, each file had to be downloaded and uploaded one by one by hand. Although the time required for each file was less than a minute for each operation, this added up to several hours of time for the more than 100 students participating in the course.
2) Due to computer difficulties, there was much time lost both in and out of class both due to unsuccessful computer operation and for explanations and consultations regarding those problems.
3) On the other hand, the instructor time involved in correcting English and making comments (especially in Japanese) on students' papers by hand was longer as compared with using the word processor. It was found that in one hour, approximately 8 papers could be corrected on computer, as compared with an average of 6 papers by hand.
4) The fact the CBE classes did not need to produce a third revision (due to the fact that English corrections were made directly into the text file) led to a well-received savings of time for both the instructor and the students at the end of the school year.
5.2. Educational-oriented observations
5. 2. 1. Quality of Evaluations
In the present program, none of the students' work was graded except for the preliminary outline, but this showed a marked difference in scores between the CBE and control courses. Specifically, students who returned outlines with computers showed a higher passing rate than those who did their outlines by hand (see fig. 4).
The percentage of students passing (gaining 8 points or better on) their first outline was 44.83% for class 1-7 and 33.34% for class 1-4 vs. 24.51% for the control classes. Part of the difference between the two CBE classes may be accounted for by the relative computer illiteracy of class 1-4 (e.g. 16.67% of the classes' outlines were not correctly received due to student error in
computer uploading, as compared to 8.62% for class 1-7) . On the other hand, the relatively high passing rate of the CBE classes as compared with control classes (in spite of computer error) may be attributed to two factors: 1) the computer failure in May led to an extension of the deadline, giving the students more time to prepare, and 2) the necessity of immediate word processing led the students to take the assignment more seriously, and give it more thorough consideration. It is also possible that the visual quality of word-processed outlines as compared to handwritten ones could have led the evaluator to grade the former more leniently, although this was consciously taken into consideration at that time.
Two beneficial side results for the CBE courses were that students learned ways of utilizing their school network in courses other than data processing, and were able to develop skills in English word processing.
One of the purposes of this project was to note the effects that the requirement of word processing from the beginning of the course would have on the length of papers and on general English ability. Toward this end, the number of words used in a representative sample (N = 15) of reports from both computer courses was taken and compared with the same number of reports from control courses, which were word processed at the final stage only. The results were as follows:
As can be clearly seen, the control groups' papers were an average of more than 30 words longer than the more computer literate class, and almost 100 words longer than the less literate one. The difference between the computer courses and control courses may be due to the fact that the lack of word processing skills discouraged the students from developing their thoughts as thoroughly as they would have done by hand. It is also likely that the CBE course students may have followed their outlines more closely (having them already in the computer) than their counterparts, so that they did not add in as much additional information. A further possibility is that the relative lack of time available to the CBE courses due to deadline setbacks related to host computer failure led the students to write less.
In addition, the difference between the two computer groups also points to the fact that the more computer-wise class (which presumably had better word processing skills) was able to produce longer works with the computer.
Perhaps the biggest disadvantage of using the computer in the present project from an educational point of view was that the students didn't have the chance to internalize the corrections of their English, either by reinputting and rewriting incorporating corrections (as was necessary with the control classes) or by using spelling/grammar checkers. As previously mentioned, this was partially due to the fact that computer breakdown leading to deadline extensions left no time at the end of the course.
5.3. Communication/data management and retrieval.
E-mail provided fast and easy communication between students who had questions either on the course material or on computer operation. Approximately 60 pieces of mail were received during the year. The following are typical of the contents of such mail:
(NO matter how often I try, I can't upload my report from my house. The number of retries keeps going up, but the byte number stays at O. What should I do? I don't think I'm doing anything wrong...)
(I'm sorry; all of the economics dictionaries in our club room are checked out because we're preparing for an economics conference on 11/27-28. That's why at this point I have only finished the introduction to my English essay. I'll upload it, but the rest may have to wait until after the economics conference).
Also several letters were recieved in English especially during the first semester, but unfortunately were lost as a result of host computer hard disk crashes.
The automatic time/date data provided by the computer provided a great help in managing, especially during the topic selection portion of the course. In the control classes, there were times when students would leave undated notes stating their preferred topic, and there was no way to settle disputes as to who was “first" other than by lottery.
In theory, courses using the computer should have had no need to ever submit anything on paper, which would have led to a decrease in time spent in organizing, arranging and shuffling pieces of work. Also, taking work home would have been made more convenient since in the author's case, this means competing with three small children (with crayons) and a dog for possession of any stray piece of paper. In practice however, the extended periods of host computer breakdown near the end of the course necessitated the submittal of reports in paper and/or floppy disk form.
In spite of poor conditions for comparison between the CBE and control classes related to unplanned deadline extensions for the former, a few tentative conclusions can be made from this project as follows: 1) A positive effect on quality of outlines was noted with the CBE classes, especially with the more computer-wise class; 2) on the other hand, a negative effect was noted on average word length of the final essays from the CBE classes. This effect was pronounced within the two CBE classes as well, with the more computer-literate class producing essays of over 50 words longer than the less-capable class.
As for time-oriented differences observed, the frequent host computer breakdowns led to a great increase in instructor time required for the CBE classes. This included all time spent in communicating with the computer center, relaying information and answering student questions in and out of class, as well as unsuccessful entry attempts via modem. It is assumed that this was true for the students as well as the instructor. In fact, it may well be hypothesized that frustration related to the many delays contributed to the negative effect observed on word length in the CBE classes as well.
AFTERWORD: CONSIDERATIONS FOR COMPUTER BASED PROJECTS
The relative failure of the present project to proceed as planned is directly related to the high computer breakdown rate; both for the host computer and for individual computers. As this was the first year of full network operation as well as the first year all students purchased their particular brand of personal computer, this type of trouble was probably unavoidable, but should have been foreseen.
In this sense, the present stage of the multimedia era may be compared to the early years of the automobile industry. If something went wrong with a Model T, the driver had no other choice but to sit and wait, or walk. Now however, there are busses and taxis, and Automobile Clubs. Support for the automobile industry has evolved to the point that glitches do not inconvenience the user in any major way.
This will certainly become more and more true of computer technology as well. There will probably come a day when networks will be able to provide a range of data processing services. Companies providing backup computer services will become routine. However, that day has not come yet, and the will few decades of relatively unsupported Model T multimedia technology will continue to produce frustrations. This must be taken into consideration when planning a project of this sort.
The circumstances which presented the unique opportunity for this type of study (availability of university network and personal computers for all students) will obviously not be duplicable at all universities, although the trend will most certainly be towards this in the future.
Hinds, John (1976). A Taxonomy of Japanese discourse types. Linguistics, 184, 45-54
Hinds, John. (1980) Japanese expository prose, Papers in linguistics 13, 117-158
Hinds. John (1984). Retention of Information using a Japanese Style of Presentation, Studies in Language 8. 1, p. 45-69.
Hinds, John (1987). Reader vs. writer responsibility: A new typology. In U. Kulik, Chen-Lin C and James A. Kulik (1986) Effectiveness of Computer-based Education in Colleges, AEDS Journal, No. 19, p. 81-108.
Leki, Ilona (1991). Twenty-five Years of Contrastive Rhetoric: Text Analysis and Writing Pedagogies. TESOL Quarterly, 25(1), p. 123-143.
Raimes, Ann (1991). Out of the Woods: Emerging Traditions in the Teaching of Writing, TESOL Quarterly 25 (3), p. 407-430.
Santos. Terry (1988). Professors' Reactions to the Academic Writing of Nonnative-Speaking Students, TESOL Quarterly 22(1), p. 69-90.
Susser, Bernard (1987). Computers and composihon : A B, bliograplic introduchon The Language Teacher, 11 (5), 10-19.
Yoneoka, Judy (1993). Teaching American Culture through English Computer Software (in Japanese), Kumamoto Shodai Ronshu, p. 135-151.
Zamel, V. (1982). Writing: The process of discovering meaning. TESOL Quarterly, 16(Z), 195-209.
APPENDIX A. SAMPLE OUTLINE (ORIGINAL)
Name something in the house that usually takes at least two people to move. -Refrigerator is #2 answer in USA.
You will find the refrigerator everywhere in our home ,but what will it have the history of the refrigerator?
The refrigerator had been into practical use since 1913 in America.
The refrigerator stand on an electric and a gas one.
The refrigerator is our life cannot misse an object.
>>No bibliography! Needs more research.
APPENDIX B. SAMPLE OUTLINE (REVISED) INTRODUCTION
Name something in the house that usually takes at least two people to move. -Refrigerator is #2 answer in USA.
B. You will find the refrigerator everywhere in our home, but what will it have the history of the refrigerator?
The Refrigerator cool food and it is apparatus that can keep food.
At first refrigerator was ice one and Most old one in Japan is used by in 720 A. D that Nintoku emperor discover a ice cave on lzumi' s mountain. But there are electric and gas refrigerator at present.
The electric refrigerator had been into practical use since 1913 in America, 1930 in Japan.
The electric refrigerator's principle is use air heat when liquid become air's condition. Generally it is used by furon gas.
The gas electric's principle will explain. It heated cold ammonia water by gas burnner first of all, and separate ammonia or water. It use air heat when ammonia become air's condition.
A. The refrigerator is our life cannot misse an object.
Nippon Daihyakkazensyo , Syougakukan, Aiga Testuo, p. 323-326.
Study OBUNSHA'S JAPANESE - ENGLISH DICTIONARY, Takahashi Gengi, 1985,0bunsha.
> >Good! Needs more in conclusion, though.
APPENDIX C. SAMPLE REPORT (CONTENT-REVISED)
Name something in the house that ususally takes at least two people to move.- Refrigerator is #2 answer in USA. The Refrigerator cool food and it is apparatus that can keep food.
You will find the refrigerator everywhere in our home, but what will it have the history of the refrigerator?
At first refrigerater was ice one and Most old one in Japan is used by in 720 A.D that Nintoku emperor discover a ice cave on Izumi's mountain > > 出典 //source//. But there are electric and gas refrigertor at present.
The electric refrigerator had been into practical use since 1913 in America, 1930 in Japan. >>出典 //source// The electric refrigerator's principle is use air heat when liquid become air's condition .??? // I don't understand! // Generally it is used by furon gas.
The gas electric's principle will explain. It heated cold ammonia water by gas burnner first of all, and separate ammonia or water. It use air heat when ammonia become air's condition.
The refrigerator uses even in refrigerator cars and ships. We eat fresh food, the refrigerator is our life cannot misse an object.
Study OBUNSHA'S JAPANESE - ENGLISH DICTIONARY, Takahashi Gengi , 1985,0bunsha.
APPENDIX D. SAMPLE REPORT (FINAL VERSION. ENGLISH -CORRECTED)
The History and Operation of the Refrigerator
My question was "Name something in the house that usually takes at least two people to move" to which Refrigerator was the #2 answer in the USA. Nowadays you will find the refrigerator in almost every home. The refrigerator not only cools food but it is indispensible for preserving food as well. What is the history of the refrigerator? And how does this invention work?
The first refrigerator was the "ice box", and it was literally that: a box with ice in it. The oldest one in Japan was used in 720 A.D, when emperor Nintoku discovered an ice cave on Izumi mountain (Nippon Daihyakka Zensyo). It was in 1877 that an artifical ice box was made in Japan for the first time , in Osaka.
But there are electric and gas refrigerators at present_ The electric refrigerator come into practical use in 1913 in America, and in 1930 in Japan. (Nippon Daihyakka Zensyo)
The electric refrigerator works on the principle of using the air heat produced when liquid changes into air. Generally it uses freon gas. This principle is as follows: Cold ammonia and water is heated by a gas burner first of all, and then separated into ammonia and water. It uses air heat produced when ammonia vaporizes (Nippon Daihyakkazen-syo). New electronic refrigerators are being developed now.
The refrigerator is used not only in the home, but everywhere; even in refrigerator cars and on ships. Because we often eat fresh food, the refrigerator is indispensible in our daily lives.