2020 Volume 4 Pages 1-20
Japanese language-oriented English sentence structure (JLOS) is a kind of English sentence structure that Japanese students often employ when thinking in Japanese rather than English. JLOS is hypothesized to have the following elements: illogical, not collocated, not harmonious, and incorrect preposition. The research questions are (1) what type of weak points does Japanese students’ writing present? and (2) does the multivariate analysis model with online learning facilitate in helping students’ writing? The sample size is 58 university students in the Kanto area. The data were collected in 2017 and 2018. POSA was utilized to evaluate whether the structure was significant in Japanese English writing (JLOS). A 3-step model was proposed to improve students’ JLOS writing.
The main aim of this paper is to analyze and categorize typical mistakes in students’ English writing. In order to improve Japanese students’ English writing skills, this case study further recommends online technologies that can be used at home and in the classroom to facilitate positive self-correction. The study primarily explores the following themes:
In the second term of 2017, a business course was run for first-year students at T University. In this course, students assumed the fictitious role of trainees at a virtual company called Widgets. In the last term of the course, students had to undergo a job interview, as they wished to be promoted from trainees to full-time workers. In preparation for the interview, they were instructed to write legitimate answers to mock interview questions (model answers were also provided to them). For example, students were required to answer the following questions:
However, grammatical errors or inappropriate phrases appeared on their papers, and it became clear that each student was struggling with the common errors. Thus, a brief survey was conducted in order to explore the students’ English educational background; it found that 97% of the students had never written an English composition before entering university. In addition, 100% of the students think in Japanese, subsequently carrying out direct translations when constructing an English sentence. Furthermore, when confronted with difficult expressions, they consult a Japanese-English dictionary or an online translator.
In relation to this result, the MEXT (Ministry of Education, Culture, Sports, Science, and Technology) declared that students’ writing skills were deemed to be lacking in training based on a 2014 investigation into the command of English at higher education institutes. Having evaluated those students’ errors and mistakes, it was hypothesized that their ways of forming sentences in English may be wrong in some way. For this reason, they may need to learn additional techniques or develop skills in order to self-correct their writing. It can be argued that the method of direct translation and the usage of Japanese- English dictionaries may negatively affect students’ writing. As a result, it is proposed that more effective ways ought to be used to enhance their writing, such as online applications. The above sentences will be used to represent Japanese language-oriented English sentence structure (JLOS), which is a kind of English sentence structure that Japanese students often employ when basing their thinking on the Japanese language rather than the English language. The fundamental ideas of JLOS are actually derived from a cultural study of my MA thesis (Kimura 2003); it is emphasized that the definition of JLOS is first assumed by this study. Therefore, JLOS literature cannot be found elsewhere. The question of how to prevent or reduce such use of JLOS will be discussed in later sections.
In the field of TESOL, there appears to have been a considerable amount of research on the teaching of writing skills. However, not many researchers have studied the practice of self-correction and its relationship to writing. The following researchers conducted studies on teacher or peer error correction from different perspectives, angles, methods, and conceptions. Based on these studies, they are inclined to claim that correction practices are important as such. Many researchers possess an understanding of the elements that are needed for good writing (Makino, 1993; Vickers & Ene, 2006; Gilmore, 2009; Quinn, 2015; Chacon-Beltran, 2017). Brooks (2008), for instance, argues that good writing must include one’s own ideas and feelings, and effectively convey their messages or main points to the reader. Writing is a form of communication, and successful communication in any language requires accurate use of words and expressions. Such communication also positively contributes to the forging of different cultures and people.
2.2 Effective WritingRegarding essay writing, Taoka (2016) analyzed essay topics in the Criterion Online Writing Evaluation service, an online writing tool developed by the Educational Testing Service, by categorizing them into personal, socio-personal, or social categories, and identifying what writing skills are required for each. From there, he gathered that:
It may be implied that elements of effective writing may be related to Western linearity, sagacity, and a logical way of thinking in English. The teacher must make students aware of this problem and show them how to face it in a rational manner.
2.3 Online SupportThis is a strong caveat that online applications can save the errors in JLOS. In addition, self-correction may be successfully encouraged by online learning; specifically, online learning has become an essential teaching tool. Many researchers have studied this trend and argue that there are various advantages to using online learning tools in class.
Trinder (2017) argues that online informal learning of English deserves more attention, and presents an empirical study surveying the practices and preferences related to new media in independent settings of Austrian university students. Results indicate a clear preference for well-established, time-honored media (film, online dictionaries, email) in self-regulated contexts, as well as diverging opinions on the use of technologies in classroom settings.
Large corpora such as the British National Corpus and the COBUILD Corpus and Collocations Sampler are now accessible (Gilmore, 2008) free of charge, and online learning can be usefully incorporated into a process writing approach to help develop students’ writing skills.
Stapleton and Radia’s study (2009) argues that L2 writing pedagogy needs to give more recognition to the impact of new technological tools such as online dictionaries and online resources. It is suggested here that developments in software and online resources are leading to improvements in many aspects of student writing, in both language and content. Moreover, efficient use of this technology could have a significant effect on the way in which teachers provide constructive feedback.
To help Japanese university EFL learners sustain their vocabulary learning, Tanaka, Yonesaku and Ueno (2016) have developed Lexinote, an e-portfolio system that allows learners to record and save the target words they encounter online; to search for them in online dictionaries; to practice them in several ways, including written and oral rehearsals according to word familiarity; and to share their own output with peers. Learners are guided to monitor and control their vocabulary learning according to word familiarity. It is discussed how Lexinote can benefit learners, instructors, and CALL researchers in terms of depth of words. It would be highly desirable to use online support, as it is one of the current trends in EFL education, making it possible to interactively check the right words and correct the arrangement of words.
2.4 Self-CorrectionIt is here proposed that JLOS through online learning leads to the development of self-correction skills. As previously mentioned, it would be very useful if students could use online technology to perform self-correction of their writing. Not many researchers have conducted studies on these types of self-correction practices—here is one example of self-correction implementation.
Lawley’s paper (2015) describes the development of web-based software at a university in Spain to help students of EFL self-correct their free-form writing. The software makes use of an 80-million-word corpus of English known to be correct as a normative corpus for error correction purposes. A cohort of students who volunteered to trial the software for a month reported positively about their experience, as it had helped them identify mistakes in their own and in their peers’ writing and had greatly accelerated the self-correction process.
One of the advantages of self-correction is that of conquering the issue of JLOS. More specifically, by not translating from Japanese into English, students may write a more proper text in terms of grammar, vocabulary and rationality, and that more accurately reflects a native speaker’s writing. In addition, being able to write without depending on others for support may be considered ideal by students.
2.5 Factors in the Weakness of Japanese Students’ WritingInstead of planning, proposing and thinking in English, L2 learners tend to translate, evaluate and revise their writing by using their own native language (Rostamian, Fazilatfar & Jabbari, 2017), which may be considered natural behavior. However, dealing with this cognitive process might limit their writing activities within the framework of their mother-tongue. Due to translation, or other cultural or linguistic structural influences, the English of native Japanese speakers is described as illogical (Oi, 1986; Nishigaki, Chujo, McGoldrick & Hasegawa, 2007; Ono & Morimura, 2007; Honna, 2012; Sutherland, 2012; Honna & Takeshita, 2014) and incongruous (Gerhand & Deregowski, 1995; Toyota, 1996; Kitayama & Ishii, 2010; Miura, 2011). In order to correct English grammatical errors of Japanese speakers, especially those relating to prepositions, Nagata, Vilenis and Whittaker (2014) conducted a study using self-feedback software. Hanazaki and Hanazaki (2015) believe that the English proficiency level of Japanese college students has deteriorated from the viewpoint of preposition. Ono and Morimura (2007) especially point to students’ wrong use of prepositions. In order to improve this deficiency, their paper argues that e-learning such as eAlps, which utilizes Moodle, can be useful in terms of presenting a variety of prepositions. Further, it argues that most students are aware of advantages of this type of learning. In addition, it has been noted that the acquisition of collocations is one of the difficulties that Japanese English learners encounter (Haruno, Ikehara & Yamazaki, 1996; Koya, 2003; Yamashita & Jiang, 2010). Kita, Kato, Omoto, and Yano (1994) used corpora in teaching collocational knowledge and found that students were able to acquire some collocations effectively.
The argument in 2.5 naturally leads to the definition of JLOS.
According to the survey discussed in the introduction, whenever the students composed written English sentences, all of them thought in Japanese first, subsequently carrying out direct translations. Furthermore, JLOS may be produced when they cannot formulate proper English sentences, consult a Japanese-English dictionary, and combine the definitions they find with their own direct translations. This mixture of direct translations and Japanese-English dictionary definitions creates a foundation for JLOS; in fact, JLOS is a form of sentence-leveled English, rather than a word-leveled one.
These sentences are illustrated in Table 1, along with a few examples of several students’ writing. Italics in Table 1 show typical errors (JLOS). It is emphasized that more students’ writing examples will be shown in 7.3, 7.4, 7.5 and 7.6.
| Element | Definition | Students’ writing examples |
| 1) illogical | They lack sense, clarity, or sound reasoning; are not suitable for the situation; are unusual in a way that is difficult to understand; and require an additional word in order to realize effective comprehension. | a) Second, I’m cleaner at imitating others. b) I want to be accountant, and father likewise. |
| 2) not collocated | Some words are not collocated with other words or as a single phrase. | a) I guess how to arouse customer’s appetite. b) I love to work at bank heartfully. |
| 3) discordant | It means that they are not in harmony or keeping with the surroundings in the context; and have a mixture of informal and formal vocabulary. | a) But I haven’t realized those ideas because I haven’t known how to materialize those ideas. b) I wanna good people’s relationship. |
| 4) incorrect preposition | Students choose a wrong preposition as there are so many different uses of prepositions which Japanese students cannot identify the right one. | a) Because I think I am suited with examining smallest details and, integrate the results of survey. b) Because I am good in mathematics. |
It is believed that these facets play a very important role in conquering the issue of JLOS. Therefore, in this chapter, the importance of those four variables will be examined using (i) facet theory; (ii) mapping sentence, which is a set of hypotheses; and (iii) POSA (Partial Order Scalogram Analysis), which is its associated MDS.
Table 2 shows the details about the participants and research methods used in this study.
| Sample | n=58, first year students in T University in Kanto region |
| Date | April 2017 to February 2018 |
| Survey | Learners’ writing script of presentation |
| Statistics | A multivariate analysis of POSA and χ2 test |
| Main question on data analysis | Why is JLOS language created by Japanese English learners? |
In order to understand the structure of the variables relating to the students’ writing errors, the facet theory was used, as it is considered the most reliable instrument (Kimura, Manabe, Yasunaga & Yokota, 2002) to evaluate the causal model underlying the factors related to the students’ writing errors. POSA, in facet theory, primarily uses a binary scale, as it is unambiguous and it clarifies a scientific definition of the hypotheses made by the researcher. A native speaker volunteered with the researcher to judge the students’ English writing. The binary scale was adopted upon agreement between the native speaker and the researcher. It is stressed here that the whole context of the students’ writing was examined, but a part of their mistakes was extracted to highlight JLOS.
4.3 POSAPartial Order Scalogram Analysis (POSA) was invented by Louis Guttman (1972) and developed by Shye (1978) and Shye and Elizur (1994). The main feature of POSA is to demonstrate the distributive disperse of JLOS. In other words, it shows the developmental process of how students deal with the variables related to the JLOS. POSA is an extension of the one-dimensional Guttman scale and is a multi-dimensional cumulative scale. There is no uniformly used stress measure.
4.4 Mapping SentenceA mapping sentence is an a priori hypothetical resume of the research themes. Four facets are shown in Table 3; these were deductively chosen, as they seemed to give us clues about the Japanese students’ written errors from the viewpoints of the previous studies discussed above.
| Facet | Content of the Questions | Answer Form |
| A | Do I tend to translate Japanese sentences into English when writing the English task and do they appear to be illogical? | yes or no |
| B | Those sentences don’t sound like native English. Do I often use wrong collocations? |
yes or no |
| C | Does some of my writing look and sound incongruous in its context? | yes or no |
| D | Does the choice of incorrect prepositions affect my writing? | yes or no |
The mapping sentence for why JLOS language can appear on learners’ writing is summarized in Table 4.
The writing manuscripts of 58 out of 61 students were used as data for this research. The data were collected from April 2017 to February 2018 during English lessons at T University in the Kanto region of Japan. Each student’s sentence data were chosen from their writing tasks. Then, the data were categorized and analyzed by the author, as well as by another lecturer who is a psychologist working at the university. The participants were all freshmen and the level of their English was a mixture of lower intermediate to upper intermediate. Under the same conditions and same theme, they were asked to answer the job-related questions:
The students were allowed to consult a dictionary or use Google Translate if necessary. The participants understood and agreed that their writing was used for research.
The following shows that how data spread on the diamond shape of POSA.
Figure 1. Diamond scalogram portraying the tendency of JLOS users
It is noted that a diamond-shaped POSA such as that in Figure 1 is the most regular configuration in the analysis; the data show a distributive disperse of JLOS. Moreover, the reproducibility is 95%, and considered to be acceptable (Canter 2004). There are 16 possible combinations (2×2×2×2) in this POSA. Referring to Figure 1, the numbers in brackets shown in the diamond indicate the frequency of a profile. In a similar manner, the vertical numbers on the right-hand side indicate the sum frequencies of the profiles of the same joint score, which is totaled from left to right. The downward direction (Joint Direction) is determined by totaling the digit values of the profiles, as shown on the lefthand side of the diamond. This Joint Direction shows the degree of mistakes in writing. Therefore, when a score increases it means that the use of JLOS strengthens. It ranges from 0 (the most use of proper English) to 4 (the most use of JLOS). The horizontal order (see Lateral Score) is determined by interpreting the aligned joint score levels. The lateral score is a psychological scale implying that the left part of the lateral direction shows dependency from the teacher, and the right part indicates the effect of independency from the teacher. For instance, the lateral score 1 of profile 1000 (20) are the people who tend to write illogical sentences. The illogicality of their writing is one of the most fundamental mistakes in English thinking and is also the most difficult to self-correct; therefore, these people are more likely to be dependent on the teacher’s help. On the other hand, the lateralscore 7 of profile 0001 (4) belongs to those who only make mistakes in using prepositions. This type of mistake is easy to tap, and can be easily self-corrected by consulting a dictionary. These students tend to be, to some extent, independent from the teacher’s advice. The profiles connected by the downward lines are one-dimensional Guttman scales. We now turn to look at the main findings of the survey. The most popular progress of the POSA is 1111(1) → 1110 (1) →1100 (7) →1000 (20) →0000 (6). The strength of POSA is to show the most popular developmental path connecting the profiles in the most frequencies by the arrow. That path is 1111(1) → 1110 (1) →1100 (7) →1000 (20) →0000 (6). This consists of a Guttman scale with cumulative progression; it implies how to improve the errors in JLOS and how to improve the skills involved in English writing for Japanese students. Those profiles are neatly arranged along with the right oblique region of the POSA format by connecting with each other. Joint score 1—having a mistake in JLOS—is found to be the majority (40 people). It is neither due to the sample size nor it is a characteristic of JLOS. It demonstrates that Japanese students are likely to only make one instead of multiple mistakes when writing in English. This may tentatively suggest that Japanese students’ prospects are bright.
We are now familiar with the structure of Japanese writing errors in terms of POSA, as well as the most popular developmental path which the students go through in JLOS. By examining the POSA patterns, a student may learn a path to self-amend errors. In doing so, a model for self-correction is suggested to make the process easier and more fruitful. The following model-chart suggests how those JLOS can be corrected by the students themselves.
Figure 2. A model for self-correcting procedure in amending JLOS sentences.
The result indicates that the sentences of 29 students were illogical (48%), and that this was the most common mistake among all students. My idea, based on my experience as a language teacher, is that these illogical sentences are quite difficult to correct, as many of them are produced by translating directly from the students’ mother-tongue. The actual meaning of these sentences may be hidden within the Japanese frame of mind, and therefore cannot often be accurately and effectively conveyed to English readers. Moreover, illogical sentences may be formed due to a lack of grammatical knowledge; as a result of this, it is often clear that the word order is not correct, or that those sentences may need additional words to make sense in English. All in all, the result implies that the root of JLOS errors may derive from the nature of Japanese language. This section will suggest self-correction before students submit an assignment to their teacher. In fact, according to the students in my classes, most of them tend to go through the following process when constructing English sentences: they first try to think of, or begin drafting, sentences in simple English or in Japanese followed by an English translation; then, whenever they come across a problem, such as not finding a suitable word or phrase, they use a Japanese-English dictionary or Google Translate. However, it should be emphasized that they must develop a correct and more complex procedure. Detailed below is my recommended process that language learners should follow (see Figure 2). This is a 3-step procedure.
The reason for going through Step 1–3 is to make students think exclusively in English. Thus, online tools such as an English-English dictionary, collocation dictionary, corpus, or Google Advanced Search are, indeed, free from the influences of Japanese language.
7.2 Examples of self-correctionThis section will demonstrate a process of self-correction, before students submit their final assignment draft to their teacher. As shown in Chapter 3, there are four kinds of JLOS sentences. These may be (1) illogically wrong, (2) not having a good collocation (3) not harmonious sentences that are inconsistent with the surrounding words, and show (4) incorrect preposition usage. These sentences will be self-corrected below (see Figure 3). From 7.3 to 7.6, the general guideline of how to go through Step 1 to Step 3 is referred to Figure 3. Going through Steps 1–3 involves a complex mixture of online tools such as an English-English dictionary, collocation dictionary, corpus, or Google Advanced Search and therefore it is meaningless, for instance, to show a simple Google method illustration.
Figure 3. A 3-step model for self-correcting procedure.
A question the teacher asked the students was, “What is your career goal?” A student wrote the following sentence. “My career goal is to do my best work of assigned work.” This is a typical example of an illogical sentence that is a direct translation from Japanese into English.
| Step 1 | Type work of assigned work into Google. These assigned a job I cannot do phrases will come up. This result shows that work of assigned work may be wrong. |
| Step 2 | Moving onto the next step, use NBC corpus and search assigned work. The following phrase will come up: “…which are often not part of one’s assigned work.” If a student knew the phrase do my work, they could develop the following sentence: “My career goal is to do my assigned work.” |
| Step 3 | Go back to Google, and search “My career goal is to do my assigned work.” The following question will appear: “What are your career goals?” This means that their sentence would be correct, as it answers the question. However, this sentence does not include the best work phrase. In this case, a teacher is working to improve it and change it. For example: “My career goal is to do my assigned work the best I can.” |
A question the teacher asked to students was: “Why do you think this job is a match for you?” A student answered: “I guess how to arouse customer’s appetite.” However, this sentence uses an incorrect collocation.
| Step 1 | Type arouse customer’s appetite into Google. The results include how to use the term one’s appetite. For instance, a question reads, “Hey! I'm looking for an expression to say when the scent of a meal is delicious, and it can make people feel hungry. Is it ok to say it ‘arouses one’s appetite’? If not, what can a good smell of food do with appetite?” It is obvious that the student does not want to discuss food. Therefore, the student may understand that to arouse customer’s appetite is not the correct collocation that they want to use. |
| Step 2 | Now look it up in a collocation dictionary. In the dictionary, students are likely to find the phrase to arouse one’s interest/curiosity. Therefore, they might say that they know how to arouse the customers’ interest. On the contrary, if the word appetite is looked up, the dictionary firstly shows that to whet somebody’s appetite is a correct collocation for appetite. It implies that if an experience whets your appetite for something, it increases your desire for it. |
| Step 3 | Go back to Google and search whets your appetite. The second column of the first page provides common usage of whets your appetite. The following examples were shown: “My wife has been listening to reggae music all day. She heard a song by Bob Marley on the radio this morning, and it really whetted her appetite for reggae.” It is assumed that they might want to write something such as: “I know how to whet customers’ appetites.” |
A not harmonious sentence is one that is inconsistent with its surrounding sentences and may contain a mixture of informal and formal vocabulary. A student wrote the following: “But I haven’t realized those idea because I haven’t known how to materialize those ideas”. The question the teacher asked was: “Why do you think this job is a match for you?
| Step 1 | If someone inputs materialize ideas into Google, only the following usage is shown: “The key to materializing an idea is sticking to it!” This result leads us to believe that materialize ideas is not common, nor does it make much sense in English. |
| Step 2 | Look up the word materialize in an English-English dictionary. The dictionary defines it as happening to appear in the way that you expected. Then, if you typed the word materialize into a thesaurus, it would show many synonyms such as carry out and put into effect. |
| Step 3 | It is inferred that the student wants to write: “I don’t know how to put those ideas into effect.” Go back to Google and search put idea into effect. Then, it is likely to become obvious that their sentence is written in correct English; thus, it can be changed to: “But I haven’t realized those ideas because I haven’t known how to put those ideas into effect.” |
A student wrote the following sentence: “Because I think I am suited with examining small details, and integrate the results of survey.” In this context, they must prove that the job is a match for their skills.
| Step 1 | If I am suited is typed into Google, I am suited for and I am suited for/to this job because will appear. Therefore, it would be understood that they could use be suited for/to. |
| Step 2 | In an English-English dictionary, the following terms are shown: best/well/ideally/perfectly suited to/for. For example, “…the candidate most ideally suited to doing the job.” Having seen the example sentence, they might change their version in the following way: “Because I think I am suited to examining small details and, integrate the results of survey.” |
| Step 3 | In Google, there are numerous examples of the I am suited to sentence. However, the sentence in Step 2 is not completely correct. Such a case requires the teacher’s intervention in order to construct a grammatically correct sentence. Therefore, the product becomes: “Because I think I am suited to examining small details and integrating the results of survey.” |
It would be ideal if students could undergo these three steps on their own. However, if teachers find that the students have not self-corrected, they may simply introduce the ideaand provide an example. In fact, the above 3-step model for proposed self-correction skill using online resource was put into practice, and the χ2 test shows that all four facets defined in JLOS have significantly improved in comparison to English writing without the guidance of the self-correction skill (χ2=12.99, df=1, p<.001). The cross-tabulation is carried out with or without the self-correction skill against four factors in JLOS in terms of the frequency. The χ2 test significantly shows that learning the self-correction skill improves English usage regardless of the four elements in JLOS.
7.7 Limitation of the studyThe validity of this finding is implicitly associated with the three conditions shown below:
The data show that almost all students write various kinds of JLOS. The online model suggests how this JLOS can be fruitfully self-corrected by the students, possibly with little help from the teacher. A year after the data were collected, I taught students how to correct their JLOS. By practicing the JLOS model, students appeared to successfully:
As a result, the number of JLOS appeared to have reduced considerably. Again, the aim of this paper is to discourage the use of JLOS and encourage students to acquire the self-correction skill instead. Although it may sound difficult or complicated for some, it is argued that their writing ability is likely to be significantly improved once they master this skill.
