Error Analysis in Composition Written by Nursing Department Students of Universitas Muhammadiyah Yogyakarta

There is more freedom nowadays for workers to move across border. One of the promising opportunities is to become an overseas nurse. However, to become an overseas nurse, one needs a certain level of language skills; thus, it is important to embed English, as the world’s lingua franca, in the nursing science course. When the students of the Nursing Department master the English, they will be able to engage in a communication with others within the international scope. It is very important to prepare them with standard, error-free English to prepare them entering the global market. This research tries to figure out the students’ language proficiency by viewing their grammatical errors especially in the S-V agreement. The data were taken from the composition written by Nursing Department students, which were then analyzed using the concordance program, namely SCP 4.05. Using the seven pronouns (I, we, you, they, he, she, it) as the key words, the sentence contexts were examined. Then, the sentences that contain errors were taken and categorized. From this research, there are 5 kinds of students’ problems reflected in the errors. They are: 1) sentence without verb, 2) errors in the use of be and stem, 3) inflectional –s for the third singular person, 4) modals, and 5) auxiliary verb.


INTRODUCTION
The advancement of technology and transportation along with the global policies, such as Asean Economic Community (AEC) or European Union, makes it possible for a person to work in another country easily.Consequently, there is a need for lingua franca because of the mingling of different person with different nationalities needs one language to communicate to each other to conduct the business (www.ncsbn.org).One of the most widely known and used language is English.Therefore learning English becomes very important nowadays, and at the same times becomes the rich source to do a research on it.The phenomenon of language produce by the learners already attract attention since

61-72
many time ago and "has long tradition" (Sridhar in Ellis, 1985, p.51).The purpose of error analysis has moved from merely checking the errors and linguistics classification to more pedagogic one, to provide information on the teaching materials and remedy (Ellis, 1985).Fauziati (2011) support Ellis view in her writing that error analysis function is broadened to view the learning process by analyzing the language produced by the learners.One way to understand English proficiency is by looking at the errors made by the learners because they are part of the learning.According to Fauziati (2011, p.26), some experts believe that Current literatures view errors not just as deviations but rather as a source for studying the process/ strategies used by the learner in learning the TL.
They are evidence about the nature of the process and of the rules used by the learner at a certain stage in learning course.Therefore, if we want to study the learners' IL system, we should find clues to the systems by analyzing the errors they make.
The term TL here is the target language; it is the language that is learned by the learner.IL is the interlanguage or the language produce by the learner which is not yet the similar to the target language and it has its own characteristics (Richard and Schmidt, 2002).
Hence, it is very interesting to analyze the errors made by the students to understand their competence in using the language.Analyzing the written text is easier because it is the tangible evidence that can be studied over and over.Therefore, the writer chooses to study the composition made by the Nursing Department Students of UMY.Based on the writer's experiences in teaching them, they still had difficulties in accom-plishing the minimum requirement of writing a sentence, such as subject-verb agreement.Some examples of the sentences are as follow: She afraid with her boy friend.They are become piece worker.She agree to do it.
They not yet married.
Sentence may have correlation in many different ways called "sentences relatedness" (Fromkin, Blair, & Collins, 2000, p. 129) and the relationship between subject and verb in a sentence is called agreement or "subject-verb (S-V) agreement" (Fromkin, et al, 2000, p. 136).S-V agreement is a basic ability to make an intelligible sentence.The students' acquisition on the rules of this S-V agreement can be used as an indication of the student language competency on the understanding of the basic sentence.Richards and Schmidt (2002, p. 184) state that "errors have been studied to discover the process learners make use of in learning and using a language".This research will figure out the student's language ability viewed from the ability to perform correct S-V agreement in their sentences.The aim of doing the error analysis is to provide information on the difficulties students have in learning English to help teachers or institution to design program (Richards & Schimdt, 2002).
To find out the errors in the S-V agreement, the subject pronoun (I, we, you, they, he, she, it) in the students' compositions were used as the key words in the concordance program to limit the sentences.The focus of the analysis was in the agreement between the subject and verb in those selected sentences.The results of the analysis can be used to investigate the pattern of S-V agreement and the problems commonly faced by the students in composing a good and eligible sentence.Thus the research question will be: (a) Do the sentences used by the students in their composition meet the requirement of the S-V agreement?And (b) what kind of errors in the subject-verb agreement that the sentences commonly have?

LITERATURE REVIEW ERROR ANALYSIS
To define errors in language learning, one should also understand interlanguage because both are interrelated.Errors are defined as the norm deviation or faulty in using the target language from the result of "incomplete learning" (Ellis, 2003, p. 260;Richards & Schimdt, 2002).When the errors occur in the language as the learners produce the target language, the specific language is made by them.This kind of language produce by the learners is called interlanguage.Brown in Hourani (2008) defines interlanguage as the language produced by the learners which is different from their mother tongue and the target language.Mitchell and Myles (2004, p. 137) support the argument by stating that …interlanguage…involved a major shift away from viewing learner language as a defective version of the target language, or as a mixture of first and second language…viewing it as an organic system with its own internal structure.Thus, interlanguage is the language produced by the language learner that has not been complete yet and contain errors.The language is also different from both the native language and the target language.Below are the examples of the interlanguage: 1) *My name is Gilang Cikal Romadhan, I am school at SMA 7 Jakarta.2) *I love biology but I love not subject language French.3) *I have a sister, she age 19 years old and she study at University Esa Unggul.4) *I like sport because with sport we can healthy and strongly.(Fauziati, 2011, p. 26) Error analysis is a methodology or study to investigate learners' errors when they learn a target language.The errors made by the learners in their language production can be the result of their learning process or the product of the language learning instructions made by the teachers.Therefore, the error analysis can be used to figure out the learner's learning process and portrait the language competence mastered by the learners.The language mastery can be treated as the input to emphasize on how to teach learners, and as a basis to prepare the material.Indeed, the error analysis has the pedagogical application or implication in language teaching (Corder in Ellis, 2003).
To study the errors, it is important to know the differences between errors and mistakes as well as the classification of errors.Errors are the result of the imperfect learning where the learners do not master the language completely yet.Meanwhile, mistakes lay on the inability to perform the language correctly not because they lack of competence.Mistakes occur because the factors that influence the performance, e.g.slip of the tongue, carelessness, or fatigue (Richards & Schimdt, 2002).The classification of errors will be explained in the next paragraph.
The classification of errors fall into "vocabulary (lexical errors), pronunciation (phonological errors), grammar (syntactic errors), misunderstanding of a speaker's intention or meaning (interpretive errors), production of the wrong communicative effect, e.g. through the faulty use of a speech act or one of the rules of speaking (pragmatic errors)" (Richards & Schimdt, 2002, p. 184).Another category is called "a surface strategy taxonomy of errors" (Dulay et al in Ellis, 2003, p.56).
The surface strategy taxonomy of errors is one way to describe errors "that focus only on observable, surface features of errors, as a basis for subsequent explanation" (Ellis, 2003, p. 54).The term surface structure comes from transformational grammar "to denote the actual syntactic realization in a language of an utterance" (Poole, 1999, p.195).Chomsky defines the surface structure as the structure which is obvious in the surface to contrast it with the deep or abstract structure (Aitchison, 2003).Therefore the surface strategy taxonomy of errors tries to describe the errors based on the linguistics categories as mentioned in table above (Ellis, 2003).
Besides the categories, there are some steps or procedures in conducting the error analysis.
Corder in Ellis (2003) and Gass and Selinker (2001) propose almost the same steps.There are similarities of both steps from step one to three and there are variation starting from steps four and five.The first step is the determining the data or what is so called language corpus that can be written or oral data along with its size.The second is the error identification, which may refer to the category or type of errors that is chosen, such as grammar (syntactic errors) or vocabulary (lexical errors).The third is to classify the errors based on the grammatical description, e.g., subject-verb agreement, verb form, etc.For steps four and five, Gass and Selinker (2001)

FORM VERSUS MEANING
It is obvious that the form of language determine the meaning.Syntax rules are needed to  Ellis, 2003, p.56) construct phrases and sentences out of morphemes as well as words, and the sequences of part of speech in a sentence must be meaningful and make sense.Once the requirement of putting the part of speech in the right order is obtained, the sentence is well formed or grammatically correct (Fromkin et.al, 2000).Hogue (2003) argues that a sentence can also be defined as a group of words that has a subject and a verb and expresses a complete thought.By this, it means that if grammar rules are too carelessly violated, communication may suffer.It is what Hogue (2003) says as having a complete thought.In addition, Harmer (2001) states that the grammar of a language is the description of the ways in which words can change their forms and can be combined into sentences in that language.Through the sentences, the meaning is conveyed or the communication takes place.Therefore, correct grammar is needed to produce an eligible sentence.
It can be summed up here that the form and the meaning are closely interrelated.The evidence of how form can affect the meaning can be seen from the active passive sentence making.The different position of subject and verb will influence the doer and the sentence meaning can be totally different.To perform a language or to put the language into the actual use, the system of language need to be acquired.By obtaining the grammatical or syntactic rules, the sentences produced will be in line with the language system; thus, the sentences will make sense and be understood.Ungrammatical sentence will lead to the disturbance of communication.As the result, the message is failed to be conveyed.It is in line with Ellis (2003, p. 707) who states that analyzing learners' language should also concern with the "correctness" and "appropriateness" where the former involves "rules of usage" or grammar and the latter involves "rules of language use" or communicative message.

MINIMUM REQUIREMENT OF S-V AGREE-MENT.
This research tries to find out learners' problems in using S-V agreement as the minimum requirement, which shows the language ability of the users.However, if the learners have not yet been able to write correctly the minimum requirement, it means that teaching learning process should solve this problem first before moving on to more complex materials.This is in line with one of the functions of error analysis proposed by Richards and Schimdt (2002) to provide information on the problems the learners encounter during their learning language to determine the 'pedagogically relevant materials' (Gass & Selinker, 2001, p. 65) In conducting an error analysis, it is very important to determine the identification and classification of errors whether it is grammatical or vocabulary errors and to classify the errors (Gass & Selinker, 2001).In this article, the correlation between subject and verb is used to classify the errors.Thus, it is important to know the different forms and types of verbs and any part that correlate with the verb such as auxiliary verb, inflectional -s, modals and stem.
To begin with, verbs can be classified into linking verb and action verb or transitive and intransitive.The component of the verb is the auxiliary verb and the stem.According to Shiach (1995) there are several points to avoid the errors when using the verb.The first is that verbs have to agree with the subject of the sentence or the part of sentence.The form of verbs varies according to who or what is carrying out the action, expressing the feeling or 'being'.Next, the two main factors that affect the form of the verb are the person or subject whether it is the first, second, or the third person, and also the plurality, whether the person or subject is plural or singular.Hogue (2003) also states that verbs must agree with their subject in number (singular or plural) and person (I, we, you, they, and so on).A singular subject (one person or thing) has a singular verb.On the other hand, a plural subject (two or more people or subject) has a plural verb.Another aspect to be considered is the tenses of the sentence, whether it is in the past, present or future.A pronoun such as 'each, anyone, everybody, etc.' takes a singular verb.Collective noun as they mean a single group, usually take a singular verb.

RESEARCH METHOD
In line with the steps proposed by Corder in Ellis (2003) and Gass and Selinker (2001) that are presented in the previous section, below are the steps: 1. Determining the data or what is so called language corpus 2. Identifying the errors 3. Classifying the errors 4. Explaining the errors First, the corpus data should be determined.The data were collected from nursing students' compositions consisting of 56 soft-files or compositions.The compositions were typed in the computer.The compositions were the final written assignment and the students could choose some topics provided by the teachers.The assignments made by the students were submitted in the form of the soft copy as part of the marking.The writer did not tell the participants that their works would be used for research because the interest to do research rose when the writer read carefully the compositions and there was no time to meet the participants again to ask their permission to use their work.
The data then were entered in the concordance program, and the results showed that there were 1482 sentences with 22.853 words.This data then was shortlisted using the key words to make further analysis and to limit the investigation within the subject-verb agreement.The key words were the seven pronouns, namely I, we, you, they, he, she, and it.After each key word was sorted, then the sentences were identified using the grammatical features involving the subjectverb agreement, such as verb form, auxiliary verb, and modals.The next step was to classify the errors which resulted in the types of S-V agreement errors found in the composition written by Nursing Department students.
The writer used concordance software namely Simple Concordance Program (SCP) 4.05.Concordance program is software used to search word queries in the linguistic corpus (www.tuchemnitz.de).The corpus serves as the database, and the concordance program provides help to search and to sort words in the database.Since this research tries to find the subject-verb agreement focusing on the verb from, auxiliary verb and modals, the pronouns of I, we, you, they, he, she, and it were chosen as the key words to see the agreement between the subjects and the verbs.After the sentences were sorted based on the key words, then the sentences were analyzed based on the types of errors that commonly occur in them.The concordance program was able to shortlist the sentences using a certain keyword(s); but, the program cannot classify the sentence errors.The writer was the one who did the analysis to identify and classify the errors.

FINDINGS
After the corpus data were put in the SCP 4.05, they were shortlisted using the keywords consisting of six pronouns.The writer then identified and classified the findings based on the relation between the subject and verb.The findings showed some problems with subject-verb agreement that the writer noticed.The problems were divided into several categories: a) sentence with no verb, b) problems with modals which is not followed by stem, c) problems with the auxiliary verbs, d) problems with the inflectional -s for the third singular person, and e) problems with be + stem or just S+V-ing without be.

SENTENCE WITHOUT VERB
They sad if we sad they happy he healty an freesh she/he still young than us, she always angry and very blazedshe afraid with her boy friend It curious if there I sure that she is we free for express we lazy to walk again we similar other people to stop if we happy When and you difficult to call him They difficult to get job you while togetherwe beside our dearIf they usually with it If they still at school,They for that small body to take limit he the lord that he the only child. it less in nutrient compositionI the activites that contribute often and they time to use in the outside housethey always health also they still kids.she/he more adult thinking than usthey permission to Aslan for if we often consumption fast foodwe consumption they always attention their diets they you often consumption of fast The table reveals that most participants had problems with the use of 'be' and its variations.After a subject, there should be 'be' form if it is followed by adjective, adverb, and noun.For example "She always angry", "We beside our dear", and "They still kids".They were also still confused about word formations since they used noun derivations for the verb.For example "they permission to Aslan", or "We often consumption fast food" Unlike the previous issue, 'be' and its variations were used excessively here for stems do not need the form of 'be', for example "I'm continue study".However when the students did not use 'be', they wrote V-ing instead of just stem, for example "We including in good peoples".If the sentences were intended to be in continuous tense, they lacked form 'be', e.g., "We falling in love".The problem found was the problem with the inflectional -s.Suppose that the sentences were in the simple present tense with the third singular person meaning that the sentences should use verbs with inflectional -s, such as "He like" or "she agree to do it".It was obvious that it was not because of carelessness or mistyped because it occurred quite frequently.

PROBLEMS WITH MODALS
Modals must be followed by stem and no preposition whatsoever.However, the data showed that most participants used 'to+stem' after modals, e.g., "We must to protect beauty".The other problem was that they used past verb after modals probably because it was intended to be past tense, but instead of changing the modal into past, they change the verb, for example "We can found many advantages".In addition, the students also had problems with modals followed by adjective and noun.They also tended to use formation as verb, for example "So we can conclusion that…"

PROBLEMS WITH AUXILIARY VERB
Do you ever goes to mountain rangeIf you ever goes you most don't you be afraid if Are you even heard about Genotipe Z Are you agree about news avian influenza Are you like freesh air?/ Would do you fell ? to you how is important become healthyHad you imagined a bad incident struck make we doesn't pleasant in thereMake we doesn't free to move, after that will they not worry sick because In love we not only find passionthat we not must see anything to we not struck they are not know negatives they not yet marriedbut they not think about they not want if they not want if their they haven't capabilityshe is not endure life in city This section shows that the participants did not have adequate knowledge about how to make a negative and affirmative sentence.This part represents the students' problem in using the auxiliary verb.They were still confused about using auxiliary, such as 'form be', do, does, and modal.The sample sentences are "Do you ever goes to mountain range?", "Are you agree about…", "we not must see anything" and "We doesn't pleasant in there".

DISCUSSION
The of the research show that the students' English was still very much influenced by their Bahasa Indonesia.In Bahasa Indonesia, a sentence can still be understood even though there is no verb in it, for example: Dia mahasiswa yang baik.or Dia adalah mahasiswa yang baik.Dia Tuhan yang Maha Pemurah.or Dia adalah Tuhan yang Maha Pemurah.
Both sentences above have the same meaning and are grammatically correct.However, when the principles in Bahasa Indonesia are put in English, the sentences become grammatically incorrect.The fact can be seen in category a) sentence without verb.Subjects in the example below are followed by adjectives without linking verb 'be'.
we can quiet and relax we will quiet They sad if we sad we beside our dear he the lord that we consumption The other problems were the problems concerning with the verb.Verbs in Bahasa Indonesia do not require as many changes as in English.In English, many aspects can change the verbs of the sentence, such as time, plurality of the subjects, as well as negation and question.The auxiliary verbs make English more complicated.The students might have not mastered all those complexity yet, but they must write in English, so they mixed up their Bahasa Indonesia system with English.
The results of the sentences made by the students whose learning is incomplete because they still made errors in writing their language and still mixed up the system in Bahasa Indonesia to write sentences in English are below: Make we doesn't free to move they not want if their that we not must see anything they not worry sick because The students usually used dictionary to help them in learning English.It seems that they often used the words in dictionary as It was probably the reasons why the students used modal+to infinitive since the dictionary used 'to' to signal that the word following it was a verb.The examples below show the results of the data: we must to protect beauty > kita harus melindungi kecantikan we can to less our smoking so > kita dapat mengurangi rokok jadi… we will to vomit, dizzy > kita akan muntah, pusing The other reason for the students' problems in creating eligible sentences is probably from the English teaching learning process used in Indonesia.The process of teaching and learning is elaborated in the curriculum.When the curriculum change, the teaching learning process will be also different and this condition will affect the result of the learning that can be viewed from the errors made by the students (Ellis, 2003;Sawalmeh, 2013).It is common knowledge that the Indonesian curriculum is changed quite often.The curriculum usually follows principles from a certain approach.One of the approaches implemented in Indonesia was the Communicative Language Teaching.In Communicative Language Teaching, one of the principles is that grammar is acquired subconsciously through the use of language during the learning process (Richards & Rodgers, 2001).Krashen in Griffith and Parr (2001) argues that languages cannot be learned, but can only be acquired through natural communication.For that reason, grammar might have been taught inductively so it is likely that the students did not master the grammar well.Ellis (2003) writes that general source of errors is the "induced errors", which occurs "when the learners are led to make errors by the nature of the instruction they have received (p.60)".
The influence of Bahasa Indonesia should be avoided when using English since Bahasa Indonesia has different language system from English.Even tough using mother tongue language system to produce target language is part of learning, but the process should be eliminated gradually.During the process of learning a language, the students have interlanguage and interference error which means "the use of elements from one language while speaking another" (Richarad in Ellis, 2003, p. 60).During the process of learning, if students, especially Nursing Department students, still have the native language influence surely that their language, especially the written one, will be hard to understand because it is unintelligible for the other.To overcome the problems, it is very important that there should be a paradigm shift from Bahasa Indonesia to English so that the students will be aware of the differences and later are able to use English system and syntactic rules when they use it.

CONCLUSIONS AND RECOMMENDATION
In conclusion, in this globalized era the mastery of foreign language especially English is very important to prepare the students to face the challenge in the workplace and to give them better bargaining position.One way to know the students' mastery is by analyzing the errors that they made during the learning.This information can be used as the basis to determine the learning materials suitable for them.The aim of this research is to know the students language competence by viewing the errors they made in their sentence viewed from the S-V agreement.There are 5 types of errors that the students made; first is sentence without verb, second is errors in the use of be and stem, third is inflectional -s for the third singular person, then modals and finally auxiliary verb.
This research is preliminary one.Many parts of speech still need to be analyzed.Researchers who want to conduct a similar research may analyze the other grammatical points as the basis to conduct the error analysis.As the influence of Bahasa Indonesia is quite obvious here, there is also an open opportunity to study the influence of first language on the second or foreign language using this method.
It is likely that teachers or curriculum makers should adjust with the students' ability in English.It is important to know the level of students' English proficiency and the goals that students should reach so that teachers and institutions as the curriculum makers can decide the best intervention for the students.The teaching should be conducted from the easy level before moving on to more complex features.It is important to make sure that the students master the Basic English first so that the students will learn language that is in line with their competency and pace.When the teacher acknowledges the students' level, it seems that the teacher can give a higher language skills to promote the learning.

TABLE 1 :
A SURFACE STRATEGY TAXONOMY OF ERRORS Source: Dulay et al in

TABLE 2 :
SAMPLES OF SENTENCES WITHOUT VERBS

TABLE 3 :
SAMPLES OF SENTENCES WITH BE + STEM AND S+V-ING WITHOUT BE.

TABLE 4 :
SAMPLES OF SENTENCES WITH PROBLEMS WITH THE INFLEC-TIONAL -S FOR THE THIRD SINGULAR PERSON SUBJECTS.

TABLE 5 :
SAMPLES OF SENTENCES WITH PROBLEMS WITH MODALS

TABLE 6 :
SAMPLES OF SENTENCES WITH PROBLEMS WITH AUXILIARY VERB