AN ANALYSIS OF GRAMMATICAL ERRORS COMMITTED BY THE TENTH GRADE STUDENTS OF SMA NEGERI 1 BATURITI IN WRITING RECOUNT TEXT IN ACADEMIC YEAR 2014/2015

The present study aimed at (1) finding the types of grammatical errors committed by the tenth grade students of SMA Negeri 1 Baturiti in writing recount text in academic year 2014/2015; and (2) finding the sources of the errors committed by the tenth grade students of SMA Negeri 1 Baturiti in writing recount text in academic year 2014/2015. This study was designed as descriptive-qualitative study. The methods used to collectdata were document analysis and interview. The instruments used were project and interview guide. The data in this study were analyzed by using data analysis theory proposed by Miles and Hubarman. The types of grammatical errors used in this study were based on the surface strategy taxonomy proposed by Dulay at al while the sources of the errors were based on the combination of theories proposed by Richards and Brown. Model of error analysis used in this study was the one proposed by Gass and Selinker. The result of the study showed that the subject committed 167 errors in their writing, namely misformation (29.94%), addition (27.54%), omission (24.55%), and misordering (17.96%). The sources of those errors were interlingual transfer (50.30%), intralingual transfer (49.10%), and context of learning (0.60%).


INTRODUCTION
English is an important language which needs to be learned. In Indonesia, English is taught as a foreign language (English as a foreign language / EFL). It is treated as a compulsory subject.
Learning the English language is about learning how to communicate using it, both in spoken and written forms. Therefore, those who want to use English communicatively have to master the four language competencies, namely listening, speaking, reading, and writing.
Among those four competencies, writing is considered to be the most difficult one (Emmaryana, 2010;Watcharapunyawong & Usaha, 2013;Panahandeh & Asl, 2014). According to Emmaryana (2010), writing needs good knowledge on the topic of the writing and good knowledge on grammar. In addition, Watcharapunyawong& Usaha (2013) states that the ability to write requires both syntactic and semantic knowledge. It means that to be able to write, knowledge of word meaning and how to arrange the word appropriately is needed. Panahandeh&Asl (2014) pointed out that writing is one of the most problematic areas in language learning. Pasand and Haghi (2013) states that writing involves the development of an idea, mental representations of knowledge, and experience with subjects. In other words, to be able to write well, one has to know words which are relevant to the topic of the writing and how to arrange them in order to express his/her ideas appropriately. The ability to produce good writings is not an inborn skill (Yahya, Ishak, Zainal, Faghat, &Yahaya, 2012). It requires a complex process involving formulating new ideas and transforming information.
According to Raimes (in Ahmed &Karunakaran, 2013), there are nine components of writing. Those include syntax, content, the writer's process, audience, purpose, word choice, organization, mechanics, and grammar. All of these components are needed to produce a good piece of writing. However, several studies related to students' writing show that students frequently commit errors and make mistakes in these components, specifically in the grammar component.
Errors and mistakes seem similar, but they are actually two different things. As cited in Abed (2013), errors are defined as violation of rules which are systemic and governed by rules. They appear because of learners' incomplete knowledge of the target language. Mistakes, on the contrary, are random deviations and unrelated to any system. They are caused by several causes including slip of the tongue, slip of the ear, false start, and non-linguistic factors such as strong emotion, lack of concentration, etc.
Errors used to be considered as a problem that should be eradicated as soon as possible, but, now they are seen as a tool which can assist in the learning process (Alhaysony, 2012). According to Corder (1967), errors are significant in three different ways. Firstly, errors tell language teachers about learners' progress towards the goal and what remains for them to learn. Secondly, errors provide to researchers important evidence of how a language is learned or acquired and what strategies or procedures the learners are employing in their discovery of a language. The third significance is that errors are important for learners themselves because the making of errors can be regarded as a device used by learners in order to learn.
Regarding to the significances of errors, teachers need to undertake a systematic analysis which is called error analysis. By systematically analyzing students' errors, teachers can find out the students' progress in learning the target language and in what way they have problems. After finding out the students' problems, teachers can try to find solution to solve the problems. Therefore, the researcher, as a future English teacher, is interested in conducting such analysis.
There are various kinds of errors committed by students in Indonesia (Bayinah, 2013;Cholipah, 2013). Bayinah (2013) conducted a study on students's grammatical errors in writing recount text. All of 31 students who were involved in her study committed errors. In total, there were 163 errors committed by the students. Another study showing that students in Indonesia commit errors in writing was carried out by Cholipah (2013). She found that the subjects of her study, which consisted of 30 students, committed 813 errors. Students of SMA Negeri 1 Baturiti, a public senior high school in Tabanan, Bali, also commit errors in writing. Based on the researcher's prior interview with the English teacher who teaches the tenth grade students in that school, it was found that the students of that school committed a lot of errors when they were asked to write English text. Therefore, the researcher is interested in analyzing the types of errors committed by these students. Out of seven classes of tenth grade students, one class was chosen randomly to be asked to produce a recount text. This type of text was chosen to be analyzed because it is frequently used in daily life as well as it is included in the school curriculum for the tenth grade students of SMA Negeri 1 Baturiti.
There are several previous studies related to error analysis (Bayinah, 2013;Limengka&Kuntjara 2013;Noviyanti, 2014). They analyzed the types of grammatical errors committed by different levels of learners in different types of text they produced. Bayinah (2013) conducted a descriptive qualitative study which aimed at (1) analyzing the types and frequencies of grammatical errors committed by the first grade students of MA Al-Khairiyah in using simple past tense and (2) finding the sources of these errors. She designed her study as a descriptive qualitative study. The subject of the study was 31 first grade students of MA Al-Khairiyah. The result of error analysis showed that the students committed four types of errors, namely, omission (41.1%), addition (4.9%), miselection (53.4%), and ordering (0.9). Meanwhile, the sources of errors were interlingual transfer and intralingual transfer. Limengka&Kuntjara (2013) conducted a similar study. The purpose of their study was to find out the types of grammatical errors committed by the fourth semester students at Petra Christian University in writing their essays for final project. They classified the errors into five categories which included addition, omission, misformation, misordering, and blends. The model of error analysis used in this study consisted of collection of a sample of learner language, identification of errors, description of errors, explanation of errors, and error evaluation. In the findings, it showed that even advance learners could commit errors. The fourth semester students who had previously taken three writing classes committed a large number of grammatical errors in their writing. Fourteen essays analyzed in this study contained a total of 266 grammatical errors, namely, addition (6.02%), omission (20.30%), misformation (68.05%), misordering (3.76%), and blends (1.88%).
Another study about grammatical error analysis was conducted by Noviyanti (2014). Her study was carried out to analyze and classify the types and the sources of students' grammatical errors in writing descriptive text. The study was designed as a case study. The subject of her study was class 8.8 which consisted of 30 students. The data were collected through test. The result of the error analysis showed that the students committed four types of errors, namely, selection (57.9%), omission (29.8%), addition (7.6%), and ordering (4.7%).
In conclusion, it was found that students committed errors in writing English text. The researchers mentioned previously expected further researches on the same topic. The study conducted by Limengka and Kuntjara (2013) was done in a natural setting, but they only investigated the types of errors committed by students without explaining the sources of the errors. On the other hand, the studies conducted by Bayinah (2013) and Noviyanti (2014) explained the sources of errors committed by their subjects, but they collected the data through a test which is not natural. Therefore, the researcher still found the need to conduct such a study in which the data were collected through a project and the sources of errors were explained based on the interview with the subject. It was hoped that the result of this study could give contribution to the development of English teaching and learning. This study was conducted at SMA Negeri 1 Baturiti in academic year 2014/2015. The subject of this study was the tenth grade students of that school. They were assigned to write a recount text. Their writings were analyzed in order to find the grammatical errors they committed. These errors were classified based on Surface strategy taxonomy proposed by Dulayet al (in Cholipah, 2013). This classification consists of addition, omission, misformation, and misordering.
Based on the background of the study above, the research questions can be formulated as follows: (1)  Derived from the research questions, the objectives of this study are: (1) to find out the types of errors committed by the tenth grade students of SMA Negeri 1 Baturiti in academic year 2014/015; and (2) to find out the sources of the errors committed by the tenth grade students of SMA Negeri 1 Baturiti in writing recount text in academic year 2014/2015.

METHODS
This study was designed as descriptive qualitative study. In conducting this research, the researcher went to a natural setting, i.e. school, to collect data. After collecting the data, the researcher identified, classified, and quantified the grammatical errors committed by subject. Besides, the researcher also analyzed the sources of errors as well as tried to remediate.
This study was conducted in SMA Negeri 1 Baturiti, a governmental school which is located in Br. Puseh, Ds. Perean Tengah, Kec. Baturiti, Kab. Tabanan, in academic year 2014/2015, precisely in June and July 2015. This school was chosen because an English teacher who teaches there said that the students of that school committed many errors whenever they were asked to write in English. The teacher tried to correct the students' error whenever they were asked to produce English text, but he still found many errors committed by the students in their writing.
The subject of this study was the tenth grade students of SMA Negeri 1 Baturiti. There were 176 students which were divided into seven classes; from X1 until X7. Because of the limitation of time and energy, out of those seven classes, only one class was chosen to be the sample of the study. Random sampling was used to decide the sample. The result showed that class X3 was the sample of this study. This class consisted of 26 students. Meanwhile the object of this study was the grammatical errors committed by the subject in writing recount text. These grammatical errors were classified into addition, omission, misformation, and misordering.
In order to collect data, there were two methods used, namely document analysis and interview. Document analysis is used to answer research question 1 and interview is used to answer research question 2.Document analysis is analyzing the recount text composed by the subject. The recount text was analyzed in order to identify, classify, and quantify errors committed by the students. After analyzing the students' writing, a face to face interview was conducted to find out the sources of errors committed by subject in writing recount text. It was quite impossible to determine the sources of errors by merely looking at the errors. Therefore, verification from subject was needed. This method was used to answer research question 2.
The main instrument in this research was the researcher herself. In carrying out the research, two other instruments were used, namely project and interview guide. A project was given to the subject instead of a test in order to make the data more reliable and valid. The class X4 students were given a project to write a recount text based on their own experience. They were given enough time to write the text as well as to revise it if they made some mistakes. After analyzing the recount text composed by the subject, the researcher conducted a face to face interview with the subject in order to find out the sources of errors they committed. While conducting the interview, the researcher brought a table consisting of errors committed by the subject and the types of the errors. The researcher asked the subject's reason in committing the errors. Then, the researcher categorized the reasons into four sources of errors, namely, interlingual transfer, intralingual transfer, context of learning, and communication strategy.
There were several steps followed in collecting data. The first step was visiting the field. In this step the researcher came to the field, i.e. SMA Negeri 1 Baturiti to ask permission from the headmaster to conduct a study at his school. After that, the researcher prepared all the instruments needed. The last step was conducting the study. Here, the researcher used all the instruments and the methods to collect data from the subject.
According to Miles and Huberman (1994), there are three current flows of data analysis in qualitative study, namely, data reduction, data display, and conclusion drawing or verification. Data reduction means eliminating unnecessary data. Data display is defined as an organized, compressed assembly of information that enables the researcher to draw conclusion. The last step is conclusion drawing or verification. It refers to drawing conclusion based on the analysis.
Data reduction is the act of selecting, focusing, simplifying, abstracting, and transforming the data that appear in transcriptions. It refers to selecting the most useful data related data to the topic of the study. This study only concerned with grammatical errors committed by the students. Therefore other types of errors such as punctuation and capitalization were ignored. To make it easier for the data analysis, the subject and the sentences they produced were coded. The subject which consisted of 26 students were coded S1 to S26, meanwhile the sentences were coded based on their occurrence in the text, e.g. the first sentence in a text was coded S1, the second was coded S2, etc. Therefore the sentences in a text produced by the subject were coded based on the students' code and their occurrence. For example the first sentence in the text produced by the first student was coded S1S1.In this step, the selected data were displayed in form of table and graph and were analyzed based on the aspects investigated. The frequency of each type of errors was counted, and the sources of errors were displayed. After that, the researcher drew conclusion based on the analysis of the data to answer the research questions.

FINDINGS AND DISCUSSION
Based on the researcher questions of this study, the findings cover the types of errors committed by the tenth grade students of SMA Negeri 1 Baturiti in writing recount text as well as the sources of the errors. The following is an explanation of the types of errors found in the students writing. The next part discusses the sources or the causes of the errors committed by the subject.
In order to find out the types of errors committed by the subject in writing recount text, a series of steps need to be followed. After the recount texts produced by the subject were collected, they were analyzed thoroughly. The researcher identified, classified, and finally quantified all the errors in the subject's writing.
Each recount text produced by the subject was read several times thoroughly in order to identity the errors committed. Any deviation from the Standard English which belongs to surface strategy taxonomy classification is identified.
After all of the errors committed by the subject in their recount text have been identified, the researcher proceeded to the next step which was classifying errors. All errors which have been identified were classified based on surface strategy taxonomy classification which consisted of addition, omission, misformation, and misordering. Addition was categorized into three, namely, simple addition, regularization, and double marking. Omission was categorized into seven, namely, omission of article, omission of to be, omission of conjunction, omission of preposition, omission of marker, omission of pronoun, and omission of verb. Misformation consisted of regularization, archi form, and alternating form. Misordering was not classified into any classififcation. The next step after classifying the errors committed by the subject was quantifying them. The errors were counted based on their type. The researcher counted how many student 1 committed addition errors, omission errors, misformation errors, and misordering errors.
The next step after classifying the errors committed by the subject was quantifying them. The errors were counted based on their type. The researcher counted how many student 1 committed addition errors, omission errors, misformation errors, and misordering errors.
From the quantification it was found that the subject committed a total number of 167 errors. The number of addition errors was 46, the number of omission errors was 41, the number of misformation errors was 50, and the number of misordering errors was 30.
Among 46 errors of addition, 39 errors were in form of simple addition, 6 errors were in form of addition regularization, and 1 error was in form of double marking. Among 41 errors of omission, 5 errors were in form of omission of article, 9 errors were in form of omission of to be, 2 erors were in form of omission of conjunction, 2 errors were in form of omission of preposition, 9 errors were in form of omission of marker, 6 errors were in form of omission of pronoun, and 5 errors were in form of omission of verb. Among 50 errors of misformation, 16 errors were in form of misformation regularization and 34 errors were in form of alternating form. The types and frequency of occurrence of each type of errors can be seen in Table 1. From table 1, it can be seen that there were four types of errors committed by the subject in their recount text. The types of errors, ordered from the highest frequency to the lowest one, were misformation, addition, omission, and misordering. Misformation errors occurred 50 times and the percentage was 29.94%. The second most frequent type of errors was addition errors. The subject committed 46 errors of this type and the percentage was 27.54%. Omission was the third most frequent error. There were 41 errors of this type and the frequency was 24.55%. The least frequent type of errors was misordering. There were 30 misordering errors found in this study and the percentage was 17.96%.
The most frequent type of errors committed by the subject was misformation. In total, there were 50 errors of this type and the percentage was 29.94%. Misformation errors are caharacterized by a presence of an unacceptable form of morpheme or structure. There are three types of misformation errors, namely, misformation regularization, archi form, and alternating form. However, in this study, the researcher only found misformation regularization and alternating form.
Regularization which falls under misformation category occurs because of the use of regular marker in place of irregular one. Examples of misformation regularization are irregular plural nouns (man  men, child  children) and irregular verbs (drink  drank  drunk, sit  sat  sat). After analyzing the students' writing, it was found that there were 16 errors which could be categorized as misformation regularization and the percentage was 9.58%. Below is the example of misformation regularization errors taken from the recount text produced by the subject. Code S4S10 : We ate the food my family bringed.(correct) Code S4S10 : We ate the food my family brought. (incorrect) Alternating form errors are fairly free alternation of various members of a class with each other. There were 34 errors of this type found in the recount text written by the subject and the percentage was 20.35% of the total errors. Errors committed by the subject which were categorized as alternating form were generally the use of incorrect form of verb. Some students used bare infinitives in place of present participles, some students used past tense forms where bare infinitives were needed, and so on. The following is an example taken from the subject's writing. Code S22S5 : After arrived there, we parked our motorbike. (incorrect) Code S22S5 : After arriving there, we parked our motorbike. (correct) The second most frequent error committed by the tenth grade students of SMA Negeri 1 Baturiti in their recount text was addition errors. Addition errors are those in which learners put an item or more which should not be put in a well-formed sentence. The total number of addition errors found in the recount text composed by the subject was 46 with percentage of frequency was 27.51%. Based on the analysis of the students' writing, it was found that there were three types of addition errors, namely, double marking (1), regularization (6), and simple addition (39).
Double marking errors are those in which learners fail to delete certain items which are required in some linguistic constructions, but not in others. The items mark for the same feature. From the analysis of the students' writing, it was found that there was only one error which could be categorized as double marking errors. Code S7S13 : It seemed ( ) ( ) did not wanted to leave this beautiful place. (incorrect) Code S7S13 : It seemed ( that) ( I ) did not want to leave this beautiful place. (correct) Addition regularization is characterized by the presence of a marker that is typically added to a linguistic item after exceptional items which do not take marker. Examples of these items are some irregular nouns (food, tea, coffee), some irregular verbs (cut, cast), etc. There were 9 regularization errors found in the students' writing. The following is one of them. Code S26S6 : But then we caught many fishes.(incorrect) Code S26S6 : But then we caught many fish. (correct) The last type of addition errors is simple addition. It refers to an error in which an unwanted element is added in a sentence. There were 39 errors categorized as simple addition found in the text composed by the subject. The following is an example of simple addition errors taken from the text. Code S3S5 : Then, at 1 o'clock in the afternoon, I went back to home. (incorrect) Code S3S5 : Then, at 1 o'clock in the afternoon, I went back home. (correct) The third most frequent errors found in this study was omission errors. Omission errors are the opposite of addition ones. This type of errors is characterized by the absence of an item which must be present in order to construct a well-formed sentence. There were 41 errors which were categorized as omission errors found in the students' writing and the percentage was 24.55 %. Five of them were categorized as omission of article, nine were categorized as omission of to be, two were categorized as omission of conjunction, five were categorized as omission of preposition, nine were categorized was omission of marker, six were categorized as omission of pronoun, and five were categorized as omission of verb.
Generally, there are two types of articles in English, namely infinite article and definite article. Infinite article consists of 'a' and 'an' while definite article consists of 'the'. Sometimes learners do not put an article in which it is necessary. From the analysis of the text written by the subject, there were found five errors which were categorized as omission of article. The following is an example of this type of errors.
Code S11S11 : It was ( ) nice experience. (incorrect) Code S11S11 : It was a nice experience. (correct) There are several types of to be depending on the subject of the sentence and the tenses used. To be (is, am, are, was, were) are used in passive voice, nominal sentence, and some tenses. In the students' writing, the researcher found 9 errors which could be categorized as omission of to be errors. The following is an example of this type of errors committed by the subject in their recount text.
Code S1S7 : That ( ) so funny. (incorrect) Code S1S7 : That was so funny. (correct) Conjunction is needed to combine two or more sentences into one sentence. In this study, it was found that there were 2 omission of conjunction errors. One of them is: Code S7S13 : It seemed ( ) ( ) did not wanted to leave this beautiful place. (Incorrect) Code S7S13 : It seemed that ( I ) did not wanted to leave this beautiful place. (correct) Preposition is an essential element of English language. There are several kinds of preposition like at, before, in, after, etc. In this study, the subject committed 5 omission of preposition errors.the following is one of them. Code S11S3 : We arrived there ( ) about 2 o'clock. (incorrect) Code S11S3 : We arrived there at about 2 o'clock. (correct) There are several markers in English language like suffix -ed/-d to form past tense verbs, -es/-s to form plural nouns, etc. Sometimes learners fail to put a marker where it is needed. After analyzing the recount text produced by the subject, the researcher found 9 errors which belonged to this type. One of them is: Code S14S6 : There were so many local and foreign tourist( ). (incorrect) Code S14S6 : There were so many local and foreign tourists. (correct) A well-formed sentence consists of at least subject and verb. Subject can be in form of noun or pronoun. Sometimes learners commit errors in which they do not put a pronoun in a sentence. There were six errors of omission of pronoun. Here is one of them: Code S4S14 : And ( ) took a rest at home. (incorrect) Code S4S14 : And we took a rest at home. (correct) Omission of verb errors are those in which learners fail to put a verb in a sentence which needs one. There were 5 errors of omission of verb. Code S6S8 : And then we ( ) around all objects in botanical garden. (incorrect) Code S6S8 : And then we walked around all objects in botanical garden. (correct) Misordering errors refer to incorrect placement of morpheme or a group of morphemes. There were 30 misordering errors found in this study and the percentage was 17.96%. The following is an example of misordering errors taken from the text produced by the subject. Code S15S4 : I and my friends rode motorcycle. (incorrect) Code S15S4 : My friends and I rode motorcycle. (correct) After collecting data, identifying, classifying, and quantifying the errors committed by the subject, the researcher proceeded to the next step which was analyzing sources of errors. It was done by conducting a face to face interview with the students about the reasons why they produced the errors found in their writing. There were three sources of errors found in this study, namely interlingual transfer, intralingual transfer, and context of learning. The frequency and the percentage of each source of errors can be seen in table 2. Interlingual transfer refers to the negative transfer of the first language. Sometimes learners use their first language's rules in making the target language sentences. Identifying an error caused by interlingual transfer can be done by translating the sentences produced by learners in the target language into the learners' first language. There were 84 errors found in this study as the result of interlingual transfer. Code S5S7 : Munduk waterfall ( ) very good and beautiful. (incorrect) Code S5S7 : Munduk waterfall is very good and beautiful. (correct) Intralingual transfer is the negative transfer within the target language itself. It is the incorrect application of within the target language. Out of the four types of intralingual transfer, there were only three types found in this study. They were overgeneralization (15), ignorance of rule restriction (16), and false concepts hypothesized (51).
When learners discover a rule in the target language, they might apply the rule incorrectly. They overgeneralize the rule, thinking that it is always applicable. The Ignorance of rule restriction is a failure to observe the restrictions of existing structures. In other words, it is the wrong application of rules to context in which they do not apply. There were 16 errors found in this study as the result of ignorance of rule restriction. Code S20S7 : After that we went back to home. (incorrect) Code S20S7 : After that we went back home. (correct) False concept hypothesized is derived from faulty comprehension of distinctions in the target language. This can be the result of poor gradation of teaching items. Examples of this source of errors are the confusion between time and tenses used in English, the confusion of verbal and nominal sentences, the confusion between go and come, the confusion between so and too, etc. There were 51 errors found in this study caused by false concepts hypothesized.
Code S13S2 : We were visited Sanur beach. (incorrect) Code S13S2 : We visited Sanur beach. (correct) Context of learning refers to sources of errors that come from outside. Examples of these sources are teachers, textbooks, the situation of the class, etc. Learners often commit errors because of misleading explanation from teachers and textbooks. In this study, there was only one error found which was caused by context of learning. Code S2S1 : It's happened on October 31, 2014 when it was evening. (incorrect) Code S2S1 :It happened on October 31, 2014 when it was evening. (correct) Communication strategies are related to strategies owned by learners in learning target language. It is the conscious employment of verbal and nonverbal mechanisms for communicating an idea when precise linguistics forms are not mastered by learners at a point of communication. This source of errors was not found in this study. The researcher assumed that it was because this source of errors is more commonly found in spoken communication.
Based on surface strategy taxonomy proposed by Dulayet al. (in Cholipah, 2013), there were four types of errors, namely addition, omission, misformation, and misordering errors. Similarly, Abed (2013) states that process based classification of errors classify errors into four, namely omission, addition, substitution, and permutation. Substitution is the same as misformation, and it can also be called misselection errors.
Meanwhile permutation is the same as misordering, and it can also be called ordering errors. All of the four types of those errors were found in this study.
The most frequent type of errors committed by the subject was misformation errors. There were found 50 errors of this type and the percentage was 29.94%. In line with this finding, the studies conducted by Bayinah (2013), Limengka and Kuntjara (2013), and Noviyanti (2014) also found that the most frequent error committed by their subjects was misformation errors. Bayinah (2013) found that there were a total number of 87 errors which belonged to misformation errors and the percentage was 53.40%. The study conducted by Limengka and Kuntjara (2013) also found a similar result. The highest frequency of errors found in their study was misformation errors. The subject of their study committed 181 errors of this type and the percentage was 68.05%. Similarly, the result of a study conducted by Noviyanti (2014) also showed that misformation was the most frequent type of errors committed by her subject. She found that there were 99 errors of this type and the percentage was 57.90%.
The least frequent error found in this study was misordering errors. It was found that there were 30 misordering errors and the percentage was 17.96%. In line with this finding, the study conducted by Bayinah (2013) showed that misordering error was also the least frequent error committed by the subject. There was only 1 misordering error and the percentage was 0.90%. Limengka and Kuntjara (2013) only found 10 errors of this type and the percentage was 3.76%. It had the lowest percentage compared to the other three types of errors. Similarly, Noviyanti (2014) found 8 misordering errors and the percentage was 4.7% which was the lowest percentage among the four types of errors.
There were some theories regarding the sources or causes of errors. The one used in this study was the combination of Brown's and Richards' theories. There are four sources of errors, namely interlingual transfer, intralingual transfer, context of learning, and communication strategy. Intralingual transfer can be classified into four classifications, namely overgeneralization, ignorance of rule restriction, incomplete application of rules, and false concepts hypothesized. However, there were only three sources of errors were found in this study. Those were interlingual transfer, intralingual transfer, and context of learning. There was no error caused by communication strategies found in the students' writing. Among four kinds of intralingual transfer, there were only three kinds found. Those were overgeneralization, ignorance of rule restriction, and false concepts hypothesized. Incomplete application of rule was not found in this study.
The most frequent source of errors found in this research was interlingual transfer. There were 84 errors caused by this type of sources of errors and the percentage was 50.30%. This finding is in line with the finding of the study conducted by Bayinah (2013). She also found that interlingual transfer had the highest percentage compared to other sources of errors, that was 56.40%. This finding was also supported by the finding by Noviyanti (2014). She found that interlingual transfer caused her subject to make many errors.
From the explanation above, it can be seen that in terms of types of errors, the finding of this study supported the theories related to the types and classification of errors. Moreover, it also supported the findings of previous researches. However, in terms of sources of errors, there was a slight difference between the finding of this study and the theories proposed by experts.

CONCLUSION AND SUGGESTION
After analyzing the recount text produced by the tenth grade students of SMA Negeri 1 Baturiti, it was found that there were four types of errors. Those were misformation , addition, omission, and misordering. There were two types of misformation errors, namely misformation regularization and alternating form. Addition errors consisted of simple addition, addition regularization, and double marking. Omissions errors were categorized into seven categories, namely omission of article, omission of to be, omission of conjunction, omission of preposition, omission of marker, omission of pronoun, and omission of verb.
Regarding the sources which caused those errors, it was found that there were three sources, namely interlingual transfer, intralingual transfer, and communication strategy. There were three types of intralingual transfer found in this study. Those were overgeneralization, ignorance of rule restriction, and false concepts hypothesized.
Based on the finding of the study, the researcher would like to give several suggestions to some parties, namely students, teachers, and other researchers who are interested in conducting studies about this topic.
Students are expected to be creative and innovative in learning English. English is not a content subject, it is a skill subject. Therefore students should practice a lot to be able to master English. Students should not be afraid of making mistakes and errors because they are inevitable, but they can try to minimize committing them by practicing. They can try writing their ideas in English even though they are not assigned by their teachers. After finished writing, they may exchange their writing and try correcting their friends' writing.
Writing is an important skill. Teachers need to continuously give writing exercise to students. A simple activity like writing a summary of today's lesson is suggested. Teachers need to give students a chance to correct their writing before submitting it. Besides, teachers should also give correction and feedback of the students' writing.
The researcher realized that this thesis is far beyond perfect. There were many weaknesses can be found in this thesis. Therefore, the researcher hopes that those who are interested in conducting a study on this topic can conduct a better and more complete study. There are other classifications of errors and other genres of text that can be studied.