Error Analysis On Efl Students’ Essay Writing (Fourth Semester Students of STMIK Sinar Nusantara Surakarta)

This research was finding out the errors made by English Foreign Language students in the essay writing using James’ error analysis. This study was qualitative descriptive research with 30 respondents. The data were collected by students’ mini research report which was then analyzed by observation and classification using Corder’s algorithm. It basically consists of three major stages: recognition, description, and explanation. Finally, the result was classified by using James’ four grammatical error categories. It was found out that there were 200 (100%) grammatical errors containing omission as many as 51 errors (25.5%), addition as many as 18 errors (9%), misformation as many as 105 errors (52.5%), and misordering as many as 26 errors (13%). An error correction activity was also conducted in order to decrease the errors. The result after the treatment was quite significant in which the total number of grammatical errors was decreasing from 200 to 92 containing Omission (22.9%), Addition (8.7%), Misformation (60.8%), and Misordering (7.6%). The main causes of the errors are the lack of the students’ understanding on the target language’s grammar rules and their tendency of using their first language’s grammar rules.


INTRODUCTION
Among four language skills namely reading, writing, listening and speaking, writing is considered to be the most difficult since it is not merely expressing an idea into a written form but also how to write it cohesively and coherencely to produce a nice writing with smooth flow of ideas. Cohesion itself links one word to another word, one sentence to another sentence, and one paragraph to the next paragraph so that it can clear up the meaning (Ismail & Linda, 2018) . This notion of study was conducted by (Mustafa et al., 2016) which found that there were significant errors either in the surface strategy taxonomy or Linguistic category taxonomy. In surface strategy taxonomy, the dominant error was the selection in which it contributed as much as 72% to the total error. The second was omission as much as 14.4), and the last was addition as much as 10.6%. On the other hand, the most dominant types in the Linguistic category taxonomy was word forms (48.4%), then articles (35%), non-finite verbs (34.9%), verb tenses (34.3%), plurals (33.3%), and prepositions (30%). Next was a recent study conducted by (Octaviani, 2021) . In her research findings, she mentioned that the students made omission error as much as 35.35%, misformation errors as much as 30.80%, misordering error as much as 24,23% and addition as much as 9,62%. All of the previous three studies agreed that the lack of understanding on the target language's grammatical rules and the tendency of using the first language's grammatical rules became the two major causes of the errors. However there was no error correction process conducted in theser studies. Thus ,the research gap that was taken by the writer here was that there would be an error correction and discussion activity after the types of the grammatical error and the error sources were found out. This phase of correction activity is considered important since it provides an opportunity for the students to know the errors they make, why and how the errors appear. The writer believes that this activity can help the students have more understanding in how to make a good English writing. Therefore it is hoped that they will find it easier to write academic English essays that will help them much in composing an abstract either for their thesis or to be published in a scientific journal or at least at the university repository as a part of the requirements to fulfill when they are at the end of their study.

METHODOLOGY
This study meets the characteristics of qualitative -descriptive research in which it deals with why and how a phenomenon has happened (Harley, 2006) . In addition, according to (Nawawi, 1993) , a qualitative descriptive research is a research trying to reveal a phenomena, an event, or a condition as what it is. In short, she stated that qualitative research is merely a "fact finding". The data in a qualitative descriptive research is collected qualitatively which is then analyzed quantitatively by using frequencies, percentages, averages, or other statistical analysis as proposed by (Nassaji, 2015) As what the researcher did in her previous study of the students' speech error analysis (Pravitasari, 2021), there were three major steps conducted in this study namely Preparation, Data Collection, and Data Analysis.

Preparation :
After reading related literature, the researchers explained the respondents about what they have to do during the data collection including teaching them about how to write an essay in the form of exposition text based on the given topic.

Data Collection
In this step, the respondents reported their research findings in the form of an exposition text and submit it in their google classroom account

Data Analysis
The data collected was then analyzed by using Corder's set of procedures for conducting error analysis which consists three major stages: recognition, description, and explanation of errors. These three stages were subsequently elaborated by Shridar (Fauziati, 2009b) into the following steps: 1. Collection of data 2. Identification of errors (labeling with varying degrees of precision depending on the linguistic sophistication brought to bear on the task with respect to exact nature of deviation, e.g. dangling prepositions, anomalous sequences of tenses, etc; 3. Classification into error types (e.g. errors of agreement, articles, verb forms, etc.); 4. Statement of relative frequency of error types 5. Identification of the areas of difficulty in the target language; and 6. Therapy of remedial lessons.

Error Description or Classification
There are a number of classification systems that have been used in error analysis. The one used in this research was that proposed by James in Fauziati, 2009. According to him, there are four types of error: omission, addition, misformation, and misordering.
Omission is a type of error which is characterized by the absence of an item that must appear in a well-formed utterance. As an example, we know that morphemes or words can be distinguished into two classes: content words (e.g. nouns, verbs, adjectives, and adverbs) and grammatical words such as noun and verb inflections (-s, -ed, -ing), the article (a, the, an), verb auxiliaries (is, will, can, may, etc.), and propositions (in, on, at, etc.). Language learners generally omit grammatical morphemes much more frequently than content words such as in John a new student, He speak English well, and He eating breakfast.
Addition is a type of errors which are characterized by the presence of an item, which should otherwise not appear in a well-formed utterance.
Misformation errors are those characterized by "the use of the wrong form of a structure or morpheme". There are three types of misformation which have been frequently reported in the literature, namely, regularization (overlooking exception and spreading rules to domains where they do not apply such as runned, womans, hitted, etc.), archi-forms (selection of one member of a class of forms to represent others in the class, for example, out of the set this/that/those/these the learner might use only one that), and alternating forms (the use of archi-forms often gives way to the apparently fairly free alternation of various members of a class with each other. For example, the learner uses he for she, him for her, they for it, her for she, etc.
Misordering errors are characterized by incorrect placement of a morpheme or group of morphemes in an utterance, such as in What daddy is doing and I don't know what is it.

Error Correction
One of the purpose of conduction error analysis (EA) is "to identify the principles which should guide effective error correction" (James in Fauziati, 2009). EA proponents believe that EA is able to help the teachers to devise remedial lessons and exercises, that is, error corrections or treatments which can help learners better learn the target language so that they can develop their interlanguage system.

RESULT
Based on the research conducted by the researcher, the data collected from the 30 respondents can be seen in the following table of error distribution as proposed by James (1998).  Table 1 shows four grammatical error classifications made by the 30 respondents. There are 200 errors in total with the biggest one is the misformation which contributed as many as 105 errors or 52.5% to the whole.
Here are some samples of the misformation errors found in the respondents' essay: 1. The result as tabulated in Table 1 was that there were 10 respondents saying that they are satisfying enough on their facilities and service. Correction : (1) Either the word 'satisfying' or 'satisfied' may belong to adjective. However, there is a distinctive usage and meaning between both of them. Adjective 'satisfying' is an active adjective used to express a condition in which it can satisfy others or make others satisfied. In contrast, the word 'satisfied' is a passive adjective in which it is usually used to express our feeling toward something. Example : The service of the cafe is very satisying. All customers are satisfied.
(2) Idiom of 'satisfied' does not come with preposition 'on' but come with 'with'. adjective 'satisfied' (3) The to be 'are' should be 'were' since the event happened in the past time. In addition, it belongs to reported speech also. Therefore to report/say/describe activities in the past time must use second verb form. (4) Therefore the correct sentences then is : The result as tabulated in Table 1 was that there were 10 respondents saying that they were satisfied enough with their facilities and service 2. There are some suggestion from the customers to add some features like send money, recharge etoll balance, and etc. Correction : (1) When a verb comes after the verb 'like', it must be in the '-ing' form.
(2) The correct sentence is : here are some suggestion from the customers to add some features like sending money, recharging e-toll balance, and etc 3. I gave the questionnaire via google form, so the customers don't have to fill it out on the spot.

Correction :
(1) . The activity of filling out the form was taken place in the past time, hence the sentence must use second verb form. The auxiliary verb of 'don't' must be 'didn't'.
(2) The correct sentence is : I gave the questionnaire via google form, so the customers didn't have to fill it out on the spot 4. There are 23 respondents were satisfied with their skill and friendly. Correction : (1) . The mini research was conducted in the past time, therefore the sentence must use second verb form as well. The to be 'are' must be 'were'. (2) . Two words or more connected by conjunction "and, or, but" must be the same part of speech. In addition, possessive adjective of 'their' needs a noun after it. 'Their skill' is correct but their friendly is not since friendly is not a noun but adjective. (3) . The correct sentence is : There were 23 respondents were satisfied with their skill and friendliness.
5. In the service aspect, there are satisfied with their fast response.

Correction :
(1) . The word 'there' must be 'they' since this sentence needs a subject.
(2) . The sentence must be in the past form since the activity had taken place. Therefore, the to be 'are' must be 'were'. (3) . The correct sentence is : In the service aspect, they were satisfied with their fast response The second biggest grammatical error as seen from Table 1 is the Omission which contributed as many as 51 or 25.5% to the total number of the errors.
Here are some examples of omission errors found in the research data: 1. On facilities of room comfort, there 12 respondent said it was good and 8 respondent said it was very good Correction: (1) . A word order said to be a sentence is the one that least consists of subject and predicate. Therefore, the sentence above is not correct since the predicate is missing.
(2) . The word 'respondent' must also be in plural form since the number of the respondent is 8 or more than one. (3) . Correction : On facilities of room comfort, there were 12 respondents said it was good and 8 respondents said it was very good 2. This mini research purpose to measure the level of customer satisfaction toward AIZA olshop.

Correction:
(1) . There is one part of the sentence missing namely the predicate of 'was' before 'to measure'. (2) . Possessive word is also missing from the sentence above : 'research purpose' should be 'research's purpose'. So does 'customer satisfaction' should be 'customer's satisfaction.' (3) . The correct sentence then is : This mini research's purpose was to measure the level of customer's satisfaction toward AIZA online shop.
3. Can be seen from Table 1, an average of 16 respondents answered point 3.
(1) . Subject 'it' is missing from the sentence above.
(2) . There must be a conjunction 'that' to connect these two sentences into a clause.
(3) . The correct sentence is : It can be seen from Table 1, that an average of 16 respondents answered point 3.
4. On the aspect of facility parking, ABC Bumes very satisfying.

Correction:
(1) A predicate is missing from the sentence.
(2) The order of the noun phrase of 'facility parking' is incorrect. An adjective comes before a noun to make a noun phrase. (3) The correct sentence = On the aspect of parking facility, ABC Bumes is very satisfying 5. The result based on a small survey conducted the beginning of July 2021.

Correction:
(1) There must be a to be 'was' put after the subject 'The result' the sentences above.
(2) The writer also did not put a preposition 'on' before the adverb of time 'the beginning of July'. This may indicate that the writer mostly used his or her first language. (3) The correct sentence is 'The result was based on a small survey conducted on the beginning of July.
The third grammatical error as seen from Table 1 is the Misordering which contributed as many as 26 or 13% to the total number of the errors.
Here are some examples of misordering errors found in the research data: 1. From the result of the table data, it can be concluded that Portalpulsa is good quite in terms of service, outlet, and facilities.

Correction:
(1) . The order of a noun phrase is modifier + noun therefore. Therefore the sentence above is considered incorrect in 'table data' and 'quite good'. THye should be 'data table' and 'quite good'.
The correct sentence above is : From the result of the data table, it can be concluded that Portalpulas is quite good in terms of service, outlet, and facilities.
2. The aspect to be measured were its crew outlet on their appearance and hospitality, working hours, facility, and response in full shift 12 hour work. Correction: 1) Seen from the context meaning, the noun phrase of 'Crew outlet' is considered to be incorrect in its order since it refers to the employee's responsible for taking care of the outlet. Therefore, it should be outlet crew.
2) The order of hour work is incorrect indeed. It should be working hours instead of hour work.
Therefore it bears three error types namely misordering, omission and addition. 3) There is an addition error in part of 'full shift 12 hour work'. 'Full shift' has the same meaning as 12 working hours. Therefore using 'full shift' or '12 working hours' is enough in stead of using full shift of 12 working hours altogether to avoid redundancy. 4) Finally, the correct sentence is that 'the aspect to be measured were its outlet crew on their appearance and hospitality, working hours, facility, and response in 12 work or the aspect to be measured were its crew outlet on their appearance and hospitality, working hours, facility, and response in full shift.
3. It is just a matter of how the owner can better serve the buyers. 1) A sentence consist of Subject -Predicate -Complement (object/noun -adverb). Therefore the sentence above does not match English structure properly. The adverb 'better' should be at the end of the sentence. 2) The correct sentence is 'It is just a matter of how the owner can serve the buyers better.' 4. The should have a bigger area parking. Correction: 1) The noun phrase of area parking is incorrect in its order since the root is area while the modifier is parking. The order of a noun phrase is a modifier + root (main noun) 2) The correct sentence is that 'they should have a bigger parking area'.

Isnan motorcycle workshop is enough crowded..
Correction: 1) The word 'enough' and 'quite' have the same meaning in Indonesian. They are interchangeable indeed. However, there is a distinctive usage. THe word 'enough' is used after an adjective being modified while the word 'quite' is before it. 2) The correct sentence is: Isnan motorcycle workshop is quite crowded. It is also acceptable to say 'Isnan Motorcycle Workshop is crowded enough.
The last grammatical error type as shown in Table 1 is addition which appeared only 18 times or contributed as many as 9% to the total.
Several samples of addition errors that have been found by the researcher are as follows : 1. Free internet access makes this place recommended place to visit.
Correction: 1) It is unnecessary to add the word 'place' after the word recommended because it has been mentioned in advance. 2) The correct sentence is 'Free internet access makes this place recommended to visit. 2. They provide suggestions so that Khair Garage can be more professional again.
Correction 1) It is incorrect to add 'again' since there has been 'more' used in advanced.
2) The correct sentence is 'They provide suggestions so that Khair Garage can be more professional. 3. However, some suggestions came up to make the service more better.
Correction 1) As it is found on number 2, it is redundant to add 'more' before the word 'better'.
2) The correct sentence is 'However, some suggestions came up to make the service better. 4. This study was done because to measure the level of customers' satisfaction with the service of Tresno Sehat Drug Store.

Correction:
1) It is unnecessary to use 'because' since the infinitive 'to measure' acts as an adverb of purpose.
2) The correct sentence is 'This study was done to measure the level of customers' satisfaction with the service of Tresno Sehat Drugstore. 5. This mini research was done to determine the customers' satisfaction of using MM Trans and to improve and improve its services. Correction 1) The use of double verb 'to improve' is redundant.
2) The correct sentence is this mini research was done to determine the customers' satisfaction of using MM Trans and improve its services'.

DISCUSSION
As illustrated above, the findings show that the most grammatical error made by the respondents was the misformation which contributed 52.5% (105 errors) to the total. It indicates that the respondents have lack of understanding on the use of proper form of English grammar such as regularization, archi-forms, and alternating forms. We can see the regularization errors on sample no. 3 in which the writer incorrectly used 'don't' instead of 'didn't' and sample no. 4 in which the writer incorrectly used 'are' instead of 'were'. Meanwhile, the respondents also made errors in alternating forms as shown on sample no. 5 in which the writer incorrectly wrote 'there' instead of 'they'. The finding also shows that the respondents have insufficient mastery on the word forms like noun, adjective, verb, adverb and so on as seen on sample no.2 in which the respondent used 'send' instead of 'sending', 'charge' in stead of 'charging' after verb 'like'.
The same result was found by (Mustafa et al., 2016) in their research which resulted in 72% of the total grammatical error was misformation. This error appear since the respondents are mostly still influenced by their first language, Indonesian. In addition, (Sermsook et al., 2017) mentioned in their research that this error types appeared due to the limited knowledge of the target language. Hence, they had a tendency to use their knowledge of their first language.
The same thing also works for the Omission error type in which it indicates that the respondents have limited understanding on the content words especially verb and grammatical words especially auxiliary. They did not realize that the role of auxiliaries is very significant to connect the words in order to be a properly standardized English sentence. The result also indicates that the respondents did not have enough mastery on derivative in which most words have its own form when they are used as a noun, verb, adverb, or adjective, and whether it is single or plural. This is in line with what (Octaviani, 2021) found in their research in which omission was their first biggest grammatical error found in their students.
For the misordering error sample data, we can see that the errors appeared due to the respondents' lack of understanding on the English grammar especially the Noun Phrase order in which it should Determiner + Modifier/Adjective + Noun or Determiner + Noun + Noun which acts as a modifier. However the respondents just simply wrote the words as the way they are; 'table data, crew outlet, hour work, area parking' instead of 'data table, outlet crew, working hour, and parking area'. In addition, the respondents also misordered the adjective phrase of 'good quite' in sample no. 1 in which it should be 'quite good' as the rule for the Adjective Phrase is Adverb + Adjective. Another respondent misordered 'enough good' instead of 'good enough'. Such misordering errors appeared since the respondents were mostly still influenced by their first language, Indonesian, in which there is no rule for a noun phrase, adjective phrase, and so on. The same thing happened on the Addition error type in which it is also obvious that the respondents just wrote as what the way they understood in their first language and ignored, or even knew nothing about the target language's grammar rules. For example to say ' lebih baik' they just wrote 'more better' .
From the discussion above, the researcher conclude that Error Analysis (EA) helps teacher to find out the students' precise lack of the target language's grammar, therefore several treatments or learning activities can be properly conducted in order to make the errors become fewer and fewer. This is in line with what James, 1998, stated that one of the purpose of conducting error analysis is "to identify the principles which guide effective error correction" (Fauziati, 2009b) .

Error Correction
After carrying out the EA on the students' essay and finding out their exact lack of the target language's acquisition, the researcher carried out a treatment (error correction process) in order to increase the students understanding of the target language so that their errors became fewer and fewer. It was done by returning the students' work, showing the error data resulted from the EA, giving more explanation and discussion about every error which were found, asking the students to write another essay based on the similar topic with the previous one, having pair work discussion and finally asking the students to submit their essay and analyzing the works. The result of the treatment on the error correction (remedial activity) can be seen in the following  Table 2 shows that the grammatical errors made by the respondents after the treatment (error correction) process was decreasing. The researcher believes that if more treatments are given, the fewer the errors made by the respondents will be.

CONCLUSION
There are several things to be concluded from the research findings: Among the four types of grammatical error proposed by James in Fauziati, 2009, the most frequent error found in this research is the Misformation (as many as 105 errors or 52,5%). The second one is the Omission (51 or 25.5%), followed by the Misordering (26 or 13%) on number 3 and finally the last error is the Addition (13 or 9%). There is a significant decreasing on the students' grammatical errors after an error correction was given: Misformation (as many as 56 errors or 60,8%). The second one is the Omission (21 or 22.9%), followed by the Addition (8 or 8.7%) on number 3 and finally on the last number is the Misordering (7 or 7.6%).
This research found out that there are two major causes of the errors made by the students of STMIK Sinar Nusantara was their lack of understanding on the target language's grammar rules and their tendency to use their first language automatically. This in in line with what Napitupulu, (2017) found in his research. However, after an error correction treatment was given and the students were asked to make another essay based on similar topic with the previous one, the result is quite satisfying. The number of the errors become decreasing. Therefore, the researcher believes that if the treatments were given more often, the errors will be decreasing and decreasing and finally disappear.
Finally, there are still many other aspects that can be done to generate researches on grammatical errors such as analyzing the effectiveness of certain instrument or teaching techniques used in class to decrease the errors that have been found out. Moreover, there should also be a research on studying the materials and the syllabus given in the class to know whether they also contribute the errors made by the students while using their English.