EFL learners' difficulties in the structure and written expression section of TOEFL test in an Indonesian university

This study examined the English for Foreign Language (EFL) learners‟ difficult topics in the structure and written expression section of the TOEFL Prediction Test, and reasons why they consider that these topics were difficult. A mixed-method research design was used in this study. Fifteen participants were selected through a purposive sampling mechanism from the seventh-semester students of the Department of English Language Education, Universitas Islam Negeri Ar-Raniry who have participated in the TOEFL Prediction Test to identify the difficult topics they encountered. Then, the semistructured interviews were conducted with six underachieving student‟s participants with the most recorded errors made in the test to know the reasons behind their difficulties. Findings indicated that students encountered difficulties mostly when * Corresponding author Saiful Akmal, Muhammad Nur Akbar Rasyid, Yuliar Masna & Cut Natasha Soraya Englisia: Journal of Language, Education, and Humanities | Vol.7, No.2, May 2020 | 157 dealing with determiners, conjunctions, adjective clauses, apposition phrases, and reduced clauses in the structure section. Meanwhile, adverb connectors, subject-verb agreement, and clause of concession, relative clause, and quantifier are the difficulties they encountered in the written expression section of the test. Thus, the findings also revealed several factors identified as the reasons behind those difficulties, namely lack of practice, grammar incompetence, vocabulary shortage, time management, and low self-confidence. Given the significant impact of this study, we suggested that the lectures and English departments should address these difficulties. It is crucial that the focus of the courses related to grammar and EFL proficiency tests be incorporated into the syllabus.


INTRODUCTION
The rapid growth of test of English as a Foreign Language (TOEFL) and its widespread uses in higher education, such as introducing TOEFL in curriculum, using TOEFL score as university admission or graduation, demand more discussions to understand and overcome English for foreign learners' difficulties in answering the test. In this light, Abboud and Hussein (2011, p.111) reported that, "more than 5,000 colleges, universities, and licensing agencies in 90 countries accept the TOEFL scores". In Asian context, the international spread of English and its proficiency tests, like TOEFL, has become part of globalization phenomena. This has paved the way for English to be used more often and considered as an emerging "Asian language" (Hamid & Nguyen, 2016).
Such developments, to some extent, were dominantly influenced by the language foreign policy imposed in international education. This phenomenon has forced the national or domestic language proficiency centres to be able to connect to the global cultures and identities . Because a language, as Hamid and Kirkpatrick (2016) put forward, is one aspect of national resources that needs to be properly addressed and managed. Some studies even argued that the introduction of English in education in Asian countries could no longer be considered only as a medium of instruction (Dang, Nguyen,&Le, 2013), yet it could also be linked to the region economic growth (Hayes, 2017), and the use of English as lingua franca .
In Southeast Asian context, Todd and Shih (2013) assess the use of English proficiency in two categories: the expanding circle of English and the outer circle of English. Most of Southeast Asian countries like Indonesia, Thailand, Vietnam, Myanmar, Laos, and Cambodia adopted English as a foreign language (EFL), and they were marked by relatively low score on international English tests. On the other hand, countries like Malaysia, Singapore, Brunei, and the Philippines fall into the outer circle of English, and they have been using English as their official language proven by the higher rating of English proficiency test's scores. They, moreover, add that the Southeast Asian countries almost have a similar heavily test-oriented education system. Indonesian higher education institutions, like their fellow Southeast Asian countries, have started to seriously introduce English proficiency tests. As a country in which English served as a foreign language (EFL), the use of English proficiency test, especially TOEFL, have been very popular amongst students and lecturers. These uses are various, ranging from TOEFL matriculation program (Herwandar, Safryono,& Haryono, 2012;Noviyenty, 2018), TOEFL preparation class (Masfufah, 2018), TOEFL school in online learning platform (Nimasari, Mufanti,&Gestanti, 2019), and graduation requirement (Aziz, 2016;Susanti, 2014). Given this massive use of TOEFL, the teaching and learning process, the syllabus construction, and the curriculum design are greatly influenced by the need to improve students' TOEFL score.
The uses of TOEFL for Acehnese universities have subsequently led to the implementation of TOEFL test requirement for educational purposes. In Syiah Kuala University, the minimum TOEFL score is used as a precondition for thesis examination (Kasim, 2016). Likewise, in Universitas Muhammadiyah Aceh, the TOEFL prediction test is required for undergraduate students (Netta & Trisnawati, 2020). Similarly, Samudra University in Langsa applies the same regulation, obligating students to take TOEFL test before graduation (Devira & Makhroji, 2017).
In recent times, students at UIN Ar-Raniry Banda Aceh must take TOEFL course and prediction test as pre-requirement before their bachelor thesis exam. The university management has set the minimum score of the TOEFL prediction score is Saiful Akmal, Muhammad Nur Akbar Rasyid, Yuliar Masna & Cut Natasha Soraya Englisia Vol. 7, No. 2, MAY 2020 | 167 475 for English department students, and 400 for non-English department students prior to graduation (Circular No: 756/Un.08/R/PP.00.9/02/2017). In response to the global call, national and international competition, from 2017 to 2020, the Department of English Language Education has gradually raised the minimum TOEFL score requirement from 470 to 500 for graduation requirement.
Prior Studies Huang (2006) piloted an interesting assessment on North American ESL students in their campuses. It showed that Chinese students experienced difficulties in understanding academic lectures and delivering presentations despite their high TOEFL scores. In a similar tone, Takanashi (2004) conducted a research and discovered that Japanese students repeatedly feel difficult to attain communicative English skills at school. Although most Japanese students were very eager to learn English, their TOEFL score are still low. The same case occurred in the context of Indonesian universities. According to Samad and Fitriani (2016), Indonesian students at Syiah Kuala University Banda Aceh also unsuccessfully fell short to meet the minimum TOEFL Score. They said that among 1916 TOEFL test-participants at Syiah Kuala University, only 53 students achieved score between 450 and 497.
Having said that students can experience difficulties in answering TOEFL test, the three sections of (1) listening, (2) structure and written expression, and (3) reading comprehension provided in TOEFL test can pose various levels of difficulties for students. Depending on many factors, each section requires different skills to achieve the desired TOEFL score. For example, an investigation conducted by Hambali (2008) disclosed that students still have many problems in answering the structure and written expression section of the TOEFL tests, namely: identifying verbs, identifying the correct words order of subject and verb in noun clause, and identifying words with derivation and function. In line with this, Mahmud (2014) also confirmed that most test-takers generally had problems in answering structure and written expression sections of the TOEFL test.
Those prior researches have proven that students' difficulty in answering the structure and written expression section of TOEFL test is apparent. Test takers think there are a lot of rules or sentence patterns that should be remembered and memorized. Ananda (2016) found that the most-recurring problem in section two of TOEFL test is questions on inversions and the least-frequent one is the use of verbs.
Another study believes that students are troubled with questions about subject-verb agreement in TOEFL (Atmojo, 2018). However, those problems are not only inevitable for the non-English Department students but also encountered by the English Department students.
Similar with such previous research, even though English Department students learned all the grammar courses or even passed the English language proficiency subjects, they remained facing difficulties when answering the structure section in TOEFL test. Thus, based on this problem, it needs to conduct a research and find the reasons why the English students are still difficult to answer TOEFL test, especially in structure and written section. Although these above mentioned current studies were valuable, they were mostly embarked on the general problems, challenges, and difficulties in answering the complete TOEFL test. It provides little information on the specific difficulties experienced by English department students at Islamic higher education institutions context in Indonesia. This present study therefore seeks to find out (1) the difficult topics of EFL learners in completing the structure section of TOEFL test, and (2) the factors behind EFL learners' difficulties in answering the structure section of TOEFL test in the given context.

LITERATURE REVIEW
Test of English as a Foreign Language or also known as TOEFL is a standardized test recognizing worldwide. Warfield, Laribee, and Geyer, (2013) argue that TOEFL is acknowledged globally and universally. It is designed to measure people's English language ability to see whether they are good enough to take a course at a university or college in an English-speaking country. This test has become one of the requisites that must be met if someone wants to continue his/her study abroad. TOEFL itself has different types such as Paper-based, Computerbased, and Internet-based test. Brown (2005) states that TOEFL is a way for assessing someone's skill, familiarity, or enactment in a given domain, i.e. listening, structure, written expression and reading comprehension. However, for the purpose of this present study, we focus only on the second section of the structure and written expression section in the Institutional Testing Program (ITP) TOEFL test.

Structure and Written Expressions
Structure and written expression is the second section in TOEFL test that focuses on written English. This segment consists of two parts: (A): Structure and (B): Written expression in which the test-takers are given 25 minutes time to answer the total 40 questions. Part A has fifteen multiple choice questions marked with A, B, C and D probable answers as part of sentence completion, whereas the second part has twenty five questions. In the second part, the test-takers will encounter erroranalysis type of questions in each sentence. Moreover, structure and written expression is not only existed in ITP TOEFL but also in PBT TOEFL. The difference between two is that PBT TOEFL has one section namely Test of Written English or known as TWE. Philips (2001) classifies the structure and written expression section into the following topics; there are at least 10 main topics and it can be seen in the table below: Table 1. Subsequently, word forms are related to the actual forms of nouns, verbs, pronouns, adjective and adverbs. The test-takers are given an incorrect form of the word and asked to identify the correct form. Afterwards, reduced clauses denote to the shortened form of a relative clause, which transforms the subject of a sentence.

Main Topics in Section Two of TOEFL Test
Reduced relative clauses can change the subject, but not the object, both in the active and passive voice sentences.
Connectors are words joining two words, phrases, clauses, and sentences together. The connectors can be categorised into four types: preposition, conjunction, and transition. The test takers are tested to recognise the correct use of connectors in words, phrases, clauses or sentences. Gerunds and infinitives are from time to time referred to as verb accompaniments. A gerund is a noun made from a verb by placing "ing" at the end of a word, while infinitive is the "to" form the verb.
A comparison expresses the degree of contrast by using adjectives and adverbs. It is also branded as comparative and superlative. The test takers are questioned to find the correct comparison to represent the likeness of the dissimilar things in a sentence. Then, clause-formation evaluates the format of dependent or independent clauses such as noun clause, adjective clause, adverb clause, negation clause, and the likes. Parallel structure is the usage of the equivalent grammatical structures for connected notions of equal importance. Related ideas of equal importance frequently happen in the form of list that are linked by conjunction, such as and, but, and or. Finally, redundancy is understood as the use of two or more words that say the similar thing. The test-takers are demanded to find the different words which have the same meaning.

Question' Types in Section Two of TOEFL Test
This section has two types of questions with special track for each. These types of questions will be explained below:

Structure
The test-takers must choose one answer that best completes the sentence.
Such sentences and their answers are mentioned (Abboud & Husein, 2011 p.124) Example: Engineers ______ for work on the new space program.
a. Necessary b. Are needed c. Hopefully

d. Next month
In this example you should notice immediately that the sentence has a subject engineers and it is not verb. Because the option B, are needed, is a verb, it is the most suitable answer for this question. Options A, C, and D, are not verb; therefore they are not correct answers. (Philips, 2011 p. 200).

Written Expression
The participants will see that each sentence has four underlined words or phrases. They must find one underlined choice that must be corrected in order for the sentence to be accurate. For example: Representative democracy seemed evolve simultaneously during A B C the eighteenth and nineteenth centuries in Britain, Europe, and the United States. D The correct choice is (A) (Sharpe, 2004 p. 112)

EFL Learners and Structure and Written Expression TOEFL Test
According to Kempa (1991) difficulty is the situation where the students fail to comprehend and understand a concepts or idea. These situations are caused by some factors such as lack of students' motivation (Wulandari, 2016), learning strategies (Mahmud, 2014;Samad, Jannah & Fitriani, 2017), vocabulary (Saadian& Bagheri, 2014), a complexity of pragmatic and grammatical competence (Xu, Case, &Wang, 2009) and lack of exposure to the foreign language atmosphere (Alrabai, 2016). Ananda (2016) revealed that students agreed that part B of TOEFL test is more difficult than part A. This finding is confirmed and put into more detailed by Fitri (2018) revealing that parallel structure is the most difficult questions in section B of written expression of TOEFL test. Given the complexities of grammar problems in TOEFL Test for EFL learners, some proposed a variety of approach for remedial. (2015) for instance, introduce a corpus and grammatical browsing mechanism to tackle grammar problem for their students,

Chujo, Oghigian and Akasegawa
whereas Seyedtajaddini (2014) uses audio input impact to enhance EFL learners' grammar learning. Using focused and unfocused, direct and indirect written corrective feedback on EFL learners' grammatical precision was also one of the attempts to address these complexities (Farrokhi & Sattarpour, 2011;Hosseiny, 2014).

METHODOLOGY
The present study applied a mixed method design in which the data were collected by using test and semi-structured interviews (Kawachi-Furlan, Amorim,& Finardi, 2017). The quantitative part of the study design used simple statistics form to analyse the TOEFL test results, and therefore it can be categorized as descriptive statistics design (Guetterman, 2015;Woods, Fletcher,& Hughes, 1986). Test, distributed to 15 students, was used in this study to find the most difficult topic in the structure and written expression. Those students were selected purposively based on their ITP TOEFL score. All participants were the seventh-semester students of English The participants had 25 minutes to complete the test.
As interview is used in most qualitative research no matter what qualitative strategies are being used (Creswell, 2007). Griffee (2012) sees interview as personto-person conversation where its purpose is to find and create meaningful data which have to be collected and analysed. There are three common types of interviews including structured, semi-structured, and unstructured interviews (Stuckey, 2013). Therefore, the second part of data collection used to denote the qualitative traits in this study was semi-structured interviews. This was used to know why the topics were regarded difficult for them and given to six underperforming students with the lowest scores in TOEFL structure and written expression test.
Resembling to the structured interview, this type of interview also has a rough draft of a topic question. Although there is a set of guided questions, the response from participants allows the flexibility to ask more enhanced questions (Stuckey, 2013). Semi-structured interviews also allow informants to express their views in their own terms and can provide reliable and comparable qualitative data (Griffee, 2012).There are some reasons why the interview was chosen as data collection technique in this study. Firstly, as stated by Gray (2004), by doing interview, there is a need to attain highly personalized data. Secondly, as Adhabi and Christina (2017) explain that the advantage of interview is that we can pinpoint on delicate subjects.
In addition, the strength of interview is "researcher can prompt and probe deeper in the given situation" (Annabel, 2005 p.6).

Technique of Data Analysis
The data from this research were analysed with different stage using descriptive statistics method (Woods, Fletcher & Hughes, 1986). For the test, answer sheets from the students were checked and classified into correct and incorrect answers.
For interview, the data were analysed by using coding procedure. As Miles and Huberman (2013, p.56) state that "codes are tags or labels for assigning units of meaning to the descriptive or inferential information compiled during a study".
Code is analytical process in both qualitative and quantitative research used to analyse the data by labelling and symbolizing them into theme to descriptive information.
In this study, open coding was used to analyse the data of interview. Esterberg (2002, p.158) suggest that open coding is "a process where you work intensively with your data, line by line, identify themes and categorize that seem of interest". First, we started writing the document transcript from interview's audio recorder. Then, the data were read to gain the information and ideas from the participants. Next, the data were coded, categorized them into themes, and interpreted.

Section in TOEFL Test
This research examined English learners' difficulties toward TOEFL in structure and written expression section. We used a test, only for TOEFL structure and written expression section as the instrument for data collection to figure out the difficult topics faced by English learners in structure and written expression section.  Table 2 shows the result of test in structure and written section in part A. The highest percentages of incorrect answer are question number 11 and 14 with 100%, followed by question 3 and 12 with 80% and question number 8 and 13 with 73%.
The medium percentage of incorrect answer is seen in number 10 with 53%.

Structure -Part A Determiner and Conjunction
The following captions are the questions of number 11 and 14 where all participants made mistake. These questions discuss about determiner and conjunction topic. As for question number 11, 100% participants failed to answer this question.
This means all participants still completely did not understand about the use of determiner. The test-takers said that the word "so" was appropriate with the sentence in the question. Besides, they did not realize that the question was actually questioning about the use of a determiner in particular context.

Subordinating Conjunction and Adjective Clause
The next most difficult questions are on subordinating conjunction and adjective clause in questions number 3 and 12 with a total 80% of incorrect answers.  The third most difficult subject of structure section is apposition phrase and reduced clause with question 8 and 13 with (73 %).

Adverb Connectors
Here, the most ratios of participants' incorrect answers are understood in question number 37 with 100%. This question discusses about adverb connectors

Subject-Verb Agreement & Clause of Concession
Then, it is followed by number 24, 36, and 38 which were 93% participants failed to answer these questions correctly. These questions discuss about Subjectverb agreement, and clause of concession

Relatives Clause and Quantifier
Then, it is followed by number 30 and 32 that there were 87% participants did not answer these questions well. It is about Relatives Clause and quantifier. The question no. 30 was about relative clause and only 87 %, or two participants managed to correctly answer the question. Test-takers said that relative clause topic was hard to be analysed in the question.
It can be said from the findings that all participants made mistake in those topics because of several reasons. First, the participants did not understand well and did not realize about the topic in the questions. Second, the participants were nervous and confused during the test. The final reason was the ways of the questions were constructed in the test.
This result was supported by Hajri, Jufrizal, and Wahyuni (2018) stating that among 10 main items assessed in structure and written expression subtest, three items were categorized as difficult items: reduced clause, parallel structure, and format of negation. Also, four items were categorized as medium items: singular and plural, verb-agreement, gerund and infinitive, and reduced in active sentence.

B. Factors Affecting English Learners' Difficulties in Answering Structure Section in TOEFL Test
Here, the result of interview from six the participants is discussed. They were selected based on the most-recorded wrong answers of the test to explore the reason of EFL learners' difficulties in answering structure section in TOEFL. Findings of the interview indicated that there were some factors affecting the difficulties for English learners' in answering structure and written expression section. In relation to this concern, other participants said: Most of the participants approved that the main factor was due to lack of practice or preparation.

Lack of Practice
Based on their answer, it can be concluded that practice is extremely important for participants, particularly English Department students either in the classroom or outside the classroom to prepare them for the TOEFL.

Grammar Incompetence
The second factor that also takes the big concern affecting the difficulties of English learners is due to lack of grammar skills. Most of the participants thought that they had bad skills in grammar and lack of grammar understanding. An example of the participant's response of grammar incompetence is shown below: Honestly, I don't really understand about grammar. My grammar is too bad. [IBR] In relation to this concern, other participants said: Yes, of course, when we learned grammar we only learned about the formula without understanding the using about grammar. Also in grammar we just learned simple sentence simple topics, but in TOEFL in the test the sentences (are) more

complex. [RZJ]
She believed that when she learned grammar course and took the TOEFL test, it was different level of questions and different level of sentence. In addition, another participant thought that the way the lecturer explained grammar course was the factor that made her poor in understanding grammar. The participant response is as following; "It seemed to me that my lecturer explained all grammar issues in grammar course very fast. It was very hard for me to understand them well". [SN] Based on the findings, it was clear that most participants still had problems on grammar even if they were already learned in grammar in their first and second year.

Shortage of Vocabulary
The majority of the test-takers sanctioned that vocabulary is crucial in answering TOEFL test. In fact, they said that they still encountered difficulties in answering TOEFL due to lack of vocabulary mastery. It appeared when some learners found unfamiliar vocabulary during the test. For example: In my opinion, we seldom learn about TOEFL test. The biggest problem is because we don't learn or practice it. Then, lack of vocabulary also. Sometimes, the sentences I don't know the meaning. As we know, in TOEFL the vocabulary is higher and it talked about science, economic etc. [IBR] Another factor is vocabulary. It is also important. Because when I know about the meaning of the sentence, I can know the best answer. For example in error analysis if someone doesn't know the meaning it's hard to choose the correct answer. [ADL] Based on the findings, the participants specified the difficulty in term of lack of vocabulary. All participants emphasized the role of vocabulary in answering TOEFL test. As noted by IBR, vocabulary used in TOEFL is higher and more difficult than those used in usual reading material. Also, vocabulary in TOEFL utilizes all of topics such as economic, social, education, health, geography etc.

Time Management
Based on the interview result, this study revealed that most participants' response still concerned on lack of time. Most of the learners stated that lack of time made them failed in TOEFL test. Some participants' responses are shown as following: And also, the time is also not enough for 40  The findings indicated that all of test-takers faced difficulties in answering TOEFL due to lack of time. They revealed the time was so limited to answer 40 questions in 25 minutes. Also, most participants spent more time in answering a question. The result of interview figured out that the majority of English Department students could not manage the time wisely during the test.

Low Self-Confidence
Most participants claimed that they were not feeling confident when taking the TOEFL test. In addition, some said they felt distracted when choosing the answer.
The participants' responses toward this aspect are described below: I think the difficult one is less of preparation. Because when I follow the test at the first time without preparation, I feel so confused and my score is under 400.
Then, I also feel nervous when taking the test, so that's why it makes me difficult.
Also, the time is not enough for me to answer the questions. [RZJ] And also, I'm a little bit nervous when I take a test. It's different when you learn and when you take the test, it's so different feeling. [ARD] As the learners' responses above, they indicated that most of the participants showed their anxiety during TOEFL test. Anxiety is a common phenomenon for every participant taking the test. However, anxiety can be resolved by doing a lot of practices and taking the TOEFL prediction test repeatedly.

DISCUSSION
From the data above, there were at least three levels of difficulties: very difficult, difficult, and medium which participants encountered in the TOEFL test, and there were five major causes of these difficulties. Those topics are discussed in the following sections. The first section elaborates the categories of difficulty found from the participants' answer sheets. Then afterwards, the factors causing those difficulties are analyzed.

Determiner
Question number 11, 100% participants failed to answer this question. This question was considered very difficult, which means all participants still did not completely understand about the use of determiner. The participants said that they thought that the word "so" was appropriate with the sentence in the question.
Besides, they did not realize that the question was in fact questioning about determiner. In the same light, Snape (2006) believes that non-native English speakers may have problems in acquisition, and therefore may have similar problems in noticing determiner.

Compound-adjective.
This topic appeared in the question number 37 in which 100% participants failed to respond this question. This question was also considered very difficult, meaning that all participants still did not recognize about the compound-adjective.
All of the test-takers chose wrong answer in this question. They admitted that they did not understand what the question discussed about. Also, they were distracted by feeling nervous when answering the question. Besides, the question number 37 has difficult level of difficulty. Thus, they could not answer the question. In a similar vein, Dehham (2014) analyzes Iraqi EFL learners in using English compound adjectives in which she found that they had serious difficulties at the level of producing compound adjective.

Paralellism
The next topic is parallelism. This topic is presented in question 24 in which 93% participants failed to find the correct answer. It pointed out that they did not know the concept of parallelism. It happened because the test-takers were not aware that they had to make sentences which had the same structure. Besides, they did not realize the question asked about parallelism and they said the way of the question constructed was difficult. This point is comparable to Ahmed's findings (2010) that EFL learners encountered difficulties in differentiating repetition and parallelism in a sentence. It was concluded that lack of coherence and cohesion understanding of English may be the main reason of such occurrences (Ahmed, 2010).

Conjunction
There are  still did not understand this basic topic very well. Despite the fact that conjunction is considered by many as one of the easiest topic in grammar, participants still made mistakes here. Some participants' revealed that they did not realize those questions were about conjunction. Although this study did not specifically look into details what type of conjunction errors were made by the participants, this finding apparently confirms Darweesh and Kadhim's analysis (2016). They said that conjunction problems often occur between EFL learners, especially because they did not entirely know the real meaning and specific uses of different conjunction types

Quantifier
Thus, the topic is about quantifier is also dubbed as difficult. It was represented in question 32 in which most participants' failed to get the correct answer with 87% percentage of error. Almost all participants could not answer this item correctly. Quantifier is also considered as the simple topic in grammar course.
Almost all participants selected the wrong answer for this part. There are two possible reasons why the participants made mistake here. First, the participants did not realize about the topic in the questions. In the same way, as Subramanian and Khan (2016) argue, explicit instruction might be recommended to overcome such problems. Second, the participants were nervous and confused during the test, as also asserted by Ebadi and Khakzar (2014). The last reason was the ways of the questions constructed in the test. This is partly because the nature of TOEFL test is lacking the communicative competence, unlike what we have seen in IELTS for example (Chapelle, et.al. 2011).

Subject-verb agreement
The first topic is Subject Verb Agreement (SVA). Then, question number 12 with 80% success rate is considered not so difficult or medium. Hence, only three participants passed this question. Most of the test-takers selected wrong answers about subject-verb agreement. It was admitted that the questions on Subject Verb Agreement (SVA) were truly difficult to answer in a short time (Alshammari, 2013).

Reduced Clause
Here, 73% participants' failed to answer the question, whereas the level of difficulty in this number is medium in question number 8 for example. It showed that the majority of the participants did not fully understand about Reduced Clause. In question no. 30, only two participants correctly answered the question (87%). This result was supported by Hajri (2018) who stated that among 10 main items assessed in structure and written expression subtest, three items were categorized as difficult items: reduced clause, parallel structure, and format of negation.
Following the discussion on the categorization of participants' level difficulty in TOEFL test, the subsequent elaboration on the factors that cause participants' difficulties in TOEFL test is described. The first main factor affecting English learners' difficulties was due to the lack of practice of grammar, especially for TOEFL test.
They said that they did not practice well in learning TOEFL when they wanted to take the TOEFL test. They took the test without any practices and preparation. The result was supported by Pan (2009) stating that practice is needed in order to make the students familiar with the questions, so they know how to prepare themselves for the test.
The second factor was grammar incompetence. All participants said they did not master the proper grammar skills or deficient knowledge on grammar concepts.
They also said they knew the theory on the grammar but when they were confronted with TOEFL questions, they did not know how to solve it. This result is reinforced by Mahmud's research (2014), indicating participants failed in answering the questions because they had low English grammar understanding.
Furthermore, the third factor causing the problem is insufficient vocabulary.
All participants agreed that vocabulary was very helpful to understand the question.
The participants could not answer the questions because they did not understand the meaning of the words. Also, the questions consisted of most unfamiliar vocabulary.
According to Alqahtani (2013) and Gu (2010), vocabulary is viewed as a crucial tool for second language learners and language development.
Then, the fourth factor is time limitation or time management. Participants said that they did not have enough time to answer those questions. They revealed that the time was incredibly limited for them to choose the answers of 40 questions struggled to answer the question in part B due to low self-confidence. This is particularly linked to the idea that one of the greatest psychological challenges for young language learners is self-confidence (Nikolov, 2016).
Apart from those five factors, we presupposed some other factors causing these participants' difficulties in TOEFL test. The fact is that joining the TOEFL training and preparation program does not always provide a final solution for participants to answer structure questions in TOEFL. This is proven by four out of six research participants having the problems with the TOEFL training program. The participants admitted that they did not get any improvement in understanding structure topics even after joining the training.
The participants said that it happened due to several factors. First, it was due to the lack of effectiveness in organizing the training class and creating conducive learning atmosphere. The staff of the Language Development Centre (LDC) arranged the class by combining the students from various majors, including mixing non-English students with the English department students. The participants revealed that this approach seemed to be less-effective. The lecturers, very often, also have to repeat the same explanation on the same basic topics causing some delays in the classroom to move forward and discuss the next topics for TOEFL preparation.
As a result the English department students will feel tedious and such lacklustre will disturb the whole process and test performance. These findings are seemingly against most studies on TOEFL preparation course (i.e. Nikolaieva, 2016;Erfani, 2012) that stressed the advantages of TOEFL preparation courses and classes for students. Second, the way lecturers convey the material was not too deep.
The participants said that some lecturers delivered the material in unstructured ways.
They argued that lecturers did not clearly explain in detail and offered lack of discussion time. It made the participants hard to understand the topic and eventually led to difficulties to answer those questions. In addition, the lecturers also rarely provided quick tricks or tips in answering the questions in the structure section.
However, communicative language teaching is believed to have improved students' motivation and of course, their final performance on proficiency test (Martono, 2013).

CONCLUSION
Having analysed the data in the previous section of findings, several conclusions can be drawn. The findings of the present study suggested that the students of the English Language Education Department urgently need to practice a lot in structure and written expression section in the TOEFL test. By practicing a lot, they will not only acquire knowledge on grammar, but also in terms of time management. The students should also be given short training and detailed orientation in TOEFL particularly in structure section, especially for students who have passed the Introduction to English Proficiency Test course. This is performed to strengthen their understanding of TOEFL particularly in structure and written section.
The training can be carried out, at least, in one month time with eight meetings. In addition to this, English Language Education Department may provide a learning module containing materials about TOEFL structure and the exercises to facilitate students in understanding these difficult topics.
Subsequently, the existing TOEFL training program at LDC should organize the class more efficiently and effectively. For instance, the class between English and non-English Department students should be separated. As a result, they can be more focused on their learning and the class is well-organized.
On top of that, we suggest that future studies should be carried out to further explore this topic by involving a large number of samples considering our relatively small sample of participants. Secondly, this study does not cover all areas or aspects