The Problems of Summary Writing Encountered by Thai EFL Students: A Case Study of the Fourth Year English Major Students at Naresuan University

The purposes of this study were to examine the quality of summary writing and type of problems made by Thai EFL English major students and also elicit opinions regarding problems encountered by these EFL learners in summary writing. 67 pieces of summaries written by fourth year English major students who enrolled in a research report writing course (205426) in the first semester of academic year 2021 at Naresuan University were collected and analyzed. It was found that the quality of summary writing produced by these EFL writers was viewed as “fair”, and there were no significant differences among all dimensions. They were all regarded as “fair”. Each dimension revealed the main problems mostly encountered. Concerning their opinions about summary writing problems, they thought that they found it difficult to find the main ideas of the text and finally lost focus. In addition, having difficulty in choosing vocabulary to replace the words from the original source when paraphrasing and having little time for revising were also their main problems. Pedagogical implications for L2 learning and teaching summary writing are discussed.


Background of the Study
Summary writing skills are crucial and required in academic writing contexts (Delaney, 2008;Plakans, 2008;Ono, 2021;Yamanishi, Ono, & Hijikata, 2019). Particularly in university contexts, writing summaries can be assigned in terms of both writing-only tasks and integrated writing tasks (e.g., listening-to-write tasks and reading-to-write tasks) (Marshall, 2017;Ono, 2021;Plakans, 2008). Summary writing is a complex process since it requires writing learners to employ processes of comprehending a source text, condensing a text, and producing a new text (Kintsch & van Dijk, 1978). As a result, writing learners, especially English as foreign language (EFL) learners, encounter a number of problems while they are writing summaries. These problems, for example, are found in many previous research studies which showed that second language (L2) writers tended to copy source texts or paraphrase insufficiently, so-called Near Copy when they write summaries in English (Keck, 2006;Shi, 2004). Moreover, word substitution and reordering strategies are also used by L2 writers who have difficulty in summarizing and paraphrasing sentences from source texts (Choy & Lee, 2012;Liao & Tseng, 2010;Sun, 2009).
In integrated writing tasks, including summary writing, the abilities to read and write are crucial since they reflect the understanding about the source text, reducing content, and creating a new text in the L2 writer's own words. The summaries need to have only the main points and main supporting points without changing the meaning of the source text. One important writing skill that is relevant to summary writing is paraphrasing (Yamanishi, Ono, & Hijikata, 2019). Both summarizing and paraphrasing need condensation of information, but with different levels (Hirvela & Du, 2013). Therefore, paraphrasing plays an important role in summary writing as shown in previous studies (e.g., Keck, 2006;Shi, 2012;Shi & Dong, 2018). Yamanishi et al. (2019) noted "paraphrasing plays a vital role in summary writing; however, this skill is considerably difficult to master and teach due to its complex characteristics and the influence of writers' cultural and linguistic background" (p. 3).
To improve the quality of L2 writers' summary writing, there have been a number of assessment measures and scoring rubrics that have been developed and research on summary writing assessment that have been conducted (e.g., Becker, 2016;Educational Testing Service, 2002;Knoch, 2009;Plakans, 2015;Sawaki, 2019;Yamanishi et al., 2019;Yu, 2007). By exploring various dimensions of summary writing (e.g., organization, sentence formation, the main ideas of the source text, the quantity and quality of paraphrasing, and the use of a writer's own language), this aims to reflect the quality of summaries written by L2 writers and the problems of summary writing by L2 writers so that writing teachers can use them as a guideline for giving feedback and improving summary writing of their students.

Statement of the Problem
Summary writing plays a crucial role in many academic tasks. Particularly, it is always accompanied by paraphrasing and quoting when L2 writers have to do research-based writing since they are required to integrate and synthesize sources (Hirvela & Du, 2013). In many textbooks for teaching English writing (e.g., McCormack & Slaght, 2009;Oshima & Huge, 2006), they describe summarizing and paraphrasing as useful skills when L2 writers have to integrate and synthesize source texts in order to support their own purposes. In addition, these skills are also required when L2 writers write academic texts, such as essays, literature reviews, or research reports (McDonough, Crawford, & Vleeschauwer, 2014).
In integrated reading-to-write tasks, writing is always based on reading (Carson, 2001;Weigle, 2004;Weigle & Parker, 2012). L2 writers are required to read source texts, synthesize the texts, and use source texts as input for idea generation in order to create their responses to the tasks in their own words. However, when L2 writers write a summary from source texts, inappropriate textual borrowing can occur if they are not familiar with the source texts or cannot paraphrase the information they need appropriately, especially in a timed writing assessment (Weigle & Parker, 2012). L2 writers may have difficulty in writing information from source texts in their own words since they may have low reading comprehension skills (Plakans, 2009) and limited vocabulary knowledge (Baba, 2009). In addition, L2 writers who have lower English proficiency may produce written summaries that contain near copies of source text information (Cumming, Kantor, Baba, Erdosy, Eouanzoui, & James, 2005;Gebril & Plakans, 2009).
In order to improve the quality of L2 writers' summary writing, it is important to examine the quality of their summary writing in various dimensions so that writing teachers can see the problems L2 writers have and provide writing lessons that can improve their students' summary writing. Additionally, to my knowledge, there has not been much research examining both the quality of EFL writers' summary writing and their perceptions about their writing abilities and problems of summaries written by English major students in Thailand. Therefore, the researcher would like to examine both the quality of summary writing and EFL writers' perceptions towards problems encountered by them when they write summaries. The findings of this present study may result in more appropriate and effective lesson plans and teaching methods for summary writing. Moreover, it is hoped that the findings of this study will help writing teachers to design their teaching contents and activities in order to improve their students' writing to become better EFL writers.

Purposes of the Study
The researcher aimed to achieve the following goals:

1)
To examine the quality of summary writing produced by Thai EFL English major students

2)
To examine what types of problems made by Thai EFL English major students in summary writing are the most repeated

3)
To elicit opinions regarding problems encountered by Thai EFL English major students in summary writing

1)
What is the quality of summary writing produced by Thai EFL English major students?
2) What are the most repeated types of problems made by Thai EFL English major students in summary writing?

3)
What are the opinions regarding problems encountered by Thai EFL English major students in summary writing?

Summary Writing
"A summary is a condensation of the main ideas in an article. The length of a summary depends on the assignment, the length and complexity of the article and the audience" (Reid, 1988, p. 110), and it is an interactive and recursive process (Reid, 1988). Hult and Huckin (2008) pointed out that, in academic writing contexts, students are required to write summaries in various types of writing assignments. They have to write the main idea from source materials in academic papers and presentations. In addition, these summaries need to be written in their own words. Mulvaney and Jolliffe (2005) noted that, for a good characteristics of summary writing, it has to be shorter than the original text, contain the main ideas, have limited and important details, and does not include opinions, evaluations, and irrelevant or additional information that are not contained in the original source.
Summarizing is considered to be an activity that aims to design independent secondary texts based on information from original texts. It can represent the compression of textual information. It is a process that requires comprehension and analytical and synthetic processing in order to produce new texts (Tareva & Tarev, 2020). As stated by Dewi and Saputra (2021), "a summary is a condensed version of a text that expresses the text's central ideas by recorded expression, which is related to paraphrasing. It must be shorter than the original text, contain the text's core concept, and, where appropriate, use the recorded expression" (p. 128).

Summarizing, Paraphrasing, and Text Borrowing
In integrated reading-to-write tasks, L2 writers are required to read the original text, comprehend it, and use this information to write a shorter version of the new text in their own words. According to Devine (1988), the interaction between previous knowledge (schema) of the reader and the information available in the text itself plays a crucial role in creating meaning from the text the reader reads. This notion is also supported by Koda (2005) in that the reader's comprehension includes integral interaction between the text and the reader. As noted by Urquhart and Weir (1998), "texts do not have unitary meanings potentially accessible to all, they rather allow for variety in interpretations by different readers, governed by factors such as purpose, background knowledge, and the relationship established between the reader and the writer" (p. 112). Therefore, it can be said that the interpretation of each reader may vary and/or be different from the meaning that the writer wants to convey or another reader interprets.
When summarizing, paraphrasing is also employed by L2 writers. However, it is not an easy process that all L2 writers can master. There are many problems found in this process. As found in the studies by Keck (2006) and Keck (2014), L2 writers tend to have a problem with paraphrasing since they tend to nearly copy information from the original text. This practice is in line with the findings of Gebril and Plakans's (2016) study which indicates that textual borrowing has effects on lexical diversity. L2 writers tend to borrow words from the original source when they do integrated reading-to-write tasks. In other words, textual borrowing determines lexical diversity of L2 writers. Shi and Dong (2018) note that paraphrasing and textual borrowing are affected by cultural and linguistic differences of L2 learners. Some paraphrasing and textual borrowing practices might be acceptable, but some might not since some L2 writers copy texts from source texts and change only some words or grammar. As a result, this affects the overall quality of summary writing of these novice L2 writers.

Assessing Summary Writing
In academic writing contexts, to assess the quality of summary writing, scoring rubrics are usually used since they can be used as assessment guidelines and can reflect the writing quality and problems L2 writers have. These can be used as information for writing teachers for giving feedback and designing writing lessons for further development. According to Knoch (2009), scoring rubrics play an important role in assessing integrated writing tasks that require students to write a summary. Additionally, the assessment of L2 summary writing is taken into account among research studies in the fields of L2 writing and language testing (Plakans, 2015;Weigle, 2004).
However, scoring rubrics may contain scoring guidelines for different criteria, so there are a number of rubrics created to use for assessing L2 summary writing, both holistically (e.g., Educational Testing Service (ETS), 2002; Yamanishi et al., 2019;Yu, 2007) and analytically (e.g., Li, 2014;Sawaki, 2019;Yamanishi et al., 2019). Holistic assessments are often used for reflecting an overall writing performance of L2 writers (Hyland, 2002;Weigle, 2002). They are usually used when a summative score of the writer's performance is needed. They can be used as placement tests or high-stakes tests since they are cost-effective, and the raters take less time to complete the assessments when they are compared with analytic assessments (Bacha, 2001;. However, holistic assessments do not provide strengths and weaknesses of each writer. That is, constructive feedback and informative feedback cannot be provided (Yamanishi et al., 2019). On the other hand, analytic assessments are used when the writing performance can be shown in several dimensions. They provide descriptors for the raters to assess each dimension. Moreover, the strengths and weaknesses of the writer can be identified. As a result, writing teachers can give diagnostic feedback, comprehensive feedback, and informative feedback on each writer's performance (Mertler, 2001;Stevens & Levi, 2013).

Previous Studies
There are various perspectives of summary writing that have been studied. For example, a lot of research studies aimed to investigate the effects of source texts on summary writing, the role of paraphrasing in summary writing, text borrowing, and assessing summary writing. Yu (2009) reported the effects of the properties of source texts on summarization. The summaries of 157 Chinese undergraduate students were analyzed in terms of students' summarization task performance and their perception. The findings revealed that the familiarity with the topics did not significantly affect the summarization performance. However, regarding the perception of L2 writers, macro-organization, frequency of unfamiliar words, topic familiarity, and length of source texts were important factors for their summary writing. McDonough et al. (2014) investigated summary writing in a Thai EFL university context with 46 undergraduate students focusing on how their summary paragraph writing changed in terms of the rhetorical organization and the incorporation of information from the source texts. They were taught with an explicit instruction about paragraph writing and paraphrasing strategies. The results showed that there were more students who referenced the source texts. Moreover, although there was the use of copied words from source texts, the length of copied strings decreased (i.e., 2-4 words) in their summaries. Keck (2006Keck ( , 2014 compared the use of paraphrasing in summary writing between L1 and novice L2 writers and found that L2 writers tended to have a difficulty in paraphrasing and employ insufficient paraphrasing which is called "near copy" when compared with L1 writers. Similarly, Shi and Dong (2018) investigated paraphrasing of 17 Chinese graduate students when they read research papers, and they had to paraphrase them. It was found that their paraphrases contained textual borrowing. There were some paraphrases that were unacceptable since novice writers copied information from source texts, and only words and grammar were changed.
Also, there was a study done by Weigle and Parker (2012) regarding source text borrowing in an integrated reading/writing assessment. The study aimed at determining the extent to which students borrow source text language. Written essays were transcribed verbatim and coded for source text borrowing. It was revealed that the majority of students did not borrow the language from source texts extensively. That is, it was still acceptable. In addition, there were few differences in the borrowing across topics, groups of students, and their levels of proficiency.
In view of summary writing assessment, a rubric for L2 summary writing was developed by Educational Testing Service (ETS) (2002). It is a holistic rubric that reflects an overall score of test takers, ranging from 1-5, and it emphasizes organization, sentence structure, language use, and language from the source texts. Baba (2009), for example, employed this rubric to examine summary writing performance of Japanese university students, focusing on the impact of their lexical proficiency on English (L2) summary writing. Also, Becker (2016) investigated the effect of developing and/or applying a rubric on summary writing performance of ESL writers using an overall score, ranging from 1-5. The findings of the study can raise awareness of involving students in the process of writing assessment. This can lead to an improvement of their writing performance. Knoch (2009) compared two rating scales for writing in an English for Academic Purposes (EAP). The first one was a developed scale with less specific descriptors that were commonly used in proficiency tests, and the other one was a developed scale with detailed level descriptors. The results indicated that more detailed descriptors allowed raters to better distinguish among different writing's aspects, and the raters' perceptions towards the more detailed scale were positive.
Recently, a study conducted by Yamanishi et al. (2019) aimed at developing a scoring rubric for L2 summary writing. The developed rubric was both holistic and analytical assessment. It contained five dimensions, namely content, paraphrase (quantity), paraphrase (quality), language use, and overall quality. The results of using this rubric in terms of quantitative examination and qualitative examination of the raters' perceptions were positive. The details of each descriptor were clear and helped in teaching and assessing L2 summary writing. Ono (2021) employed this scoring rubric to investigate paraphrasing the source text in integrated tasks and summary writing of 70 Japanese EFL students who were novice L2 writers. This scoring rubric can reflect paraphrasing and summarizing performance of these students and also difficulties they have since it contains descriptors with specific details for assessing summary writing.
Concerning lecturers' perception of using a scoring rubric for assessing summary writing, Putri (2020) conducted a study aiming to elicit perceptions of writing teachers using analytical rubric in scoring summary writing. The rubric contains five dimensions of assessing summary writing, namely length, accuracy, paraphrasing, focus, and conventions. It was revealed that the perception of the lecturers towards using scoring rubric was positive.
Additionally, using an analytical rubric was helpful and could enhance the summary writing of L2 writers since they knew what criteria they have to achieve.
Apart from assessing the quality of summary writing, eliciting the opinions regarding problems encountered by L2 writers in summary writing is also crucial. By doing so, the problems of summary writing can be found, and these problems will be beneficial to writing teachers for their lesson and activity design and improvement. Therefore, this present study also investigated this perspective of L2 summary writing.

Participants
The participants of this study were 67 fourth year English major students who enrolled in a research report writing course (205426) in the first semester of academic year 2021 at Naresuan University. All participants were Thai EFL writing learners, and they were chosen by purposive sampling. Their English proficiency was between lower intermediate and intermediate.

Data Collection
Regarding ethical issues, this research was reviewed and approved by the Institutional Review Board (IRB), (IRB No. P2-0322/2564). At every stage, the participants' names would remain confidential, and the results of this study were used for academic purposes only.
In this study, all participants were asked to do a summary writing test. They were instructed to write a summary of the given passage, in a paragraph of 120-150 words, describing second language acquisition (SLA), within one hour at one sitting and under test conditions (see Appendix A for more details). This source text was used because it was the topic in their field of study. All participants were English major students, so they might be familiar with it, and they were able to connect their ideas with this topic. This allowed them to perform their summary writing ability. The participants were allowed to consult dictionaries. Then, the written scripts were collected for further analysis.
For data collection of the opinions regarding problems encountered by Thai EFL English major students in summary writing, 12 students were asked to have a face-to-face interview with the researcher. They were chosen by systematic random sampling. The interview was conducted after the summary writing test. It was a semi-structured interview, and it took approximately 30 minutes for each interviewee. The interview questions were divided into three stages of writing, namely prewriting, while writing, and revising. The emphasis was put on the problems these participants encountered when they wrote summaries from beginning to end. The details of the interview questions are shown in Appendix B.

Data Analysis
After the data were collected, each summary was graded using the scoring rubric for L2 summary writing which was developed by Yamanishi et al. (2019). This scoring rubric was chosen since it was both holistic and analytical assessment. There were five main dimensions that were analyzed (i.e., content, paraphrase (quantity), paraphrase (quality), language use, and overall quality). Content is judged by whether the writer can grasp and develop the main ideas using secondary information. Paraphrase (quantity) refers to the extent to which the expressions in the summary are paraphrased in the writer's own words. Paraphrase (quality) identifies whether the writer can demonstrate effective paraphrases by using different sentence structure and vocabulary from the original text. Language use is judged by whether the writer can demonstrate an effective and appropriate use of vocabulary and sentence structure. Overall quality identifies the overall quality of the summary holistically as a response to the task requirements. Each dimension was assessed according to writing performance levels, ranging from 1-4. The full mark was a score of 20. The details of this scoring rubric are shown in Appendix C. After that, the data were converted into mean (x̄) and standard deviation (S.D.).
For interview data, content analysis was employed. After the interview data was collected, the data were transcribed and analyzed. Then, the opinions regarding the summary writing problems were then summarized and categorized according to three main stages of summary writing, namely prewriting, while writing, and revising.
In the following section, the quantitative results of the summary writing test and qualitative results of the interview are reported.

Results and Discussion
The results and discussion of this research are presented according to the three research questions (RQs) of this study. To answer RQ 1, Table 1 reveals the findings concerning rating of the summaries and the quality of summary writing for each dimension. Regarding the quality of summary writing produced by these EFL writers, overall, the total score is 9.40 out of 20. This result indicates that the summary writing produced by these EFL students is regarded as "fair". When considering each dimension, content is the dimension that they performed the best (x̄=2.03, S.D.=0.58) when compared with other dimensions. The mean scores for other dimensions were between 1.75-1.94 which are still similar and seen as "fair" as well. According to the scoring rubric developed by Yamanishi et al. (2019, p. 19), a score of 2 indicates "fair". This means that, in the dimension of content, these EFL writers "can grasp only limited main ideas, cannot demonstrate an adequate development of the main point, and noticeably include incorrect information or information beyond the original text". Concerning paraphrase (quantity), they "can paraphrase only from 25% to less than 50% of the expressions included in the summary in one's own words". Regarding paraphrase (quality), these writers "include few expressions consisting of more than 4 words in a row copied from the original text, can only demonstrate paraphrases using vocabulary from the original text, and delete expressions partially or change the word order". In the dimension of language use, they "can demonstrate only a limited range of vocabulary, word/idiom choice and usage and can demonstrate simple sentence construction. The meaning is obscure due to frequent major errors". For overall quality, "as a response to this task, the overall quality of this summary is fair." The reason for this might be because these EFL writers lack reading skills and strategies, so they cannot find the main ideas of the original text when they read. Additionally, they probably have problems with grammar and vocabulary. As a result, they cannot write their summaries in their own words appropriately, especially within limited time and a limited number of words (i.e., 120-150 words). To avoid changing the meaning of the source text, they copy or borrow texts from the source text or only change the sentence structure or order of words in the source text. These reasons can lead to incorrect information and inappropriate textual borrowing in their summaries.
RQ 2: What are the most repeated types of problems made by Thai EFL English major students in summary writing?
To answer RQ 2, according to Table 1, the results reveal that there are no significant differences among all dimensions, ranging from 1.75-2.03, which are in a fair level. This means there is no the most repeated type of mistakes made by these EFL writers in summary writing. They all are regarded as "fair" equally. However, based on written scripts, when considering each dimension in greater detail, there are main problems made by these EFL writers.
In the dimension of content, the main problem is that incorrect information or irrelevant information is included. An example of this problem is shown below.
Original source: Second Language Acquisition (SLA) refers both to the study of individuals and groups who are learning a language subsequent to learning their first one as young children, and to the process of learning that language. The additional language is called a second language (L2), even though it may actually be the third, fourth, or tenth to be acquired. It is also commonly called a target language (TL), which refers to any language that is the aim or goal of learning. The scope of SLA includes informal L2 learning that takes place in naturalistic contexts, formal L2 learning that takes place in classrooms, and L2 learning that involves a mixture of these settings and circumstances.

Summary:
Second language acquisition (SLA) cites to the study of the learning and procedures the first language of both personal and group when as a child. The additional language is called the second language or target language. In addition, the formal language occurs in naturalistic, and the informal lingual occurs in a classroom. (S42) In this example, incorrect information is included, and the main ideas are not fully included. This implies that the writer failed to comprehend the information in the source text. As a result, incorrect ideas are presented. This finding supports Plakans (2009) who explored the role of reading strategies in integrated L2 writing tasks in that reading plays an important role in the process and performance of integrated writing tasks. In other words, the writers' difficulties in comprehending the task can limit their ability to use the source text, and this directly affects their writing.
For paraphrase (quantity), the main problem is that these students can paraphrase only from 25% to less than 50% of the expressions included in the summary in their own words. Below is an example of this problem.
Original source: The scope of SLA includes informal L2 learning that takes place in naturalistic contexts, formal L2 learning that takes place in classrooms, and L2 learning that involves a mixture of these settings and circumstances. For example, "informal learning" happens when a child from Japan is brought to the USA and "picks up" English in the course of playing and attending school with native English-speaking children without any specialized language instruction, or when an adult Guatemalan immigrant in Canada learns English as a result of interacting with native English speakers or with co-workers who speak English as a second language. "Formal learning" occurs when a high school student in England takes a class in French, when an undergraduate student in Russia takes a course in Arabic, or when an attorney in Colombia takes a night class in English. A combination of formal and informal learning takes place when a student from the USA takes Chinese language classes in Taipei or Beijing while also using Chinese outside of class for social interaction and daily living experiences, or when an adult immigrant from Ethiopia in Israel learns Hebrew both from attending special classes and from interacting with co-workers and other residents in Hebrew.

Summary:
This SLA will have three types of classes. 1.) Studying in an informal way, such as studying directly with native speakers 2.) Formal learning, such as studying with people with special skills or studying in the classroom and finally, 3.) Mixed learning, such as studying in a room with a special skilled teacher and outside the room to chat directly with native speakers. (S8) Regarding paraphrase (quality), the main problem is that these students mostly copy words/information from the original text, including using vocabulary or changing word order from the original text. An example of this is as follows.
Original source: In trying to understand the process of second language acquisition, we are seeking to answer three basic questions: a) What exactly does the L2 learner come to know? b) How does the learner acquire this knowledge? c) Why are some learners more successful than others?
There are no simple answers to these questions -in fact, there are probably no answers that all second language researchers would agree on completely. In part this is because SLA is highly complex in nature, and in part because scholars studying SLA come from academic disciplines which differ greatly in theory and research methods.

Summary:
In studying the process of second language acquisition, we must answer three questions: What do the learners know, how do they know, and why do they gain more knowledge than others? There are no definitive answers to these questions, as SLAs are very complex, and the scholars studying them come from different disciplines.
As mentioned in the dimension of content, since the writer failed to comprehend the information in the source text, this eventually results in inappropriate paraphrasing for both paraphrase (quantity) and paraphrase (quality). Additionally, these novice EFL writers have some difficulties in using grammar correctly. Therefore, these significantly affect the quantity and quality of paraphrasing. As noted in Keck (2006Keck ( , 2014, novice L2 writers tend to employ insufficient paraphrasing or mostly copy and employ textual borrowing. This is in line with Shi and elt.ccsenet.org English Language Teaching Vol. 15, No. 6;2022 22 Dong (2018) who point out that novice writers' paraphrases contain textual borrowing. They tend to copy excerpts from source texts, and only words and grammar are changed.
For language use, the main problem is simple sentence structures are used. Also, there is incorrect use of grammar, so this leads to obscure meaning. Below is an example of this.
Original source: Applied linguists who specialize in SLA may take any one or more of these perspectives, but they are also often concerned with the implications of theory and research for teaching second languages. Each discipline and subdiscipline uses different methods for gathering and analyzing data in research on SLA, employs different theoretical frameworks, and reaches its interpretation of research findings and conclusions in different ways.

Summary:
Each discipline subdisciplines uses different methods for gathering analyzing data in research on SLA, employs theoretical frameworks, reaches its interpretation conclusions ways. Applied linguists specialize take any more these perspectives, they also concerned with implications theory teaching languages. (S52) As can be seen, language use in this example is not perfect, so the meaning of the ideas in the summary is unclear. This might be because these EFL writers lack knowledge of English grammatical rules. The problems of incorrect use of grammar are also found in a number of research studies conducted in academic writing contexts (e.g., Boonyarattanasoontorn, 2017;Hamed, 2018;Hussain (2019); Murad & Khalil, 2015;Nuruzzaman, Islam & Shuchi (2018); Phuket & Othman, 2015;Sermsook, Liamnimitr, & Pochakorn, 2017;Sychandone, 2016). The findings of these studies indicate that grammar is one of the main error types made by novice L2 writers. Olsen (1999) and Weigle (2002) note that limited knowledge of English grammar leads to making errors in L2 writing.
In the dimension of overall quality, as all the main problems mentioned above, this leads to Score 2 (Fair) for this dimension.
RQ 3: What are the opinions regarding problems encountered by Thai EFL English major students in summary writing?
To answer RQ 3, the findings regarding the opinions regarding problems encountered by these EFL writers in summary writing are divided into three stages as shown in Table 2. Table 2. A summary of opinions regarding summary writing problems

Stages of writing Summary writing problems
Prewriting • Overall, the main problem in this stage of writing results from reading and comprehending the source text unclearly.

•
The majority of the interviewees mentioned that when they had to read a longer text, they could not find the main ideas of the text and finally lost focus, especially when they had to write a summary on an unfamiliar topic and had limited time.

•
Someone spent too much time reading the source text.
• In addition, difficult vocabulary played an important role in comprehending the text since they directly affected their understanding.

•
Many of them could not connect ideas within the text, so they focused on only the main ideas they were clear about. Some interviewees stated that they needed to read the source text many times in order to link all information in the text.

While writing
• While these EFL writers were writing a summary, they found it difficult to choose appropriate words to replace the words they wanted to use from the original source when they paraphrase the text. They were afraid that the meanings of words they changed and used in their summary were different from those in the original text.

•
The next problem is that when they were writing a summary in their own words, they were not sure that their summaries conveyed the same meaning as the original text.
• Also, they felt that they had a problem with grammar, so this would affect their summaries. As a result, they wrote them by changing the order of words or information in the source text. For example, they changed the sentence structure from passive voice to passive voice and vice versa, or they changed from one type of sentence to another (e.g., compound sentence or complex sentence). It can be said that writing a summary in their

Stages of writing
Summary writing problems own words and sentence structures make them unsure whether or not the meanings they try to convey remain unchanged.
• Lastly, since the integrated task required these EFL writers to write within a limited number of words (i.e., 120-150 words), they thought that they could not organize their ideas or select appropriate main ideas well to put in their summaries.

Revising
• For revising their summaries, these EFL writers stated that they spent most of their time on pre-writing and while writing stages, so they did not have much time for revising. Therefore, when they revised their summaries, the main revising was on checking the correctness of punctuation marks (e.g., comma and full stop), spelling, subject-verb agreement, and word choices.

•
Additionally, with limited time, some of them did not reread and revise at all.
• Another main problem they mentioned was that they were not sure if their grammar used in their summaries was correct. This caused some of them to be unsatisfied with their summaries.
In the stage of pre-writing, the findings shown in Table 2 correspond to the findings of Plakans (2009) who explored the role of reading strategies in integrated L2 writing tasks and pointed out that the writers' problems in comprehending the source text negatively affect their writing performance. According to Devine (1988) and Jovari (2020), when the reader reads a text, he/she tries to create meaning from the text based on the interaction between previous knowledge (schema) of the reader and the information available in the text itself. Jovari (2020) mentioned "knowledge of text content can facilitate learners' comprehension during the encoding/decoding process by providing a knowledge structure to which readers can compare and fit pieces of incoming information" (p.29). The reason for this might be due to a lack of reading exposure. Especially, these EFL learners might not read a variety of academic texts (e.g., academic journals, research paper, and conference paper) outside the classroom. They do not probably recognize the structure and language use in the academic texts. In addition, they might not be familiar with the text content or have background knowledge about the text. Thus, if the EFL writers in this present study were not able to comprehend the source text clearly, this significantly affects the quality of their summary writing.
For while writing, the above findings are in line with a number of research studies (e.g., Keck, 2014;McDonough et al., 2014;Shi and Dong, 2018;Weigle & Parker, 2012) which claim that, for novice L2 writers, when they write summaries and paraphrases, their paraphrases tend to contain textual borrowing. There are some paraphrases that are unacceptable since they copy information from source texts and only words and grammar are changed. Moreover, Weigle and Parker (2012) indicate that when L2 writers write summaries from source texts, inappropriate textual borrowing can occur if they are not familiar with the source texts or cannot paraphrase the information they need appropriately, especially in a timed writing assessment. The findings of this present study also lend support to Yu (2009) who found that, regarding the perception of L2 writers, macro-organization, frequency of unfamiliar words, topic familiarity, and length of source texts were important factors for their summary writing. One new finding emerging in this present study is that the length of the number of words required in the integrated task is also another important factor that affects the summarization performance. As shown in Table 2, these EFL writers thought that, with a small number of words required to write in their summaries, they could not organize their ideas or select appropriate main ideas well to put in their summaries.
In revising stage, the findings of this present study support Silva's (1993) findings which indicate that writing in L2 is more difficult than writing in L1. They also find it difficult to organize ideas, write, and revise effectively. Their texts show more errors, and their responses to the task requirements are less effective. Additionally, their sentences tend to contain grammatical errors and discrete patterns in the use of cohesive devices. As a result, these errors certainly affect the quality of the whole text. The cause of this might result from their limited knowledge of English grammar. When they reread and check what they have written, they might not know what grammatical errors they have to correct. They might be able to notice and correct only some mistakes, such as punctuation marks or spelling, but grammatical errors in sentence structure (e.g., fragments, run-on sentence, and comma splice) and the use of conjunction (e.g., coordinating conjunction, subordinating conjunction, and correlative conjunction) might not be checked and corrected appropriately. This shows the importance of revision in L2 summary writing since it plays a crucial role in summary writing and quality of writing. According to Khaled (2007), revising is an important skill, but many novice L2 writers do not take it into account. They may find generating ideas easier than revising. As a result, they fail to revise effectively. The findings of Choi (2010) also show the importance of revision in L2 writing and positive effects of revision training on L2 writing in terms of the quality and product of writing, especially in less proficient writers.

Pedagogical Implications for L2 Learning and Teaching
In terms of writing instruction, there are six important implications for pedagogy that have emerged from the findings of this study. The first implication of the findings is that, in integrated reading-to-write tasks, reading plays a crucial role in summary writing since EFL writers need to use the information from the source text in their summary writing. The findings of this presents study showed that these writing learners had a problem in comprehending a source text. As a result, this negatively affects the quality of their summaries. Writing teachers, thus, need to take improving reading skills and strategies (e.g., guessing meaning from contexts, improving vocabulary skills by learning prefixes and suffixes, and identifying main ideas) into account so that their students are able to understand the source text clearly and can produce effective summaries. Secondly, an emphasis on recognizing and avoiding plagiarism or copying needs to be put in writing classrooms since many novice EFL writers do not take this issue seriously as can be seen from the findings of this present study. Therefore, the lessons on how to avoid plagiarism and copying the source text need to be taught explicitly. Thirdly, L2 writing students should be encouraged to do paraphrasing activities and exercises since they can help to improve paraphrasing skill which is necessary when these EFL writing learners write their summaries. Fourth, the knowledge of some particular mechanics and grammar (e.g., spelling, prepositions, articles, capitalization, singular/plural, punctuation, fragments, and subject-verb agreement) need to be provided. If these novice EFL writers can use mechanics and grammar correctly, they can significantly develop their summary writing. Fifth, revision in L2 writing is critical. Novice EFL writers need to manage their time to have enough time for revising, for example, checking for idea organization, word choice, fragments, singular/plural, and subject-verb agreement. Therefore, lessons and activities on time management and revision are needed. This will definitely help to improve the quality of their summary writing. Lastly, the scoring rubric developed by Yamanishi et al. (2019) which was employed in this present study was found to be effective and easy to use. Besides, it includes all salient dimensions for assessing summary writing in terms of both holistic and analytical assessment. Therefore, it is recommended for use in L2 summary writing assessment.

Conclusions and Recommendations for Further Research
This study examined the quality of summary writing and type of problems faced by Thai EFL English major students and also elicited opinions regarding problems encountered by these EFL learners. The findings showed that the quality of summary writing produced by these EFL writers was viewed as "fair", and there were no significant differences among all dimensions. They all were regarded as "fair" equally. Each dimension revealed the main problem mostly found. In the dimension of content, the main problem was that incorrect information or irrelevant information was included. For paraphrase (quantity), the main problem was that these students could paraphrase only from 25% to less than 50% of the expressions included in the summary in their own words.
Regarding paraphrase (quality), the main problem was that these students mostly copied words/information from the original text, including using vocabulary or changing word order from the original text. For language use, the main problem was that simple sentence structures were used. Also, there was incorrect use of grammar, so this led to obscure meaning. In the dimension of overall quality, as all the main problems mentioned above, it was considered as fair for this dimension. Regarding their opinions regarding summary writing problems encountered by these EFL writers, the main problem in the prewriting stage was that they could not find the main ideas of the text and finally lost focus. The problems in comprehending the source text negatively affected their writing performance. While writing, the main problems were that they found it difficult to choose vocabulary words to replace the words they wanted to use from the original source when they paraphrased the text. Also, when they wrote summaries in their own words, they were not sure that their summaries conveyed the same meaning as the original text. Moreover, they thought they had a problem with grammar. As a result, their summaries and paraphrases tended to contain textual borrowing. Some paraphrases were also unacceptable since information from source texts were copied, and only words and grammar were changed. For revising their summaries, these EFL writers stated that they did not manage time well, so they did not have much time for revising. Therefore, this study sheds light on the problems of summary writing encountered by Thai EFL English major students. These can help writing teachers to use the findings of this study and scoring rubric used in this study to improve summary writing of their students. Additionally, this study reflects the opinions these EFL writers had when they wrote summaries. These opinions need to be taken into account when writing teachers design their writing lessons and activities.
In this study, there are two limitations. First, the participants in this study were 67 English major students who enrolled in a research report writing course. Since it was not possible to manage them in one study group, they were divided into 4 groups and taught by 2 different teachers. Such a circumstance may have some differences in terms of teaching style and extra assignments. However, the provided instruction, including teaching materials and the main textbook were the same. Therefore, this may not significantly affect the overall result of the study. The other potential limitation is that the number of words required the participants of this study to write a summary is 120-150 words. This length of summaries may affect the quality of summary writing produced by these EFL writing learners.
For further research, the researcher would like to recommend conducting research by examining learners' summaries of different levels of proficiencies. Furthermore, since the length of a summary of this present study ranges from 120-150 words, it would be interesting to conduct another study using different lengths of summaries and compare the differences in terms of the quality and problems encountered by EFL writers among those different lengths. These would provide worthy avenues for future research and fill the possible gaps existing in the literature of summary writing.

Appendix A The Summary Writing Test
Directions: Make a summary of the passage below, in a paragraph of 120-150 words, describing second language acquisition.
What is SLA?
Second Language Acquisition (SLA) refers both to the study of individuals and groups who are learning a language subsequent to learning their first one as young children, and to the process of learning that language. The additional language is called a second language (L2), even though it may actually be the third, fourth, or tenth to be acquired. It is also commonly called a target language (TL), which refers to any language that is the aim or goal of learning. The scope of SLA includes informal L2 learning that takes place in naturalistic contexts, formal L2 learning that takes place in classrooms, and L2 learning that involves a mixture of these settings and circumstances. For example, "informal learning" happens when a child from Japan is brought to the USA and "picks up" English in the course of playing and attending school with native English-speaking children without any specialized language instruction, or when an adult Guatemalan immigrant in Canada learns English as a result of interacting with native English speakers or with co-workers who speak English as a second language. "Formal learning" occurs when a high school student in England takes a class in French, when an undergraduate student in Russia takes a course in Arabic, or when an attorney in Colombia takes a night class in English. A combination of formal and informal learning takes place when a student from the USA takes Chinese language classes in Taipei or Beijing while also using Chinese outside of class for social interaction and daily living experiences, or when an adult immigrant from Ethiopia in Israel learns Hebrew both from attending special classes and from interacting with co-workers and other residents in Hebrew.
In trying to understand the process of second language acquisition, we are seeking to answer three basic questions: a) What exactly does the L2 learner come to know? b) How does the learner acquire this knowledge? c) Why are some learners more successful than others?
There are no simple answers to these questions -in fact, there are probably no answers that all second language researchers would agree on completely. In part this is because SLA is highly complex in nature, and in part because scholars studying SLA come from academic disciplines which differ greatly in theory and research methods. The multidisciplinary approach to studying SLA phenomena which has developed within the last half-century has yielded important insights, but many tantalizing mysteries remain. New findings are appearing every day, making this an exciting period to be studying the subject. The continuing search for answers is not only shedding light on SLA in its own right, but is illuminating related fields. Furthermore, exploring answers to these questions is of potentially great practical value to anyone who learns or teaches additional languages.
SLA has emerged as a field of study primarily from within linguistics and psychology (and their subfields of applied linguistics, psycholinguistics, sociolinguistics, and social psychology), as a result of efforts to answer the what, how, and why questions posed above. There are corresponding differences in what is emphasized by researchers who come from each of these fields:

•
Linguists emphasize the characteristics of the differences and similarities in the languages that are being learned, and the linguistic competence (underlying knowledge) and linguistic performance (actual production) of learners at various stages of acquisition.
• Psychologists and psycholinguists emphasize the mental or cognitive processes involved in acquisition, and the representation of language(s) in the brain.
• Sociolinguists emphasize variability in learner linguistic performance, and extend the scope of study to communicative competence (underlying knowledge that additionally accounts for language use, or pragmatic competence).
• Social psychologists emphasize group-related phenomena, such as identity and social motivation, and the interactional and larger social contexts of learning.
Applied linguists who specialize in SLA may take any one or more of these perspectives, but they are also often concerned with the implications of theory and research for teaching second languages. Each discipline and subdiscipline uses different methods for gathering and analyzing data in research on SLA, employs different theoretical frameworks, and reaches its interpretation of research findings and conclusions in different ways.