A Comparative Study on Sentence Characters of Research Introductions by Indonesian Authors in Applied Linguistics

This article presents the findings of sentence characters in article introductions written by Indonesian authors in Applied Linguistics published in national accredited and international reputable journals. This study aimed to identify sentence characters in article introductions used by Indonesian authors in Applied Linguistics published in national accredited journals and reputable international journals. The method employed in this study was a mixed-method to describe and compare the data. This study’s total number of articles was forty from eight different journals. The results indicated that the average sentence lengths in terms of words number per sentence in the international reputable journals articles were 22.38 words, and those in national accredited journals articles were 21.84 words. Also, the results revealed that the most commonly used sentence pattern in both groups of articles was complex sentences, with 414 sentences in national journals and 378 in international journals. It is followed by simple, compound-complex, and compound sentences. Therefore, it can be concluded that the Indonesian authors from both groups prefer to use long words and use complex sentences more often than the other sentence patterns. Hence, there is no important difference in sentence characters between national accredited and international reputable journals.


Introduction
A research article published in journal is one form of the outcome of research academics in the universities and other researchers from various institutions. General Director of DIKTI in Circular Letter No. 152/E/T/2012 specifies that the lecturers, academics, researchers who conduct research activities at the college must publish journal articles. Besides, the General Director of DIKTI in Circular Letter No. 20150/E/T/2011 on the Publication Policy requires that the researcher publish scientific papers or online journal articles through the e-journal.

Ministry of Research Technology and
Higher Education (2016) explains that Indonesian researchers in the field of social sciences and humanities, including the field of Applied Linguistics or English language education, are underneath other Indonesian researchers in the fields of science, agriculture, medicine, computers, and engineering on publishing in reputable international journals. The condition is certainly unexpected because the general assumptions said that professors or researchers in English majors would not experience difficulty writing articles in English for their English language proficiency which is relatively better than researchers or lecturers in other fields of science.
According to the data of DIKTI (2016) most Indonesian academic publications are the fields of science, technology, health, and medicine such as Engineering (15.14%), Medicine (10.64%), Computer Science (10.2%), Agriculture, and Biological Sciences (9.57%), Physics and Astronomy (5.39%), while publications in the field of Social Sciences were only (4.74%) and in Arts and Humanities (0.91%). The data indicates that the least publication in a reputable international journal in social sciences and humanities includes English language education (Applied Linguistics). This situation is contrary to people's expectations, where English language lecturers and researchers experience less difficulty because they are mastery the language and academic writing skills more than lecturers and researchers in other disciplines.
The first possible cause of low publication of lecturers in Social Sciences and Humanities in international journals is some Indonesian academic authors unaware of the rhetorical pattern and the standard of linguistic features in every section and subsection of research journal articles in English. Swales (1990Swales ( & 2004 describes differences in rhetorical styles of scientific articles in Indonesian and English as found by Safnil (2001), Mirahayuni (2002), Adnan (2009), Arsyad & Wardhana (2014), and Arsyad & Arono (2016). Blagojevic (2015) says that in practice, especially humanities field, the writers regularly use the first language (L1) style when writing in English (L2) for international journals.
Another possible cause is Indonesian lecturers and researchers in the field of social sciences, especially in the field of English language education, unaware of linguistic features in their writing such as; sentence patterns, phrase patterns, specific word choices (specific lexicons), cohesion, and coherence patterns, selection of citation verbs (reporting verbs) and the use of metadiscourse markers which are often used by international journal writers as found in articles that have been published in reputable international journals.
The research articles' introduction is in great demand by Indonesian researchers to analyze because the introduction section is a form of a case being investigated that provides information to make the readers understand the purpose and specification in a larger theoretical framework. This section includes background information on the problem, such as summarizing the current study and previous studies. The introduction section is the most important part of the scientific text section, which is the first part that the readers should read after the abstract.
Related to the explanation above, the important problem experienced by the author of research articles is to write a research article. The linguistic feature becomes the problem in writing a research article which is the sentence character such as sentence length (in terms of the number of words and term of the number of sentences) and sentence patterns such as simple sentence, compound sentence, complex sentence, and complex compound sentence in research articles. However, the study is still rarely done, especially in research articles' introduction sections about sentence characters in Applied Linguistics journals. As the introduction is an important part of research that will be read to know the gap of study in the publication journal, so the logic of this study is to investigate the sentence characters or the stylistic trends used by Indonesian authors in the research articles Introduction sections of national accredited journals and reputable international journals.
Many studies of scientific activities done by researchers worldwide, much of the literature have been published in English by Non-Native Speaker (NNS). Publications by these scholars are often the equivalent of publications by Native English speakers (NSE) (Buckingham, 2008). However, there is evidence from research that is the focus to the difficulties faced by NNS when compiling scientific texts.
Englander (2014) found that the NNS English writing was on average 21% more difficult than writing in their native language. This is caused by vocabulary, syntax, organization, and methodological problems. Another study by Weijen (2014) found that writing in English can be more timeconsuming for NNS than for NSE because it requires more cognitive effort in processing language. They mostly lack clarity in explaining, grammatical errors, organizational inconsistencies, and problems with the format of the thesis. Dean (2000) notes that sentence length choice can help writers establish their position. Mamishev & Williams (2010) recommend that writers use different types of sentences to make their writing more sophisticated.
Moreover, the study by Deveci (2019) found that the average sentence length of education research articles written by Anglophone and Turkish authors was 24.7. It is the average sentence length belonging to the average length by readability experts which suggests an average of 20-25 words (Garner, 2000). Meanwhile, journal publication guidelines advise that scientific writing should be around 12-17 words per sentence (Griffies, Perrie, & Hull, 2013).
Research on rhetoric style and linguistic features of journal articles (AJ) written in Indonesian and English by Indonesian researchers and published in Indonesian journals and international journals is frequently conducted. These studies were conducted by Safnil (2001), Adnan (2009), Arsyad (2013. a), Arsyad (2013.b), Arsyad and Wardhana (2014), and Arsyad and Arono (2016). The results of this study are very important, especially in how the authors of Indonesian journal research articles should modify their research articles when writing English research articles to be published in a reputable national and international journal.
The research on journal articles in Indonesian and English was mostly conducted to analyze the rhetorical patterns of article sections such as abstracts, introduction, results, and discussion. Still, only a few comparative research studies were carried out on the linguistic features, especially the sentence character of English education research articles published in national accredited journals and reputable international journals. This is the main reason for conducting this research, namely to see the sentence characters in the introduction section of English research articles in applied linguistics published in an accredited national accredited journal and internationally reputable journal. Three research problems guided this study: 1) What are the average sentence lengths in research articles introduction written by Indonesian authors published in national accredited journals and reputable international journals? 2) What are the common sentence patterns found in research articles introduction in Applied Linguistics written by Indonesian authors published in national accredited journals and reputable international journals?
Journal of Language and Literature ISSN: 1410-5691 (print); 2580-5878 (online) Ameitha Pratiwi, Safnil Arsyad, & Syafryadin 3) What are the differences between the nationally accredited and reputable international journals regarding the number of sentence patterns?

Methodology
The method employed in this study is a descriptive quantitative study to describe and compare the numerical data. This study used a quantitative method. Sugiyono (2015) interprets the quantitative method as a research method to examine a population or a particular sample; data collection using research instruments, quantitative or statistical data analysis to trial the hypothesis. Furthermore, the study is following Deveci (2019) by calculating the average sentence length in the introduction section of research articles on the research corpus and identifying sentence patterns (simple sentence, compound sentences, complex sentence, and complex-compound sentence) in the introduction section of the research articles on the research corpus.
The corpus of this research involved forty articles from reputable international journals and national accredited journals. The articles were taken from eight different journals in Applied Linguistics published by Indonesian authors in national accredited journals and reputable international journals. The research articles were taken from eight journals, namely IJAL, TEFLIN, SiELE, IJoLE, JEELS, IJELTAL, JOALL, & Linguists. The journals were categorized in Table 1. This research used a checklist as the instrument containing sentence patterns.
This study uses Weber and Brizee (in Deveci, 2019) to determine the characters, such as average sentence length and sentence pattern. First, the researcher downloads the data which consists of forty research articles from eight journal sites. Second, the data that has been downloaded is converted from .pdf files into .doc format files using online software available at http://pdf2doc.com. The analysis in this study was completed by identifying the linguistic features, namely, sentence characters such as sentence length and sentence patterns in the introduction section. There are several stages used in this study to identify sentence characters.
First, the data is categorized into five sections (abstract, introduction, method, result and discussion, and conclusion) because this research only investigates the introduction section. Second, the researcher read the entire introduction carefully. Third, the researcher manually corrects the format defects such as spelling and punctuation in articles. Tables,  table  captions, acknowledgments, and reference lists are deleted. Direct quotes from participants or texts analyzed in the quantitative studies are also removed because the numerical data and the sentences in these sections are outside the study's scope.
The next stage was identifying the average sentence length in the introduction section. The researcher uses online software available at http://countwordsworth.com/wordspersente nce to compare the sentence length between nationally accredited journals and reputable international journals. The trial of this software showed that it recognized all of the periods, like the end of a sentence (Deveci, 2019). After calculating using online software, the data is entered into an excel sheet. To get the results, it has been inputted into excel, the data is processed using a formula available in the excel program. In the last stage, the researcher identified the sentence pattern by Waber and Brizee (in Deveci, 2019) as simple, compound, complex, and compound sentences. The sentence pattern was identified through the result found in the research article. After that, the researcher obtained and calculated the data and input it into the table. The last, the researcher analyzed the result.
This study used a co-rater to validate the result of the data analysis. In this study, Cohen Kappa's coefficient analysis is used to evaluate the inter-rater reliability of the number of sentence patterns in introductions of the articles.

Results and Discussion
This part presents the research results and discussion based on analyzing the sentence characters such as average sentence length and sentence pattern in the introduction section of the national accredited journal and reputable international journal written by national authors in Applied Linguistics.
There are 10 corpora sentence patterns taken randomly which need to be analyzed by the researcher and co-rater. After the researcher and co-rater completely analyzed the data, it was found that 276 out of 336 data was same agreement while 50 out of 336 data was random agreement. The result of Cohen's Kappa value is 0.91. Based on the category in the previous chapter, it can be concluded that the data agreement between researcher and co-researcher is excellent. As it has high reliability, the researchers continued the next research step.

The Average Sentence Length in Introduction Section
The first research question aimed to identify the average sentence length of the Introduction section in journal articles written by national authors published in internationally reputable journals and national accredited journals regarding the number of words per sentence and terms of the number of sentences per section. The tables below show the number of words per sentence. First, the results of reputable international journals are summarized in Table 2. The second is calculating the average length of introduction in terms of the number of sentences per section in research articles introduction written by Indonesian authors published in international reputable and national accredited journals. The results of the calculation are shown in Table 3.  Table 3 shows the number of sentences in the research article introductions in national accredited journals and reputable international journals. It shows that the average is 38 sentences per section in national journal articles and 34 sentences per section in international journal articles. Specifically, JEELS articles are the longest, with an average of 44 sentences in the introduction section. The second is from IJELTAL articles with an average of 38 sentences. The third is from the LINGUISTS articles with an average of 36 sentences. The least is from JOALL articles with an average of 35 sentences. Moreover, the longest sentence in international journal research articles is from the TEFLIN with an average of 45 sentences. The second is from IJoLE articles with an average of 39 sentences. The third is from IJAL articles with an average of 34 sentences. The shortest is from the SiELE articles, with an average of 19 sentences.
The first objective of this research is to calculate the average sentence length of the introduction section in Applied Linguistics research articles written by Indonesian authors. The first result indicates that the average sentence length in terms of the number of words per sentence in research articles introduction of national journals and international journals is very close (21.83 in national journals and 22.38 in international journals). The scientific writing should be around 12-17 words per sentence (Griffies, Perrie, & Hull, 2013). Meanwhile, the average sentence length of those research articles is included in the readability level which the average of each sentence is around 20-25 words per sentence (Garner, 2000). It can be implied that Indonesian authors have followed the international conventions in writing sentences in terms of the number of words per sentence. The result in line with Deveci (2019) found that the average sentence length within 24.17 in the corpus of the research.
Furthermore, the second result indicates that the average length of introduction in terms of the number of sentences in the introduction section is about 38 sentences for national journals and 34 sentences for international journals. Author guidelines in JOALL (2021) as a national journal indicate that writing articles should use more than 318 sentences or more. Technical guideline in TEFLIN (2021) as an international journal implies that the number of words used in the article is 218 to 348 sentences or more. It indicates that Indonesian authors in national and international journals have followed the Journal of Language and Literature Vol. 22 No. 1 -April 2022ISSN: 14102580-5878 (online) guidelines and used long sentences in the introduction section. The choice of sentence length in the journal articles can help the writers to establish the position (Dean, 2000). However, making sentences shorter or longer does not indicate a certain level of proficiency (Lestari & Lertira, 2013). Using longer or shorter sentences alone does not help present a good writing composition. The long and short sentences should be combined to present an ideal variation of easy-to-read patterns.

Sentence Pattern in Introduction Section of National Journals and International Journals
The second result of the study is aimed to identify the common sentence patterns found in research articles written by Indonesian authors published in nationally accredited and internationally reputable journals. The sentence patterns found are simple sentences, compound sentences, complex sentences, and compound-complex sentences. The researcher analyzed forty introduction sections of research articles from national accredited journals and reputable international journals. The results of identification are summarized in Table 4 below. Table 4 shows that sentence patterns, namely simple, compound, complex, and compound-complex sentences, are found in nationally accredited and internationally reputable journals' introduction section. Furthermore, the most common sentence pattern is complex sentences, with 414 sentences in national journals and 378 sentences in international journals. The second is simple sentences with 170 sentences in national journals and 136 in international journals. The third is compound-complex sentences, with 106 in national journals and 112 in international journals. The minor sentence pattern used in the introduction section is compound sentences with 76 sentences in national journals and 59 sentences in international journals. The second research objective identifies the common sentence patterns found in the research articles introduction in Applied Linguistics written by Indonesian authors published in national and international journals. This study found that complex sentences become the typical sentence patterns that Indonesian authors use in the introduction section of national and international journals. Simple sentences followed it as the second common use of sentence patterns. The result implies that Indonesian authors in national and international journals could make complex sentences because they used simple sentences and combined with at least one dependent clause. It also can make sentences more efficient and avoid the redundant of the same sentences or repetition in writing. Good writing in the scientific paper should be clearly and precisely in defined opinion to avoid imprecise words (Wyrick, 2017). Thus, the result of this study, in line with the result of research by Deveci (2017), indicates that language learners use complex sentences more commonly than other sentence patterns to display writing ability. He also found that simple sentences were indeed preferred by Journal of Language and Literature ISSN: 1410-5691 (print);2580-5878 (online) Ameitha Pratiwi, Safnil Arsyad, & Syafryadin authors who were familiar with English compared to national writers to improve writing quality.
On the other hand, the Indonesian authors in national and international journals rarely use compound-complex sentences and compound sentences in the introduction section. According to Weber and Brizee (in Deveci, 2019), compound sentences are used to combine more than two independent clauses or parallel sentences, and compoundcomplex sentences are multiple independent clauses and one or more dependent clauses. It is probably caused by the introduction of the research article containing more overview of the general topic, available research information, and literature, so they rarely use very long and parallel sentences. Wijen (2014) observes that the sentence pattern in the introduction section is used according to the idea or thought of the author. Moreover, mixing all kinds of sentence patterns such as simple, compound, complex, and compoundcomplex sentences in writing can have a good writing style (Oshima and Hogue, 2006). The result was in line with Deveci (2019), indicating that compound sentences were used the most in the method section because the author tended to be direct and concise with information describing how to obtain results. Compound-complex sentences are used the most in the discussion section. So, this is possible that compound sentences and compound-complex sentences are rarely found in the introduction section.

The difference between National and International Journals
The last objective aimed to answer the third research question about the differences between the nationally accredited and reputable international journals regarding sentence characters. The results indicate that complex sentences are the most common sentence pattern found in both groups, and it followed by simple sentences, compoundcomplex sentences, and compound sentences. Mamishev & Williams (2010) imply that using sentence patterns by authors can make writing more sophisticated. It also found that Indonesian authors in national and international journals used long sentences in the introduction section. It implies no important difference between national accredited and international reputable journals regarding sentence characters. The results found more similarities in sentence characters while only a few differences. The differences could be because the authors lack concentration on the writing quality caused by unfamiliarity with English (Wallwork, 2016). It makes the author less meticulous with the word used (Newell, 2014).
However, the researchers in Indonesia rarely study sentence characters, namely sentence lengths and sentence patterns. So, the results of this study are expected to provide some advantages to help the author write the research article introductions. The advantages of this study can help the author make various sentences with complexities to make sentences more efficient. Combining the author's average sentence lengths can help determine whether the level of articles is made following national or international writing conventions. The last analysis in this study is to establish the differences of sentence patterns in the national accredited journal and reputable international journal in terms of the number of sentence patterns.

Conclusion
From the results and discussion of this study, several points are concluded as follows. First, in the introduction section, Indonesian authors in the national and international journals used long average sentence length in terms of words per sentence. Furthermore, the average length of introduction in terms of sentences per section found that Indonesian authors in national accredited journals and reputable international journals use long sentences in the introduction section. It can be implied that Indonesian authors have followed the community conventions of writing and technical guidelines in writing sentences.
Second, the most common sentence pattern found in national accredited and international reputable journals was complex sentences, followed by simple sentences. Complex sentences were consistently found in the introduction section of research articles. It Journal of Language andLiterature Vol. 22 No. 1 -April 2022 ISSN: 1410-5691 (print);2580-5878 (online) interprets that the complex sentence is an obligatory sentence pattern used in the corpus. Otherwise, the rare sentence patterns found in the corpus were compound-complex sentences and compound sentences. This results can be assumed that the authors have used varied sentence patterns in writing the introduction section.
Third, there is no important difference between Indonesian authors in national accredited journals and reputable international journals regarding sentence characters. It is found that Indonesian authors in national and international journals used complex sentences more often, and both groups used long sentences in the introduction section. Thus, this implies that Indonesian authors in national journals probably have written following international writing standards.