The Use of Linguistic Features in Indonesian Students’ Texts at Hungarian Universities

Students enrolled in a higher level of education may face challenges in producing assignments, including writing an essay. They need to present appropriate linguistic features in the text to demonstrate their writing quality. Comparing linguistic features during university study is one fundamental aspect of administering writing quality and showing student writing development. The topic presented in this study describes an initial statistical analysis and the frequency of linguistic features in the texts produced by Indonesian graduate students enrolled at Hungarian universities. This study proposed two research questions: 1) How frequently do pre-selected linguistic features appear in the texts of Indonesian graduate students? 2) How do these features appear in a paired T-test statistical analysis? Seven MSc and MA graduate students took part in the study to meet the goal. They are students at three different universities in Hungary, which all have a Social Science Faculty. During their second-year studies, fourteen essays with a minimum text length of 2000 words were collected. The findings revealed an increase in conjunctions, adjectives, abstract nouns, concrete nouns, noun phrases, expanded noun phrases, active verbs, verb phrases, and passive forms. Meanwhile, the frequency of linguistic features such as prepositions, definite articles, indefinite articles, noun clauses, adverbial clauses, and adjective clauses decreased. In addition, both the rise of nine linguistic features and the decrease of six features in the second semester are not statistically significant.


Introduction
Internationalization has been exalted as a boon to global education, notably higher education. This objective is consistent with the mission of the Stipendium Hungaricum scholarship program, which was established in 2013 by the Hungarian government to promote Hungarian higher education globally and to attract outstanding international students from all continents who can develop personal and professional ties to Hungary while receiving a high-quality education in the heart of Europe (Web1, 2020 https://stipendiumhungaricum.hu/ about/). This program is based on bilateral education agreements between Hungary and the sending countries' governments on five continents and is already available in nearly 80 countries, including Indonesia.
In the early stages of the agreement between the two countries in 2016, 50 students were eligible for the scholarship program. Meanwhile, Indonesia has had the benefit of a doubled increase since 2020. According to the data released by the Indonesian students association in Hungary, the total number of students enrolled at Hungarian universities is more than 200 students (Gariahub, 2021). Due to this circumstance, English has become fundamental for the students to communicate during their academic work. Hence, English as a medium of instruction is a necessary component of education, even more so when students study in non-native Englishspeaking environments (Macaro et al., 2018).
As international students register in Hungarian higher education, they must meet the university's English proficiency requirements.The Council of Europe published the Common European Framework of Reference for Languages (CEFR), which describes language learners' speaking, reading, listening, and writing abilities at six 132 Subandowo,D. reference levels (University of Cambridge of ESOL examinations, 2011, p. 4). The levels are A1 (breakthrough), A2 (wastage), B1 (threshold), B2 (vantage), C1 (effective operational proficiency), and C2 (mastery). In most cases, Hungarian universities require students to have a B2 or higher (or an IELTS score of 5.5 -6.5, see Stipendium Hungaricum Programme Operational Regulations, 2021) to enrollin a master's program in the social sciences faculty. Nonetheless, the requirement for an English proficiency certificate does not appear to be the only high-stakes assessment of students' success. Singh (2015) discovered that the stakes are high for international graduate students enrolled in Master's programs. For non-native English speakers, English as a medium of instruction may add to the academic difficulties they face. Additionally, Hyland (2006) stated that most graduate students will encounter lectures, seminars, and exams and are required to take notes, present, and write assignments. Accordingly, the latter issue considers being pertinent to academic writing.
Indonesian students may encounter enormous difficulties in academic writing involving language functions, including the use of linguistic features. Linguistic features are critical for indicating text quality in academic writing and academic success (Fang, 2005). The texts with a higher score are more likely to contain linguistic characteristics associated with sophisticated language (McNamara et al., 2010). Staples et al. (2016) and Swales (1995) observed that informational density and nominalization appear to be inherence to linguistic features. Hence, the current study looks at the progression of linguistic features used in Indonesian students' English academic texts over two semesters. The findings are expected to give prominence to the differences in linguistic features students use in their residency.

Aim and research questions
The primary objective of this study is to compare Indonesian graduate students'academic writing across two semesters in terms of pre-selected linguistic features and to determine whether these features change during their residency. To accomplish this goal, two preliminary questions are formulated as follows: 1) How frequently do pre-selected linguistic features appear in the texts of Indonesian graduate students? 2) How do these features appear in a paired T-test statistical analysis?

Literature study
This session concerns issues about the current topic. The initial discussion begins with English as a medium of instruction in the European countries involving Hungary. It is followed by a general overview of writing and English academic writing. The last session highlights linguistic features in L2 English academic writing.

English as a Medium of Instruction (EMI) in Higher Education
The spread of English is intimately connected to globalization (Earls, 2013). Globalization has an impact not only on language use but also on the economics of higher education. The process of globalizing English appears to be a wild and woolly affair, the result of a slew of decisions made by editors, teachers, students, parents, writers, publishers, translators, officials, scholarly associations, corporations, and schools with an equally diverse set of motivations (Montgomery, 2004). It appears to be a complicated phenomenon with positive and negative social consequences, encompassing economics, culture, identity, politics, and technology (Block, 2010, p. 290).
EMI is relevant to "the use of the English language to teach academic subjects (other than English itself) in countries or jurisdictions where English is not the first language (L1) or is not spoken by the majority of the population" (Macaro et al., 2018, p. 9). Due to the increasing dominance of English and the intensifying internationalization of higher education, EMI in higher education has grown too muchover the last two decades or so (e.g., Coleman, 2006). There appears to be no doubt that English has become the standard medium of instruction in higher education in many countries, including several where the language does not yet have official status (Crystal, 1999, p. 6).
Many tertiary institutions, including universities, are increasingly using EMI (e.g., H. Coleman, 2011;J. Coleman et al., 2018;Macaro et al., 2018). It seeks to increase the mobility of both students and faculty members to make them more competitive and employable in international settings and achieve attractive and reputable institutions. Furthermore, Macaro et al. (2018) explain that the growing phenomenon of EMI in higher education establishes clear links with institutional aspects involving perceived internationalization, foreign student attraction due to decreasing enrolment numbers of home students as well as due to demographic changes, national cuts in higher education investment, competition between state and private sectors, and the use of English as the international language, specifically in the academic sphere.
It is now a fact that an increasing number of universities are willing to offer both undergraduate and graduate programs in English (Doiz et al., 2014). Björkman (2011) investigated the increasing use of English as a lingua franca as a medium of instruction for higher education in continental Europe and elsewhere. The study focused on new group learners who primarily require English to communicate with speakers from other first language backgrounds. Another field of study, conducted by Petzold and Berns (2000), shows that English is increasingly being used as a powerful medium in education in Hungary. It has a significant impact on learning, particularly in Budapest higher education. The study results reported that university students had contact with English through current knowledge by reading various texts such as professional journal articles, reference works, and textbooks.
EMI is being implemented at all higher education levels, notwithstanding a program for international students that includes Indonesian students. According to Lamb et al. (2021), the requirement of standard English proficiency levels and investment in staff training, including appropriate educational technology, are critical factors in the spread of EMI in HEIs.
Dearden (2016) has undertaken a well-cited survey on the growth of EMI in 60 countries and discovered that there was a widespread phenomenon of rapid expansion of EMI and an expectation that it would continue to grow with both private and public education. In Europe, for example, there has been a significant shift away from mother tongue-medium higher education courses and toward English-medium courses. According to Broggini and Costa (2017), almost all postgraduate programs in Scandinavia are now taught in English. Wächter and Maiworm (2014) also state that between 2007 and 2014, the growth of EMI courses in Europe increased by 1,000 percent to accommodate an increase in the number of international students and demonstrate EMI courses. It is worth pointing out that the expansion of EMI in Europe is based on relatively strong foundations in terms of secondary school English language proficiency; for example, universities in Hungary require international students to have at least an upper-intermediate level of English (B2 on the CEFR or 6.0 of IELTS, see, e.g., Stipendium Hungaricum Programme Operational Regulations, 2021). Furthermore, EMI in Denmark has been discovered to serve as a marker or social distinction for middle-class students with higher self-confidence in English and as a means of career advancement (Lueg et al., 2015).

Writing
Writing is a paramount skill that all language learners should cultivate (Baghbadorani & Roohani, 2014). It is a thinking tool that allows students to express themselves, understand, and share their perspectives on their world (Clark, 2014, p. 6). Accordingly, writing is regarded as one of the most critical skills for students, particularly those in higher education. Chien (2012) asserted that writing is usually required for papers, reports, and theses, and it plays a vital role in both teaching and learning. Not to mention that writing also poses significant challenges for second language (L2) learners (Shofiya, 2004).
Writing development is required for university students because it has become a critical skill for their academic development. It is regarded as an essential skill for knowledge production and dissemination in any disciplinary discourse (Raoofi et al., 2017). It is the case that gauging students' academic achievement is heavily reliant on their ability to communicate their language knowledge and ideas and how they complete their assignments, improve critical thinking skills, and develop their cognitive performance (Graham et al., 2007). It is also important to note that the ability to write proper English has become critical for academic success in higher education, where English is the medium of instruction. Moreover, Raoofi et al. (2017) opine that English writing in academic circles enables students to share their research findings with global readerships and publish their ideas and research in an international outlet.

English academic writing
Writing refers to a skill needed by a language learner in various contexts for the whole of their life (Smith, 2008). It includes the writer's ability to compose something that could be documented and use that particular language wholesomely (Mohajeri, et al., 2013). To be more specific, the learner needs to convey clarity, coherence, logic, conciseness, and straight to the point to the readers. For example, L2 English students at the university level must have this skill because they need to write academic texts, including summaries, essays, research reports, article reviews, and theses and dissertations, which are part of their assessments. Several studies reported that good writing skills support someone's academic success, whether writing up research reports, preparing research papers, or taking essays test (e.g., Brun-mercer et al., 2015;Tsingos-Lucas et al., 2017;Gebhardt et al., 1989).
Academic writing, even so, dramatically differs from personal writing because of its own set of rules and practices. The word "academic" means having to do with higher education or a career at the university (Irvin, 2010;D. R. Russell et al., 2012). Biber and Gray (2016) maintain that the language of academic writing is peculiar, different from everyday speech, and different from most other registers of English. Furthermore, Tardy describes academic writing as a "transformation of knowledge," including readers' persuasion, significance, and credibility (2005, p. 325). Hyland (2002Hyland ( , p. 1092) explains that academic writing is "an act of identity" conveying disciplinary content and a representation of the writer. Hartley, for example, mentions that some characteristics of academic writing include technical, impersonal, authoritative, humourless, and more accessible for non-native speakers to follow (2008, p. 4). Taş (2010), on the other hand, states that academic writing is not merely a linguistic process but recognition in the social community they write for. According to Xu and Zhang (2019), academic writers need to confine their writing to disciplinary norms and conventions by choosing particular discursive features. Similarly, Staples et al. (2016) argue that academic writing deals with planned and edited language that is specific in a concise format. Hence, academic writing is a process in which the ideas are organized by specific rules to convey a writer's representation of disciplinary contents.
In the case of academic success, students at the university level are instructed to submit writing assignments. Al-Zubaidi (2012) argues that academic writing at the university level is relevant to students' success. Wilson and Glazier (2011) also state that successful foreign language learners in English writing will have better chances and benefits in their life-long careers. It indicates that writing is an essential skill that students need in their academic lives. Irvin (2010) points out that students who produced a successful piece of writing depended mainly upon their writing tasks, including an essay, thesis, or dissertation. On the other hand, Paltridge and Starfield (2007) argued that thesis writing is a challenging process, especially for L2 English students, due to limited language proficiency in critical thinking, genre, and social knowledge.

Linguistic features in L2 English academic writing
Writing is a fundamental component of higher education. It assists students in communicating their ideas and promotes academic success. On the other hand, writing an academic text appears to be challenging for English learners (Shofiya, 2004). They must demonstrate the ability to write in a second language in addition to mastering grammatical structures and amassing extensive vocabulary, as well as to distinguish academic and conversational English in terms of conventional words, phrases, and sentence structures (register knowledge; Biber et al., 2019). Besides that, students must write within their academic disciplines, utilizing specific linguistic features (Fang, 2005;Hyland, 2000;Swales et al., 1996).
Writing an academic text is inextricably linked to the academic genre in higher education. Hyland defines a genre as expressing oneself through language (2003, p. 18). He divides academic genres into several categories, one of which is student essays (see Hyland, 2008, p. 11). Correspondingly, Crossley (2020) stated that students' academic writing assignments are useful indicators of assessing linguistic features.
Regarding specific linguistic features, Fang (2005) examined some characteristics of scientific writing. The study discovered that lexical density and abstraction could be classified into distinct characteristics. The term lexical density refers to the ratio of different words to the total number of words in a text, sometimes used to measure the difficulty of a passage of text (Richards et al., 2010, p. 336). The lexical density illustrates how a text is lexically rich (Gregori-Signeset al., 2015). Subsequently, this feature can gauge students' progress in their writing.
On the other hand, abstraction refers to the writer's use of language features to transform assumptions about the world into abstract entities (Hyland, 2009, p. 7). Ferris (1994) asserts that L2 writers produced a more significant number of passives, nominalizations, relative clauses, adverbial clauses, and conjuncts in sentences. According to Hinkel (2009), English language learners (ELLs) from China, Japan, and South Korea used significantly more necessity and obligation models (must, should, ought to, have to, and need (to)). In addition to linguistic characteristics, Russell's (2014) comparative study demonstrates that university students frequently use modals, verbs, adverbs, adjectives, clauses, phrases, perfect aspects, passive forms, and adjective clauses.
Based on existing research, linguistic features in L2 academic writing are likely to vary. It looks likely impossible to examine all of the research's findings, as each researcher proposes distinct categories. However, for this study, pre-selected linguistic features based on the most occurrence cited by the prior studies focused on prepositions, conjunctions, definite articles, indefinite articles, adjectives, abstract nouns, concrete nouns, noun phrases, expanded noun phrases, action verbs, verb phrases, noun clauses, adverbial clauses, adjective clauses, and passive forms are considered from the literature based on their frequency of occurrence.

Research design and method
This present study is a quantitative study that includes statistical data analysis using the SPSS application and a paired T-Test study to compare two groups (Kim, 2015). Using a paired T-test in this study is to examine the difference in linguistic features between the first and second semesters.Seven Indonesian MSc and MA graduate students participated in the study. The students were enrolled in the faculty of Social Science at three different universities in Hungary, where they studied International Relations, Social Integrity, and Regional and Environmental Economics. All invited students who participated in the study had B2 English proficiency levels or higher and had signed the consent letter.
The selection of Indonesian students for the present research was motivated by several aspects. First, an increasing number of Indonesian students have had the opportunity to study higher education abroad through scholarships provided by the Hungarian government. Second, as international students, they are expected to write high-quality papers, such as essay assignments. The assignments are high-stakes assignments designed to help students succeed academically. Because English has become a second language for considerable Indonesian students enrolled at Hungarian universities, writing an academic paper seems to be an issue during their studies. Furthermore, research on Indonesian graduate students' English academic writing is still relatively small, particularly in a non-English speaking environment outside of the country's borders.
In terms of data collection, all students must submit a high-stakes assessment essay with a minimum text length of 2000 words, submitted and marked by their lecturers. Prepositions, conjunctions (e.g., coordinate and subordinate conjunctions), articles (e.g., definite and indefinite articles), adjectives, nouns, verbs (e.g., active verbs), phrases (e.g., noun phrases, expanded phrases, and verb phrases), clauses (e.g., noun clauses, adverbial clauses), and passive forms were among the 14 essays collected and analyzed from students. As for data analysis, an application such as Atlas.ti was used to count the frequency of selected linguistic features by uploading the essays to the program. Linguistic features were classified into fifteen occurrences of selected linguistic features. The analysis was begun by exploring students' frequency of linguistic features in two semesters and demonstrated the paired T-test statistical results. At the final stage, the analysis highlighted increases and decreased occurrences of the features between the two semesters.

Results and discussion
The data are typically distributed according to the statistical analysis (P-value > 0.05, Shapiro-Wilk test). The paired T-tests employed that the data fall into three major areas. Nine variables, or 60% of data, are increased in the second semester. Second, six variables, or 40% of data, are eliminated during the second semester. Third, however, the difference between the rise and fall in the writing development of Indonesian graduate students during their two-semester studies is not statistically significant. Table 1 shows that the three most frequent items in students' first semester texts are noun phrases, abstract nouns, and prepositions, equivalent to approximately 150 items. Additionally, at least three frequency of occurrence is sure to occur for noun clauses, adjective clauses, and passive forms with a total frequency of only about 20 items. On the other hand, a slightly different frequency of occurrence in the second semester makes the number frequency to more than 140 items, including abstract nouns, noun phrases, and prepositions. While the frequency of occurrence of adverbial clauses, indefinite articles, and noun clauses decreases during the second semester, they appear to be subjected to ten items.

Indonesian graduate students' frequency of linguistic features in two semesters
Overall, the data in Table 1 have shown little difference in the frequency of occurrence of linguistic features between students' first and second text semesters. Nonetheless, the increase in linguistic features such as conjunctions, abstract nouns, concrete nouns, expanded noun phrases, active verbs, and verb phrases occurs significantly in the second semester. Thus, the arrangement distribution of pre-selected linguistic features in two distinct semesters can be classified according to their frequency of occurrence, beginning with abstract nouns, noun phrases, prepositions, adjectives, concrete nouns, conjunctions, expanded noun phrases, active verbs, verb phrases, adverbial clauses, passive forms, adjective clauses, indefinite articles, and noun clauses.

Paired T-test statistical results
The preceding paragraph demonstrates that students' essay assignments have changed slightly in terms of linguistic features, not to mention statistical data analysis, as illustrated in Table 1. The frequency in which abstract nouns occur appears relatively high during their residency. The following paired T-test analyses highlight critical aspects of the study, including the normality test (Table 2), the paired sample test (Table 3), and paired sample statistics (Table 4).   Table 3 exposes the results of the paired sample test, which is used to compare two variables about the same subject. The test frequently consists of two variables separated by time. The findings show that the differences in Indonesian students' linguistic feature frequencies in the first and second semesters are not statistically significant (P-values > 0.05). The fifteen pair groups in table 4 illustrate the progression of the linguistic features for two semesters. The table demonstrates that the group falls into two broad categories: increase and decrease. The following chart proves the different categories with the total percentage of each group. The test showed that Indonesian graduate students used more linguistic features in the second semester. Hence, it is presumable that the development of students' academic texts does exist during the study.

Diagram 1. Paired sample statistics results
Based on the preceding discussion, the study finds that prepositions, conjunctions, articles, adjectives, nouns, verbs, phrases, clauses, and passive forms are the most frequently utilized pre-selected linguistic features in Indonesian graduate students' L2 English academic writing at Hungarian universities. The linguistic feature distribution consists of fifteen variables: abstract nouns, noun phrases, prepositions, adjectives, concrete nouns, definite articles, conjunctions, expanded noun phrases, active verbs, verb phrases, adverbial clauses, passive forms, adjective clauses, indefinite articles, noun clauses.
Diagram 1 demonstrates that the frequency of nine variables or 60 % of data, including conjunctions, adjectives, abstract nouns, concrete nouns, noun phrases, expanded noun phrases, active verbs, active verbs, and passive forms, is likely to increase. On the contrary, the frequency of six variables or 40 % of data relevance with prepositions, definite articles, indefinite articles, noun clauses, adverbial clauses, and adjective clauses decreases. The rise and fall frequency distributions in two semesters are not statistically significant.

Conclusion
The present study envisions two advantageous implications for Indonesian graduate students' academic texts during two semesters. First, the findings indicated that students paid close attention to producing English academic texts. They improved their writing quality by utilizing a variety of linguistic features. The use of abstract nouns and noun phrases as academic writing features, on the other hand, appears to be the focus of students' attention when writing academic texts. In this regard, it is worth noting that writing instruction focusing on features can assist Indonesian students enrolled in Hungarian higher education, particularly in the three disciplines, in improving their writing quality. Second, the study can examine the development of academic writing in a non-native English-speaking environment by conducting further writing assessments both as a product and a process. Moreover, corpus data analysis may help determine the development of students' English academic writing. The fifteen pair groups in table 4 illustrate the progression of the linguistic features for two semesters. The table demonstrates that the group falls into two broad categories: increase and decrease. The following chart proves the different categories with the total percentage of each group. The test showed that Indonesian graduate students used more linguistic features in the second semester. Hence, it is presumable that the development of students' academic texts does exist during the study.

Diagram 1.Paired sample statistics results
Based on the preceding discussion, the study finds that prepositions, conjunctions, articles, adjectives, nouns, verbs, phrases, clauses, and passive forms are the most frequently utilized pre-selected linguistic features in Indonesian graduate students' L2 English academic writing at Hungarian universities. The linguistic feature distribution consists of fifteen variables: abstract nouns, noun phrases, prepositions, adjectives, concrete nouns,