JURNAL BASICEDU An Analysis of Translations Techniques by Using (GT) Google Translate’s English into Indonesia in English Simple Sentences

The goal of this study was to examine the correlation between the accuracy and acceptability or also readability of Google Translation’s (GT) English to Bahasa Indonesia translations of simple English sentences. This study applied a correlational quantitative research approach using survey research. The samples of this study included all students majoring in the English Language Education Study department of the fourth semester and used a purposive sample technique of around 35 students from STKIP PGRI Sidoarjo. The results showed 3 sentences rated ‘very accurate’ (6%). 29 sentences (58%) were deemed ‘accurate’. Then, 11 sentences were ‘less accurate’ (22%). Lastly, 7 sentences (14%) were deemed ‘not accurate’. Furthermore, based on the SPSS 26 results showed the positive constant value of 78.884 indicates positive effect of independent variable (GT translation). If the independent variable changes by one unit, so will the GT translation variable. If GT translation (X) rises by one unit, questionnaire results (Y) rise by 0.324, or 32.4%. Regarding to the findings, translation using GT through basic English sentences on selected reading texts utilized at STKIP PGRI Sidoarjo was mainly positive (64%). The investigation revealed a remarkable occurrence that GT’s translations were mostly accurate.


INTRODUCTION
People can't avoid engaging with people from various countries who speak other languages when they live in the world today. As a result, interacting with people from various countries might be challenging. Bahasa Indonesia, for example, differs significantly from English in daily use among Indonesians. As a result, it will be difficult for Indonesians to communicate with those who speak English and may misinterpret their meaning . To overcome this issue, translation is the most effective tool available. Many people are in need of broadening their worldview as a result of global progress. Learning resources like books can help you learn about many countries. Foreign-language books make things harder because the information is unclear. To understand the books, learn the language. The best way to gain knowledge is to read translated books. As knowledge grows, so does the need for translators, driving up translations (Duff, 1994).
It is appropriate for educational institutions to simultaneously introduce and begin using AVT and (ICT) as a foundation for more advanced learning because translation has pushed the world toward a universalization of knowledge (Zafitri & Harida, 2017). That's significant since innovative growth of ICT is a key factor in facilitating the speed and transformation of knowledge to learners across the country more broadly. When it comes to the establishment of education policies, it is necessary to ensure that students have access to cutting-edge technology that is comprehensive.
The validity and effectiveness of translation as a method for learning a second language have long been debated. Translation had fallen out of favor with the language teaching forum until lately. For a long time, learning a foreign language through translation was deemed an inappropriate method of instruction (Brown, 2002). It was condemned for its close ties to traditional grammar translation. In the modern world, translation is still viewed as a form of mechanical linguistic transfer. Because it is not a conversational activity, it is still overlooked as a good technique for learning a foreign language. Many people find translation to be monotonous, pointless and time-consuming.
Translation is the replacement of one language's text with another's (Catford, 1974). A translator must locate equivalent source-to-target text when translating. A translator must understand the message's context before rewriting it in the target language. In order to successfully translate from one language to another, the translator must be fluent in both the source and (TL) target languages, which is not a simple undertaking. Furthermore, it is explained, regarding to the two types of translation based on the amount of text that has been translated. Both full and partial translations exist, in which the source language text is fully translated into the target language, as well as translations in which only a portion of the source language text is translated into the target language (Catford, 1978).
In the field of translation, machine translation (MT) has been developed in order to speed up the translation process. Translators' work evolves in lockstep with the advancement of technology. The information and technological revolutions have been the driving forces. Automated translation has seen a dramatic increase in demand, as well as new methodologies and requirements. It is clear from the history of MT research and implementations that several different types of MT have been the focus of quality assessment study (Ghasemi & Hashemian, 2016). These include example-based, open source, pragmatic, rulebased, and statistical MT. More than 100 languages supported by Google Translate (GT), which is a tool for translating various written documents from a language to another. An expression, text section, or entire web page can all be translated using this tool in addition to individual words.
GT uses human translations to identify the best possible translation pattern for a given text by scanning through multiple sources (Farahsani, Rini, & Jaya, 2020). A Google Translation's degree of accuracy is directly proportional to the number of human-translated texts that the service searches for. When we talk about rules for language, we're usually referring to the grammar principles that regulate English sentences. Subject, object, complement, and/or object of preposition all appear in a sentence. As the subject and verb agreement, a simple is the fundamental structure that may be added to a clause by employing the object or complement of a phrase in that form (Catford, 1974). Taking into account Catford's suggestion, the author has decided to utilize only a partial translation, concentrating on the translation of English basic sentences into Bahasa Indonesia.
Machine Statistics Translation (MST), on which GT is built and which operates by analyzing hundreds of millions of pairs of natural bilingual texts, is the foundation upon which MST operates (Koehn, 2009). This natural companion has the potential to act as a credible illustration of the application of language from one of the options that are offered. Translators, methodologists, and educators all have various views on translation. Controversy and criticism exist when it comes to its employment in foreign-language education . Several studies have been done throughout the years that either favor or completely reject the use of translation as a learning tool. That's the fundamental reason that translation was viewed as ineffective, unreliable, and pointless. The use of translation (TL) in language acquisition had to be avoided due to significant restrictions.
It is defined by Rahimi (2004) as a precise and complete description of the source message and its transmission. According to Rahimi, a translation is considered faulty if it accidentally leaves out material, includes information that isn't present in the source text, or makes errors in interpreting the text's meaning. According to Khomeijani Farahani (2005), the degree of correctness in a translation relates to the author's ability to accurately and exactly translate the source text. The degree of accuracy of a translation can be assessed in two ways, according to this scholar. As an initial step, you should look for and underline the essential (or content) terms in both your source text and the translation to see how similar their meanings are. In this case, the focus is on finding the most accurate translations for the source language keywords. It can be determined that the translation is satisfactory in terms of accuracy criterion, if the translator has succeeded in conveying the same message as the source text.
A regenerating device, like as a machine translator, needs to be reviewed in order to understand the system that the machine uses. This evaluation could serve as a jumping off point for the development of new machine translator models. Humanika (2002) mentioned there are five different types of evaluation that can be done based on the MT assessment objectives. These evaluation types include feasibility evaluation, internal evaluation, diagnostic evaluation, declarative evaluation, and operational evaluation. Nababan (2010) explained the term 'acceptability' refers to whether or not a translation has been disclosed in accordance with the norms, norms, and cultures that are prevalent in the target language, either on a micro level or on a macro level. This can be said about the translation on either level. The grammatical compatibility of the source material with the target language and the reader's reaction to the translated text both play a role in determining whether or not the text may be considered acceptable. When compared to acceptance, accuracy emphasizes the precision of message delivery, while acceptance is more concerned with reasonableness.
Related to the translation and its implementation toward MT, many researchers have done in conducting those focuses. Garcia & Pena (2011) revealed that these students benefited more from machine translation than high-fluency students, and they preferred to use it against their teachers' wishes. Using machine translation helped starting students converse better, too. Josefsson (2011) evaluated vocational students' translation strategies and attitudes. As a mobile phone tool for students, GT was faster and more accurate than standard dictionaries, especially for collocations, phrases, and technical words. GT makes students aware of their own learning, leading to more coherent texts. GT was less beneficial for grammar solutions. GT is the second most used online dictionary (Jin & Deifell, 2013). Due to a lack of grammatical explanations, learners utilize GT as a supplement to the online dictionary. Students say using online tools like GT enhances their reading and writing in foreign languages while minimizing their learning anxieties. Researchers were cautious about the new findings because the online lexicon lacked clear explanations and ignored context. Translation is crucial, especially in science, because it helps find equivalent terms in a discipline. Researchers have studied GT and most of them analyzed GT's errors in translating phrases or sentences (Costa, Ling, Luís, Correia, & Coheur, 2015), compared GT to human translators (Halimah, 2018), and used GT (Gestanti, Nimasari, & Mufanti, 2019). GT struggles with mechanical engineering jargon. Furthermore, research was done using descriptive-comparative human translation analysis with Keshavarz's model (Ghasemi & Hashemian, 2016). Following the paradigm, 50 English and 50 Persian sentences from Motarjem Hamrah were selected and translated using GT from Persian to English (TT1) and from English to Persian (TT2). He categorized errors as lexicosemantic, tenses, preposition, word order, verb group distribution, and passive voice. The most error-prone translation was lexicosemantic, with 42 English to Persian and 26 Persian to English elements. Concerning to its essential element of translation, so, this current research aimed to examine the correlation between the accuracy and acceptability of Google Translation's English to Bahasa Indonesia translations of simple English sentences.

METHODOLOGY
A quantitative design was used for this kind of research, and it was conducted in the field (survey research). The researchers wanted to conduct their field study using questionnaires so that they could collect field data. This sort of study falls under the category of quantitative research in terms of the measurement and analysis of the data. This is due to the fact that the data is presented in the form of numerical data, and it is evaluated through the use of statistical techniques such as SPSS 26. While descriptive interpretations were applied in the process of collecting data or information to be sorted, the nature of the problem was taken into consideration. This quantitative research was based on the research purpose of investigating, discovering, and describing the correlation between the accuracy and acceptability of Google Translation's English to Bahasa Indonesia translations in order to provide descriptive data. Next, the information was entered into the form's description field. The findings from the analysis of the correlation between the accuracy and acceptability of Google Translation's English to Bahasa Indonesia translations of simple English sentences data were given in a descriptive manner in order to provide a comprehensive and correct account of the situation.
In order to collect data from informants, the researchers utilized primary data sources, also known as primers and related to Sekaran (2013), the term "primary data" refers to the material that was gathered by researchers from first-hand experience. Sources of primary data were collected in the form of field data and were directly received from study subjects or informants who were students at STKIP PGRI Sidoarjo. These data were related to the accuracy of Google Translate (GT) in the translation process. After that, the secondary data for the translation process employing GT were obtained from selected reading texts that were utilized in the "English Translation" subject. The researchers used a population consisting of all students majoring in the English Language Education Study department of the fourth semester and used a purposive sample technique of around 35 students from STKIP PGRI Sidoarjo to conduct their study.
Through the use of a questionnaire, the method for collecting data was gleaned from the outcomes of the translation process that involved GT. The questionnaire, which was a method for collecting data, consisted of presenting the respondents with a set of questions or statements in the form of a Google Form and requesting responses from them. The utilization of the questionnaire was the method of data collecting in the field that was the most fundamental. In order to make conclusions about the study based on the findings of this questionnaire, the responses would be converted into numerical form (quantification), organised into tables, and statistically evaluated. Questionnaire was one of the data collection procedures that was employed in the research that was based on the Rensis Likert scale. The Likert scale divides responses into four categories: very accurate, accurate, less accurate, and not accurate. Scale and definition of the translation quality assessment of accuracy-rating instrument and acceptability are presented on the following The target sentence's meaning is not faithfully translated from the source sentence. Lexical items and links between phrases, clauses, and sentence elements are problematic.

not accurate
There are several aspects of translation in general, as well as the use of GT and the applications that are included in GT, which are utilized as indicators in the process of determining how accurate GT is. After that, an indicator variable serves as the beginning point for the construction of things formed instrument that statements based on the table mentioned.

RESULTS AND DISCUSSION
This study used a quantitative technique and a correlation research design to find out the correlation between the accuracy and acceptability of Google Translation's English to Bahasa Indonesia translations of simple English sentences. The following table provides more analysis of the results of the accuracy of translation at word level through 50 sentences on the selected reading texts.
The results of translation using GT: Very accurate: (3/50) x 100% = 6% Accurate: (29/50) x 100% = 58% Less accurate: (11/50) x 100% = 22% Not accurate: (7/50) x 100% = 14% The table 2 showed there were 3 sentences classified into 'very accurate' translation (6%). Next, it was found that there were 29 sentences (58%) were classified into 'accurate'. Then, in the category of 'less accurate', it was found 11 sentences (22%). Meanwhile, there were 7 sentences (14%) which classified into 'not accurate' category. Related to the findings, it was argued that the translation using GT through the English simple sentences on the selected reading texts as the English materials used at STKIP PGRI Sidoarjo was mostly in positive tendency (64%). Meanwhile, the negative tendency showed the percentage of 36%. The findings of the study, on the other hand, reveal a phenomenon that is quite remarkable. The greatest part of half of all of GT's translations were accurate.
Following the results mentioned, the purpose of the online questionnaire was to collect information from respondents who were presently enrolled or had completed the subject of "English Translation" at STKIP PGRI Sidoarjo. All of the students in the English Language Education Study department were given the questionnaire in order to ascertain the outcomes of the GT translation and the correlation between the accuracy and acceptability of Google Translation's English to Bahasa Indonesia translations of simple English sentences. As a result of the validity test that was conducted for this study, which depended on statistical analysis that was carried out with SPSS 26, it was determined that all 10 questions on the questionnaire were valid.
The reliability test was carried out with the Cronbach's alpha method, which was incorporated into the SPSS 26 software. The results of the test are presented in table 3. A variable's dependability is evaluated using a statistic called Cronbach's alpha. It is possible to evaluate the dependability of an indicator by determining whether or not it is greater than the threshold value of 0.7. (Ledesma et al., 2002). The following table contains the information that was used to arrive at the conclusion that the Cronbach's alpha value is 0.884. Because the value is more than 0.7, it is reasonable to believe that the data is "Highly Reliable". This is because the value above the threshold. The following findings were acquired from the survey that was analyzed, and the results of the reliability test are summarized in the table that can be found below. As was mentioned previously, the purpose of this research was to examine the correlation XXX. In terms of its correlation, the findings of the questionnaire were used to make a representation between the accuracy (the dependent variable) in this part of the study and the GT translation, which was collected in order to (independent variable). After that, we looked into them using a technique called linear regression. The findings of the statistical investigation of SPSS 26 are presented in the table that can be seen below.  Table 6, it could be argued the correlation (R) value is 0.382. A coefficient of determination (R 2 ) of 0.34 was generated from the output findings, indicating that the independent variable (GT translation) influences the dependent variable (the results of questionnaire) by 13%, with the remaining 87% influenced by other variables outside the variable X. As a result, an ANOVA test was conducted to see whether there was a statistically significant correlation between variable X (GT translation) and variable Y (the results of questionnaire). The following table summarizes the findings of the ANOVA study. The ANOVA test results from SPSS 26 show that F count = 2.072 with a significance level/probability of 0.036 < 0.05, which indicates that the regression model may be utilized to predict variable Y (the results of questionnaire). Based on these findings, it can be concluded that there is a correlation between variable X and variable Y. Lastly, the coefficients result from SPSS 26 was presented on the following table. Y = a + bX Y = 78.884 + 0.324X The result could be described that the positive constant value of 78.884 indicated that the independent variable (GT translation) has a positive effect. The GT translation variable will increase or be fulfilled if the independent variable increases or has an effect in one unit. The X regression coefficient of 0.324 indicated that if the GT translation (X) increases by one-unit, the results of questionnaire (Y) rising by 0.324, or 32.4%.
The fourth-semester students at STKIP PGRI Sidoarjo were introduced to the "English Translation" subject and associated classroom activities at the beginning to the end of the course. Based on a study by researcher Duff (1994), well-structured translation exercises can aid in the development of accuracy, clarity, and flexibility, three of the most important characteristics of any language learning: Instead of preparing future translators, translation activities aim to help students improve their command of the English language. Various ways of integrating different activities exist. Grammar and lexical exercises, on the other hand, are used to help students strengthen their grammatical and lexical abilities before they begin translation tasks. As a starting point for further speaking practice, translation is useful.
In agreement with what Dagiliene (2012) and Cook (2010) have indicated about the rise of the translation approach as a result of the necessities of the current world, this research provides further support that translation technology can be advantageous for foreign language learners. It also confirms other previous research (Groves & Mundt, 2015;Garcia & Pena, 2011), which stated that GT can be very effective for  (2011) that GT may not be the best instrument for grammar learning. Based on the data mentioned above there is significant evidence that GT can be utilized as a complementary tool to assist students with interpreting English language. These arguments supported the findings which showed 3 sentences rated 'very accurate' (6%). 29 sentences (58%) were deemed 'accurate'. Then, 11 sentences were 'less accurate' (22%). And 7 sentences (14%) were deemed 'not accurate'. According to the findings, translation using GT through simple English sentences on selected reading texts utilized at STKIP PGRI Sidoarjo was mainly positive of accuracy (64%). Negative trend was 36%. The investigation revealed a fascinating-phenomena that GT's translations were mostly accurate for English simple sentences. There are many different types of translation processes and products that include literary translation, technical translations, subtitling, and machine translation (Basil, 2004). Additionally, the term "translation" is sometimes used to include interpreting, which refers to the transfer of written texts. "Equivalence" is a phrase that may be translated from the definition provided above. Source language's meaning and context are identical to those of its closest natural language counterparts in both cases. Meaning and style are the first two things to consider. The message conveyed by the source language must be the same. If the target language is influenced by the source language, the reader of the translation will be confused. The explanations mentioned were support the findings of this study which indicated that readers are expected to understand the target language's meaning in order to accurately convey the meaning of the source language. This means that the final product of the translation must be comprehensible. The data of this study showed that the regression model may be utilized to predict variable Y (the results of questionnaire) and it can be concluded that there is a correlation between variable X and variable Y.
According to the results, the independent variable (GT translation) had a positive effect on the constant value of 78,884. If the independent variable rises or has an influence on a single unit, the GT translation variable will rise or be fulfilled. When the GT translation (X) is increased by one unit, a 32.4 percent rise in questionnaire (Y) findings can be expected. According to Leonardi (2010), a good translation should have a natural flow, re-create both the style and the context of the original text, and respect the conventions of the target language and its correlation. These findings were related to Leonardi's (2010) claims. Due to the fact that translation is considered to be a communicative activity, it requires dialogue to take place between the instructors and the pupils. Learners are strongly encouraged to talk about both the correct and incorrect ways to approach the translation work, as well as any issues that arise (Leonardi, 2009).
The conclusion that can be drawn from this finding is that the independent variable (GT translation) has a favorable influence, as shown by the fact that the constant value was positive (78.884) If the independent variable grows or has an influence in one unit, then the GT translation variable will either increase or be fulfilled. Based on the value of the X regression coefficient, which was 0.324, it was determined that an increase of one unit in the GT translation would lead to an increase of 0.324, or 32.4 percent, in the results of the questionnaire. However, in order to get rid of the issues that were brought up by the participants, more research needs to be done to provide concrete supporting evidence and practicality in the process both from the method and method in integrating renewable technology in translation, such as GT, can be the appropriate alternative for students who are not skilled in English to understand elements of English.

CONCLUSION
Everything can be done with technology and machines in this advanced day. The translation procedure is no different. A variety of system can be utilized to assist and facilitate the process. Among translation applications, Google Translate (GT) is the most famous and sustainability widely used. This machine translation (MT) program is supplied by Google. Based on the phenomenon and discussion findings, it can be