StudentsAo perception of Google Translate as a media for translating English Diva Aprisa Pangestika1.
Budi Kadaryanto2.
Dian Shafwati2 English Education Student FKIP.
University of Lampung English Education Lecturers FKIP.
Universitas Lampung Correspondence E-mail: divaap6690@gmail.
ABSTRACT
This research aims to describe students' perceptions of using Google Translate to translate English material.
In this research, researchers used the ex-post facto method.
The methodology used is quantitative, using a questionnaire with 40 Likert rating statements with 5 choices: strongly disagree, disagree, neutral, agree, and strongly The population in this research was English Language Education Students at the University of Lampung, and the sample consisted of four batches, namely the 2019, 2020, 2021 and 2022 classes.
Data analysis was tested using the Rasch model via Winsteps software.
Interesting findings were found that students showed different response patterns, while item analysis showed that there were six mismatched items and items in one construct measuring the same logit instrument with unidimensional and multiple item requirements as well as assessment validity, which indicated that ranking simplification would produce more measurement results.
Data shows that all students know about Google Translate as a medium for translating from one language to Students consider using Google Translate to translate and use it to learn pronunciation and increase vocabulary.
This means that students know other functions of Google Translate.
In the learning process, students of the English education study program have a lot of English language material.
In understanding English material, students admitted to using Google Translate to translate it into Indonesian and vice Keywords: perception, google translate INTRODUCTION Translation is not just changing one language to others, but also provide the right information to deliver the content well.
According to Newmark in Pujiati .
7: .
The concept expressed by Newmark refers more to translating not only diverting messages from the source language to the target language but also having to understand the target language the author wants to convey to avoid mistakes.
The same thing was expressed by Roger T.
Bell .
1: .
who defines translation as AuTranslation is the expression in another language .
r target languag.
of what has been expressed in another, source language, preserving semantics, and stylistic equivalencesAy.
Translation as a form of expressing a language in another language as the target language, taking into account semantics and equivalence.
This means that it is not only the message that needs to be considered when translating, but the translator needs to pay attention to the semantic and equivalence aspects.
Catford .
uses a linguistic approach in viewing translation activities and he defines it as Authe replacement of textual material in one language (SL) by equivalent textual material in another language (TL)Ay namely .
eplacing text material in the source language with equivalent text material in another language targe.
In researcherAos opinion.
Google Translate is the easiest and most translating tool for translators of various foreign languages it can even translate words, phrases, clauses, sentences and discourses.
Especially for some people who don't like to read.
Google Translate is like a thick dictionary that is more updated.
Because it takes a long time to translate if we use a dictionary.
We only need to use our smartphones, then continue by typing a few words that we want to translate Historically, translation was carried out by someone who is bilingual and equipped with special knowledge of the topic to be translated.
In the mid-20th century, textual theory and discourse analysis paved the way for modern translators who have linguistic transfer skills to apply the whole text approach .
ext/discourse as a whol.
In carrying out translation activities.
This was coupled with the digital revolution that started in the 90s and led to an increase in translation requests.
In the end, translation tools emerged, which in principle are translations based on Information Technology and Communication (ICT) (Garcia, 2009: .
Students or learners who need theory or learning materials from foreign sources can easily obtain translations through machine translation facilities, such as Google Translator.
Bing Translator.
Yandex Translator.
Babelxl, or Dictionary.
Even though there are several studies that mention weaknesses in the quality of the results of this Google Translate machine translation, the people and even students still feel the use of this Google Translate.
Translation using the Google machine is very practical, fast and cheap, but not yet close to a high level of accuracy.
Especially among students today who use translation services via Google because the internet can be accessed easily.
This proves that the translation results from Google Translate do not provide the right information.
However, most students still rely on Google Translate in translating English material or text.
Google translate can provide the right translation when the text being translated is only one word, but if the text is in the form of a sentence, the translation results are sometimes inaccurate.
This has no effects on the students of Lampung University so that until now they still use Google Translate in working on a text.
II.
METHODS
This research used Ex-post facto method.
The researchers obtain the data in the Lampung University.
The subject of the research are 2nd to 8th semester.
The data need in this are 70 students.
The data collecting are divided into two main steps.
In the first step the researcher spread the questionnaire.
The participant is completed the questionnaire asking about the use of Google Translate focusing on perception, problem, effectiveness and weakness with the aims to find out studentsAo perception on Google Translate as a Media for translating English Materials at Lampung University.
RESULTS AND DISCUSSION
Results of Quetionnaire In this research, researchers distributed a questionnaire of 40 statements.
The data obtained from this research were taken by distributing questionnaires to four The respondents of this research were students at the University of Lampung.
Based on the table, it shows that of the 70 respondents a total of eleven male respondents or about 7.
7% while the number of female respondents was fifty-nine U-JET.
Vol 12.
No 3, 2023 people or around 41.
Based on the table, it shows that forty-two respondents or 4% of the 2019 batch, seven or 4.
9% of the 2020 batch, four or 2.
8% of the 2021 batch, seventeen or 11.
9% of the 2022 batch answered the questionnaire.
Responden
Gender
Batch
Amount
Percentage Amount
7,7%
41,3%
Percentage 29,4% 4,9% 2,8% 11,9% Table 1.
Respondent Demographics Table 1 showed on the left side is the distribution of subject abilities while on the right side is the distribution of items.
From this map it can be seen that the variable map depicts a map of the distribution of respondents and the items answered.
From Table 1 it can be seen that the distribution of respondents and items that are at logit 0, e S2.
S3.
S1.
S4, is the average of the respondents' answers given.
There are statements that are difficult for respondents to answer, e.
i S3 the majority of respondents answered with low scores while for statements S8 and S2 respondents answered with high scores.
But from this variable map it can be concluded that the majority of respondents and items are at logit 0 or the average score of the respondents.
Person - MAP - Item
<more>|<rare> | S3
S1 S2
|S S1 S4
AP T| S1 S2 S2 S2 S2 S3 S3
EP NP | S2 S3 S6 S9
TP | S1 S1 S2
AL AP DP DP NP PL RP S| S7
GL JP LP NP RP M S2 S3
AP AP AP AP AP BP DP LP ML MP MP NP PL VP | S1 S1 S2 S3 S3
AP AP EP FP FP LP ML PP QP SP M| S3
AP IL MP MP SP | S1 S1 S3 S3
AL AP AP AP EP HL NP SP WL | S1 S4
AP DP DP FP IP KP ZP |S S2 S5
AP ML MP S
U-JET.
Vol 12.
No 3, 2023
AP RP | S3
DP | S1
IP T|T
NP |
AP S8
| S2
<less>|<frequ> Graph 1 distribution map of respondent and items According to Yazid .
stated that perception is an internal process that allows a person to choose, organize and interpret stimuli from its environment and these processes affect its condition.
this is related to students' experience of using Google Translate in the learning process, so from the results of the questionnaire students know that Google Translate is a media for translating.
The distributions of items after the questionnaire test is carried out is as follows:
Total No.
Aspect Translation The use of Google Translate (GT) Google Translate Aplication Sub.
Aspect Interest in Translating Difficulties in Translating Intensity of use Function Effectiveness How to use Impact of use Efficiency Facility Quality Benefit Lack Excess Frequency Percentage According to Bond & Fox .
the item's average logit value is 0.
0 logit which indicates that the instrument as a whole can measure.
The item average value of 0 logit is a random value assigned to indicate a 50:50 probability which is nothing but the same measure of the respondent's level of ability and the level of difficulty of the U-JET.
Vol 12.
No 3, 2023 item if it is found that the average logit item is not 0.
0 then the overall instrument is not Unidimensionalty In this case the Rasch model analysis uses Principal Component Analysis (PCA) of the residuals, which measures the extent to which the diversity of the instrument measures what should be measured.
Total raw variance in observations Raw variance explained by measures Raw variance explained by persons Raw Variance explained by items Raw unexplained variance .
Unexplained variance in 1st contrast -- Empirical -Modeled 4% 100.
0% 60.
4% 13.
Table 3 Standardized Residual In the table above it can be seen that the results of measuring the raw variance of the data are 46.
0%, not much different from the expected value of 40.
This shows that the minimum requirement of 20% unidimensionality is fit, but at the same time the Rasch limit of unidimensionality is fulfilled, which is above 40% (Reckase, 1979.
Linacre, 2.
Another thing that supports the unexplained variance is all below 8% which indicates the level of item independence in a good instrument.
------------------------------------------------------------------------------------------|ENTRY TOTAL TOTAL
MODEL| INFIT | OUTFIT |PT-MEASURE
|EXACT MATCH| |
|NUMBER SCORE COUNT MEASURE S.
|MNSQ ZSTD|MNSQ ZSTD|CORR.
EXP.
| OBS% EXP%| Item | |------------------------------------ ---------- ---------- ----------- ----------- ------| | 8 293 70 -2.
A-.
S8 |
| 16 216 70 -.
B-.
S16 |
| 20 190 70 -.
C .
S20 |
| 13 196 70 -.
D-.
S13 |
| 9 158 70 .
E .
S9 |
| 25 298 70 -2.
F .
S25 |
| 23 234 70 -.
G .
S23 |
| 38 191 70 -.
H .
S38 |
| 24 134 70 1.
I .
S24 |
| 36 208 70 -.
J-.
S36 |
| 5 228 70 -.
K .
S5 |
| 7 171 70 .
L .
S7 |
| 3 147 70 .
M .
S3 |
| 2 159 70 .
N .
S2 |
| 1 166 70 .
O .
S1 |
| 34 253 70 -1.
P-.
S34 |
| 35 211 70 -.
Q-.
S35 |
| 22 151 70 .
98 -.
R .
S22 |
| 6 157 70 .
98 -.
S .
S6 |
U-JET.
Vol 12.
No 3, 2023 | 40 140 70 .
92 -.
T .
S40 |
| 19 197 70 -.
94 -.
94 -.
S19 |
| 4 222 70 -.
93 -.
93 -.
S4 |
| 27 151 70 .
93 -.
92 -.
S27 |
| 18 262 70 -1.
89 -.
89 -.
S18 |
| 26 169 70 .
S26 |
| 14 213 70 -.
S14 |
| 39 148 70 .
S39 |
| 12 134 70 1.
S12 |
| 29 180 70 .
S29 |
| 15 218 70 -.
S15 |
| 11 145 70 .
S11 |
| 37 199 70 -.
S37 |
| 33 126 70 1.
S33 |
| 17 169 70 .
S17 |
| 10 150 70 .
S10 |
| 31 198 70 -.
S31 |
| 32 154 70 .
S32 |
| 28 151 70 .
S28 |
| 21 152 70 .
S21 |
| 30 184 70 .
S30 |
|------------------------------------ ---------- ---------- ----------- ----------- ------|
| MEAN 185.
00 -.
02 -.
| 46.
| S.
| 10.
------------------------------------------------------------------------------------------Table 4 Conformity of Item .
tem misfi.
Table 4 shows the order of misfit items that provide interesting information.
There are 6 items that have a negative Point Measure Correlation value with a very small average measurement error 0.
05 logit.
To find out which items do not fit, it can find out by adding up the average mean squared infit .
ean Infit MNSQ), which is 1.
with its standard deviation (Infit MNSQ S.
), which is 0.
41, so the value is 1.
= 1.
so that the MNSQ Infit value greater than 1.
45 is an indication of inappropriate Discussion Overall, it can be said that the studentsAo perception about using Google Translate ass a media have positively impact.
In addition, there are several previous research about studentsAo perceptions on Google Translate as a media for translating English The first previous research.
Alam .
in his research state that there is a positive response in increasing vocabulary mastery and pronunciation.
This improvement is indicated by the better ability of users to understand foreign language texts and pronounce the correct vocabulary.
Besides that, it helps speed up vocabulary mastery because it can be use directly and practically, without the need to use an English dictionary.
It means that the use of Google Translate as a learning medium can improve one's vocabulary mastery and pronunciation.
The second research.
Krisna .
in his research, researchers want to prove whether Google Translate can help people improve their translation skills or not.
Based on the results of the researcher's field observations, the increase in the mastery and pronunciation of the community's U-JET.
Vol 12.
No 3, 2023 vocabulary is help by the Google Translate application.
Maulida .
in her research all students know about Google Translate as a service for translating from one language to another.
Nevertheless, they can be used to learn pronunciation.
There was only one student who cite other uses of Google Translate besides translating, which is to learn pronunciation and increase vocabulary.
They are greatly helped by this application to discuss lecture material delivered in English, to understand which they have to translate it first.
Therefore, the differences of this research with previous research is that researchers focus on the easiness and ability of Google Translate in Translating English Material.
However, some students still don't understand the working system of the Google Translate application.
They don't understand that the working system of this application is translating words.
Therefore, there are still students who translate sentences using Google Translate by directly entering the translated sentence and then trying to match the translation results with the context.
It can also be concluded that the findings of this research are different compared to several previous studies which only focused on the use of the Google Translate application (Alam .
Krisna .
, and Maulida, .
IV.
CONCLUSIONS AND SUGGESTIONS
Conclusions This research present on the development and validation of students' perceptions of google translate as a medium in learning English.
A total of 40 statements were identified, where each construct consisted of two items arranged in this research instrument by providing five answer choices in the form of a semantic differential.
many as 70 Lampung University students from four batches voluntarily filled out research instruments.
The results of the analysis using Rasch modeling show that at the instrument level, all items and respondents show a good level of reliability at the respondent's level, it appears that the pattern of responses with variants from strongly agree to disagree with the use of Google Translate as a learning media in English.
The results of item tested show that there are items that are misfit and items from the same construct that have the same logit value so that the previous two items need to be rearranged, while items with the same logit need to create a new item.
According to Linacre .
research instrument in this study shows that it measures what it should measure, that is it fulfills the unidimensionality requirements.
Suggestions From the findings presented in chapter four, there are several suggestions as follows:
In this research, most of the 2019-2022 English Education Department students agree that Google Translate is appropriate for use as a media for translating English material.
Lecturers should use Google Translate as a media for language learning teachers must also use other learning media and collaborate with Google Translate to make the learning process better.
In this research.
Google Translate makes it easier for students in the learning For students, students should use the Google Translate feature well in language learning so that students can increase their knowledge and abilities in language learning.
In this research, researchers focused on the ease of acceptance of Google Translate and the performance of Google Translate.
For future researchers, it is better to U-JET.
Vol 12.
No 3, 2023 focus on other cases of Google Translate in language learning and use this research as a preference to find out more about Google Translate in language learning as a learning media.
REFERENCES