Journal of Languages and Language Teaching, Vol. 5 No.1, May 2017 6 THE VALIDITY ANALYSIS OF ENGLISH SUMMATIVE TEST OF JUNIOR HIGH SCHOOL Baiq Della Triastiwi Putri SMP Darul Hamidin Padamara, Lombok Timur, NTB dellaputri@gmail.com Abstract This research was aimed at analyzing the English summative test validity at SMP Darul Hamidin Padamara. The object of the research was the English summative test given to seventh grade students in the academic year 2016/2017 used descriptive method. The instrument used to analyze the data is documentation such as the English summative test, syllabus, blueprint and students answer sheet. The researcher matched the English summative test and syllabus to find out the content validity, for construct validity the researcher use blueprint and the last is students answer sheet to find out the criterion validity, to support the data the researcher used questionnaire. The finding of this research showed that the English summative test of SMP Darul Hamidin had validity in term of content and construct validity because 96% of test matched with indicators and for criterion validity was not valid because out of 50 items only 10 items were valid. Key Words: Analysis, Validity, English Summative Test INTRODUCTION One of the most important aspects of teaching learning process is evaluation. According to Harlen (2007: 12), evaluation is more often used to denote the process of collecting evidence and making judgments about programmed, systems, materials, procedures and process. Through evaluation teachers are able to find out the effectiveness or failure of a method and also students’ achievement in mastering the lesson. The students and their learning process is the main focus of classroom evaluation while the teachers need to hold an evaluation to measure the students’ ability in teaching learning process. Test is one of the forms to evaluate students’ ability which is usually used by the teachers to know how far the students master the lesson. In constructing the test, teachers should know and master the principles and also the steps that will be done in making a test because it is will help the teachers increasing the teaching learning process. In order to measure accurately, the teachers should use a good test which the one of characteristics of good test include validity. According to Hughes (2003: 26) that a test is said to be valid if it measures accurately what is intended to be measures. In other word, validity is the most important consideration in test evaluation. In fact, the result of researcher’s observation in teaching practice on 25th July 2016 at SMAN 1 Keruak there were some problems occur that dealing with the test itself. During the observation, the researcher found that there were some students could not pass the test. Moreover, for the reason of students’ failure, the students and the teacher blamed each other. The teacher argued that it was caused by the students’ limitation in knowledge and they did not study for the test. On the other side, almost all of students gave a contrast reason with their teachers. Although, some of students confessed that their failure was caused by not study on the test. But most of them did not satisfy with Journal of Languages and Language Teaching, Vol. 5 No.1, May 2017 7 the test that the teachers made. They thought that it was too difficult and the content of the test is not appropriate with materials that taught before. As the researcher elaborated in the previous paragraph, it is really important to make a test to be well constructed because it will make the teacher know their students better. The seventh grade is the beginning class of Junior High School. So, by giving the well constructed test for the students, the teacher will know the general information about students’ ability as whole and it will help the teacher find the suitable steps in treating the students in class. Based on those problems, the researcher takes an initiative to write a proposal about the analysis summative test in order to find out whether the quality of the English summative test that the teachers made will good or not. Therefore, the researcher decide to choose this subject for her thesis proposal, entitle “The Validity Analysis of English Summative Test of Junior High School” RESEARCH METHOD Research Design This study used descriptive research. According to Kothari (2004: 2) descriptive research includes surveys and fact-finding enquiries of different kinds. Descriptive research can be either quantitative or qualitative. It can involve collections of quantitative information that can be tabulated along a continuum in numerical form, such as scores on a test or the number of times a person chooses to use a-certain feature of a multimedia program, or it can be describe categories of information such as gender or patterns of interaction when using technology in a group situation. Based on description above, the researcher aimed to analyze the validity of Summative English Test of Junior High School. Sources of the Data The data of this study taken from the test administered for the seventh grade students which used in the final examination test at SMP Darul Hamidin Padamara, the answer sheet and syllabus. Teacher’s information related to the document also used as the secondary data. All of the data get from the school. Population and Sample Population According to Suharsimi (2010: 173) the population is all the subject of study in which the data can be obtained. Therefore, the population of this research was all of the seventh grade students of at SMP DH Padamara and then there were 21 students taken as the population of this research. Sample. The researcher used no-rule sampling as a sampling technique. It means that the researcher take a sample without any rule since the population was homogeneous and have no selection bias (Barreiro and Albandoz, 2001: 4). Therefore, the sample of this study was 21 of seventh grade students of SMP Darul Hamidin Padamara. Techniques of Data Collection The data collected by using documentary. It means that the data which obtained in the form of something written or document. The researcher collected the data from the school and the teacher who made English test in the research place. Kinds of the data that collected such as blue print, English summative test, syllabus and students answer sheet. To support the data the researcher also used questionnaire. This study used a questionnaire to express opinions held by the subject. The total of questionnaire was 10 items of questions. Techniques of Data Analysis After collecting significant data, the researcher analyzed content, construct and criterion validity by using different way. To analyze the content validity of test, the Journal of Languages and Language Teaching, Vol. 5 No.1, May 2017 8 researcher use some processes in analyzing the data such as: First, the researcher determined indicators based on basic competencies which stated on English syllabus administered to seventh grade student. Second, the researcher matched the item of the test with indicators for each basic competency. After finding out the percentage of content validity, it was investigated by following table: Table 3.1 The level of content validity 76 – 100% Good 56 – 75 % Sufficient 40 – 55% Less good >40% Bad (Suharsimi, 1992: 313) RESEARCH FINDINGS AND DISCUSSION Research Findings The result of criterion validity of a test The researcher used students answer sheets to analyze the criterion validity and to calculate it, the researcher used point biserial correlation formula because kind of test for English final examination at seventh grade students is multiple choices. Based on result of the item analysis above, shows that criterion validity of English summative test for seventh grade students was not valid because from 50 items only 10 items which valid. The result of content validity of a test. For the content validity analysis, the researcher used the indicators of each basic competency which were elaborated in the syllabus to measure the coverage content of the test. The content validity of English summative test administered to seventh grade student of SMP Darul Hamidin Padamara was analyzed by comparing the item of the test with indicators of each basic competency which was reflected on syllabus or in other word by determining to what indicators the item administered. Most of the items measure the reading skill. The following table will give clearly description of the content validity of the test. Table 4.1 Analysis of Content Validity Basic Competency Indicators No. of item tested 1. Merespon makna yang terdapat dalam teks tulis fungsional pendek sangat sederhana secara akurat, lancar dan berterima yang berkaitan dengan lingkungan terdekat 1. Me ngidentifikas i berbagai informasi dalam teks fungsional pendek berupa: daftar barang 5, 6, 22, 23 2. Me ngidentifikas i berbagai informasi dalam teks fungsional pendek berupa: ucapan selamat dan surat pribadi 2, 3, 4, 19, 20, 21, 34, 35, 36, 37, 41 2. Merespon makna yang terdapat dalam monolog sangat sederhana secara akurat, lancer dan berterima untuk berinteraksi dengan lingkungan terdekat dalam teks berbentuk deskriptif dan prosedur. 3. Me ngidentifikas i berbagai informasi dalam teks fungsional pendek berupa: pengumuma n 7, 8, 9, 10, 24, 25, 26, 42 4. Me ngidentifikas i berbagai informasi dalam teks fungsional pendek berupa: instruksi 1, 18, 38, 39, 40 Journal of Languages and Language Teaching, Vol. 5 No.1, May 2017 9 5. Me ngidentifikas i berbagai informasi dalam teks fungsional pendek berupa: deskriptif 11, 12, 13, 14, 27, 28, 29, 45, 47, 48, 49, 50 6. Me ngidentifikas i berbagai informasi dalam teks fungsional pendek berupa: procedur. 15, 16, 17, 30, 31, 32, 33, 46 TOTAL 6 indicators tested 48 items Based on the table above, it showed that each items of the test covered the indicators of the English lesson. However, from 50 items, there were only 48 items included and 2 items were not in syllabus. The question number 43 and 44 ask the student to arrange the jumbled word into a good sentence. Thus, it can be concluded that these question not appropriate to syllabus because the indicator is not concluded into recommended indicator in syllabus at seventh grade student. It was found that the content validity of the English summative test administered to seventh grade student of SMP Darul Hamidin Padamara was 96%. Therefore, based on the level of content validity the test categorized good content validity. The result of construct validity of the test Table 4.2 Blueprint of Testing Reading Skill Indikator Jenis tes No. soal Total 1. Me ngidentifikas i makna dalam teks fungsional pendek: daftar nama barang belanjaan. Pilihan ganda 5, 6, 22, 23 4 items 2. Me ngidentifikas i makna dalam teks fungsional pendek: ucapan selamat dan surat pribadi. Pilihan ganda 2, 3, 4, 19, 20, 21, 34, 35, 36, 37, 41 11 items 3. Me ngidentifikas i makna dalam teks fungsional pendek: pengumuma n Pilihan ganda 7, 8, 9, 10, 24, 25, 26, 42 8 items 4. Me ngidentifikas i makna dalam teks fungsional pendek: instruksi Pilihan ganda 1, 18, 38, 39, 40, 43, 44 7 items 5. Me ngidentifikas i informasi dalam teks lisan monolog berbentuk: deskriptif Pilihan ganda 11, 12, 13, 14, 27, 28, 29, 45, 47, 48, 49, 50 12 items 6. Me ngidentifikas i informasi dalam teks lisan monolog berbentuk: procedur. Pilihan ganda 15, 16, 17, 30, 31, 32, 33, 46 8 items Based on table above, the researcher can see that summative test which administered for seventh grade student of Journal of Languages and Language Teaching, Vol. 5 No.1, May 2017 10 SMP Darul Hamidin Padamara has reached good construct validity because the questions and the indicators connected to each other. However, there are some questions not appropriate with indicators in syllabus such as item number 43 and 44 which the researcher talked before. Then, there are two questions were not having the answers such as item number 36 and 40. The answer for research question as presented on the first chapter was based on the result of analysis. It showed that the English test of seventh grade students in SMP Darul Hamidin Padamara has reached good validity because the content and construct validity was covered in final examination test. However, the criterion validity was not good it does not mean the whole test was not good too, it is proven with almost all of the questions connected, relevance, and also matched with indicators in syllabus. From the questionnaire, the researcher was found there were some students confess that all of the material which administered to seventh grade student was taught before, because from 21 students only three students argued that content of the test was not taught by teacher. Therefore, the researcher concluded that the content of the test was valid. For criterion validity also showed from the result of questionnaire they agreed that the test was suitable with their ability. From 21 students there were 5 students argued that the test was not suitable with them. For construct validity there were 5 students argued that the test seem definite each of the indicator or material that taught before. As presented on the background of study, this research was conducted because almost half of the student did not pass the test. Therefore, the researcher wanted to find out the reason for the students’ failure. Through the result of this study, it could be assumed that the reason for almost half of student did not pass the final exam was caused by they did not study on a test. DISCUSSION The result of this study that is the English summative test of SMP Darul Hamidin Padamara has valid in term of content validity because 96% test matched with the indicator. According to Brown (2003: 22) if a test actually samples the subject matter about which conclusions are to be drawn, and if it require the test-taker to perform the behavior that is being measured, it can claim content-related evidence of validity, often popularly referred to as content validity. This research is similar to research that was conducted by Nofiyanti (2011: 52) already proved that from 50 questions the test has only 5 invalid items. Hastuti (2009: 184) most of the items of English National Final Examination for Junior High School 2006/2007 match with Competence Standard and the Basic Competence of the English Syllabus. Septin (2016: 58) the teacher-made test constructed by the English teacher at SMP DH Padamara was classified as a good test in term of quantity because the percentage of content coverage was 88.23% which was classified had very high level of content validity. The result analysis in term of construct validity was valid because the questions and the indicators connected to each other. However, there are some questions not appropriate with indicators in syllabus such as item number 43 and 44. Then, there are two questions were not having the answers such as item number 36 and 40. According to Brown (2003: 25) any theory, hypothesis or model that attempt to explain observed phenomena in our universe of perceptions it called construct validity. The result of criterion validity analysis was not valid because its only 10 items which valid and 40 items invalid. According to Hughes (2003: 27) criterion validity is the degree to which results on the test agree with Journal of Languages and Language Teaching, Vol. 5 No.1, May 2017 11 those provided by some independent and highly dependable assessment of the candidates’ ability. The validity of the test also supported with the result of questionnaire. CONCLUSION AND SUGGESTION Conclusion As presented on research question, “Does the English Summative Test used of seventh grade student at SMP Darul Hamidin Padamara have validity or not?” Yes, valid in term of content and construct validity because 96% the test was matched with indicators. Then for criterion validity the test was not valid because from 50 items there were 10 items which were valid. Suggestion Based on conclusion above, The suggestions can be gives as follows: This research provides the information for those who are interested in developing a test. Actually, there are some characteristic of a good test and this research was only focused on validity. Therefore, for the next researcher, the researcher suggests to analyze the other characteristic of a test since there was only few of previous study analyzed the other characteristic of a good test. REFERENCES Baiq Septin Pebriantika. 2016. The Analysis of the reliability and validity of teacher-made test at the first grade students of SMP DH Padamara in academic year 2015/2016. Mataram: Universitas Mataram. Barreiro, P. L., & Abandoz, J. P. 2001. Population Sample (Sampling techniques). MaMaEusch. Brown, H. D. 2003. Principle Language Assessment. San Francisco: Addition Wesley Longman, Inc. Faturrahman Imran. 2015. Penelitian Experimental. Unpublish. IKIP Mataram. Hamid Darmadi. 2011. Metode Penelitian Pendidikan. Bandung: Alfabeta. Harlen, W. 2007. Assessment of Learning. London: SAGE Publications. Hastuti Handayani. 2009. An Analysis of English National Exam (UAN) For Junior High School Viewed from School Based Curriculum (KTSP). Semarang: Diponegoro University. Hughes, A. 2003. Testing for Language Teacher: Second Edition. United Kingdom: Cambridge University Press. Kothari, C. R. 2004. Research Methodology (Methods and Techniques). India: New Age International (P) Limited. David, M. M. 2009. Measurement and Assessment in Teaching (Tenth Edition). United States of America: Pearson. Nofiyanti, 2011. An Analysis on the Content Validity of the Summative Test for the First Year Student of Junior High School At SMPN 87 Jakarta. Jakarta: Syarif Hidayatullah University. Suharsimi Arikunto, 2010. Prosedur Penelitian (Suatu Pendekatan Praktik). Jakarta: Rineka Cipta. Weir, C. J. 2005. Language Testing and Validation. London: Palgrave Macmillan.