THE DEVELOPMENT OF THREE TIER TEST (TTT) INSTRUMENTS TO IDENTIFY STUDENTS’ MISCONCEPTIONS ON MATERIAL IN ELEMENTARY SCHOOL

The purpose of this study is to develop the assessment instruments used in testing the level of science misconceptions in high grade elementary school. The instrument used is a Three-tier test (TTT) This research is a Research and Development (R&D) research that is a method of conducting research for research that produces a product, and testing the effectiveness of the product. The research stages are as follows: (1) Data Collection and Literature Studies, (2) Product Description and Design, (3) Small Scale Trials, (4) Product Analysis and Revision, (5) Broad-Scale Trials, and (6) Product Analysis and Revision. The research conducted at SDN 5 Sukagalih, Tarogong Kidul District, Garut on a small scale and for a large scale at SDN 1,2,3,4 and 5 Sukagalih, Tarogong Kidul District, Garut. Based on the results of the research obtained data that by using TTT instruments teachers can find out how much quantity of students are misconceptions, trials are conducted on a small scale and large scale and each is carried out 2 stages, namely the stage before the revision and after the revision of the instrument. The results of phase I tests on a large scale many students experienced misconceptions of 62.7% and the phase II test after instrument revision showed a significant decrease to 32.65% of students who experienced misconceptions, the presentation of a decrease in the quantity of misconceptions for large scale was 52.86%. the result was quite significant compared to the small scale which decreased by only 24.61%. The results showed that by using TTT techniques teachers will easily identify the quantity of students who misconception and for the next the teacher will provide treatment to students to lower the quantity.


A. INTRODUCTION
In classroom learning especially science, not a few students build conceptions that are irrelevant to scientific facts. So in the end the concept that was built is wrong and it is commonly called misconceptions 1 , some other terms such as preconception 2 and common sense concepts 3 . The erroneous knowledge is owned continuously and is believed to be the truth to explain various phenomena 4 . According to Berg the main principle in misconception correction is that learners are given a learning experience that shows their concept conflict with natural events, a contradiction of new experiences with the old concept 5 . Another point of view is that misconceptions are the idea children "have incorporated into their cognitive structure that they use to understand and make predictions about the world 6 . At the elementary school level in identifying misconceptions and understanding concepts, a special test is also required, one of which is the Three Tier Test (TTT). The advantage of TTT is that it does not take much time and can be used to know students who do not understand the concept of answering questions by guessing 7 . There are categories of TTT as in the form of the following  Research at the basic level on material forms of objects has also been carried out in MIN Cicendo Bandung 9 and the results are many siswqa that also experience misconceptions and caused by several factors, one of which is the method used in teaching teachers as long as kuran material can be understood by students.
According to the standard content of subjects 10 material form objects in elementary school in general consists of: (1) Various forms and properties of objects (2) Changes in the form of solid objects, liquid and gas consisting of melting, disfies, freezes, evaporates, condenses and crystallizes.
The expected result of this research is the extent to which the instruments developed can know the condition and quantity of misconceptions of elementary school students so that if the numbers found are alarming then it needs to be  This research is a research and development (R&D) is a method of conducting research for research that produces a product, and testing the effectiveness of the product. The research stages are as follows: (1) Data Collection and Literature Studies, (2) Product Description and Design, (3) Small Scale Trials, (4) Product Analysis and Revision, (5) Broad-Scale Trials, and (6) Product Analysis and Revision. The resulting product is a form of Three Tier Test instrument in the form of objects in high-grade elementary school, which has been tested both on a small and large scale.
The instrument used is a form of Three Tier Test (TTT) where each question item has 3 stages of answer. After the results of the answers collected and then processing the data using quantitative analysis. The subject of this study was a high-class student at SD Negeri in the tarogong area of Garut.

B. RESULTS AND DISCUSSION 1. RESULTS
This research was conducted in the high class of SDN 5 Sukagalih for smallscale data collection, then for largescale data collection was conducted at SDN 1, 2, 3, 4, and 5 Sukagalih Tarogong Kidul Subdistrict in Garut Regency in March-June 2020. Data Number of Instrument Items used to measure the quantity of misconceptions consists of 19 question items as seen in the following table:  , 5, 6, 7, 8, 9, 10, 15, 16, 17 1, 2, 3, 11, 18 12, 13, 14, 19 19 The validity test of the instruments in this study was based on expert judgement. As Fraenkel and Wallen stated that validity can be done by someone who will see the content and format of which instruments are appropriate and which are not, a person in question is a person who knows about what will be measured so that the instrument is feasible, therefore the validity of the question and indicators is carried out by lecturers who are experts in their field 11 . In identifying the occurrence of misconceptions, the Three Tier Test (TTT) method with three levels of testing is applied.
The decrease in the quantity of misconceptions in question is a decrease in the number of students who experience misconceptions in each concept tested. To find out the decrease, the formula adapted from the normalized gain formula developed by Suhandi is: The distribution of scientific concepts and misconceptions contained in the instrument details are as follows: Air consists of various gases and is not separated between the dangerous and the harmless Air consists of one substance and differs from gas, because the air is good and the gas is toxic.

12, 13
12 Gas is composed of material that is very small in size and has a very small mass so that the overall mass of gas is very light.
Gas has no mass or weight. 14, 17 13 Dissolution involves two types of so-called sugar dissolving substances in hot water as solvents, while melting is different from dissolution.
Dissolution and melting are the same events. 15 14 Decomposing objects, the material will decompose into simpler substances and be released into the environment in The rotting object will disappear. In addition to solar heating, the wind can also help evaporation of sea water.
Evaporation occurs only when there is sunlight. 18 The following is a recapitulation of the results of TTT analysis that can distinguish students who experience misconceptions, understand concepts, and do not understand concepts. Based on data in Table 4, the percentage of students on a small scale who experience misconceptions decreases. The decrease in the number of students experiencing misconceptions occurred in the whole concept of changing the form of objects. To find out the significance of differences in the quantity of misconceptions in the concept of changes in the form of experimental class objects and control classes, a statistical test of the hypothesis has been conducted. The following is a diagram that presents the percentage of the number of students experiencing an average decrease in the quantity of misconceptions per concept on the concept of tangible objects on a small and large scale.

Figure 1 The Percentage Decrease in Student Misconception Quantity
In Figure 1 shows that all concepts tested on a large and small scale about the concept of the form of objects experience a decrease in the quantity of students who misconceptions. The average percentage comparison of pretest, posttest, and decreasing student misconceptions between large and small scales is presented in Figure 2 below:

Figure 2 The Percentage of PKM Students large scale and small scale
The results of phase I tests on a large scale many students experienced misconceptions of 62.7% and the phase II test after instrument revision showed a significant decrease to 32.65% of students who experienced misconceptions, the presentation of a decrease in the quantity of misconceptions for large scale was 52.86%. the result was quite significant compared to the small scale which decreased by only 24.61%.

DISCUSSION
The degree of understanding of concepts and indicators in particular is as follows: (1)  The stage of data collection begins with making a type of question instrument from the editor of the sentence must be adjusted to the level of elementary age ability in reading and understanding the reading of the question. After that the question is distributed to first responders students on a small scale and then on a large scale. Furthermore, the stage of data processing with quantitative techniques calculates how in a learning environment, a tested, valid and reliabel instrument is required. In addition, the type of instrument given is not just a multiple choice question and answer or a 1-level fill, but it takes more than 2 levels to determine the validity of the answer and the respondent's confidence in the answer. Types of problems that have more than 1 level or more can identify misconceptions, and based on the findings, there are still many students who experience misconceptions in the material form objects.

ACKNOWLEDGMENTS
Researchers thanked the relevant parties including the Ministry of Education and Education who have provided research funds and schools that have given permission to carry out research. Not to forget also the whole team that has helped this research.