Microsoft Word - 4. Ijirana 13563-42543-2-ED 15 08 2021-JM (REVISI AGUST21).docx Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 194 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Ijirana1, Jusman Mansyur2, Muh. Rizal3, Sitti Aminah4 1-4Faculty of Teacher Training and Education, Universitas Tadulako, Indonesia DOI: 10.23917/ijolae.v3i3.13563 Received: February 16th, 2021. Revised: April 20th, 2021. Accepted: May 28th, 2021 Available Online: August 18th, 2021. Published Regularly: September 1st, 2021 Abstract This study aims to longitudinally describe the metacognitive skills and external representation in the context of problem-solving at Chemistry Education, Tadulako University. The qualitative study respondents were selec- ted from a number of first year students for two consecutive semesters. Two respondents were selected based on results of selection using a metacognitive skills assessment questionnaire (MCAI) and we categorized as high and medium. Three problems were resolved by respondents at intervals of one to two weeks in each semester through a one-on-one thinking-aloud and it was followed by a semi-structured interview. Data collection was recorded using a video camera. Metacognitive skills and external representation data from the odd and even semesters were deeply analyzed. This analysis technique was carried out by examining the results of problem solving and semi-structured interviews in detail and matching them with the indicators of metacognitive skills and external representation used by both respondents. The data analysis results showed that the thinking ability of respondents with high metacognitive skills in problem solving, experienced developments over time along with the development of their cognitive regulation and external representation. On the other hand, respon- dents with intermediate metacognitive skills during problem solving, were relatively the same for two consecu- tive semesters. Therefore, teachers need to consider the use of learning strategies by taking into account the number of students who have dominant characteristics in classroom learning, in terms of metacognitive skills and external representation. Keywords: context of problem-solving, external representation, longitudinal study, metacognitive skills, one- on-one thinking aloud, problem solving Corresponding Author: Ijirana, Faculty of Teacher Training and Education, Universitas Tadulako, Indonesia Email: ijiranarizal.untad@gmail.com 1. Introduction Problem-solving is an activity carried out by someone who is directed at achieving goals using a method or strategy and requires proper mental representation (Metallidou, 2009). A person’s problem-solving capabilities achie- vement can be measured from four compo- nents, namely; resources, heuristics, control, and belief (Yen & Lee, 2011). These capa- bilities are strongly influenced by the under- standing of materials and a person’s metacog- nitive skills to find solutions to problems (Iji- rana & Supriyadi, 2018) that will provide new knowledge, skills, and other components on the person (Fischer & Neubert, 2015). Howe- ver, the incomplete use of metacognitive skills (only using planning skills, not using monitoring and evaluation), will cause a po- tential for failure in problem-solving (Iji- rana & Supriadi, 2018). Therefore, there are two aspects that influence problem-solving, namely the use of complete metacognitive Indonesian Journal on Learning and Advanced Education http://journals.ums.ac.id/index.php/ijolae Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 195 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving skills and external representation. That is why it is necessary to investigate the use of these two thinking skills in the context of problem solving. Some findings from study on metacogni- tive skills in the context of problem-solving states that regulation of metacognition and monitoring of proper thinking can increase success in problem-solving, including analyt- ical and technical problems (Rickey & Stacy, 2000; Shubber, Udin, & Minghat, 2015). This metacognitive skill has become the basic ap- proach and strategy of students to achieve goals, organize, monitor, and modify opera- tions in problem-solving (Metallidou, 2009). Other results also discovered that awareness of cognition, planning, monitoring, and self- checking, as well as self-assessment (self-ap- praisal), is a metacognitive strategy often used by students (Haidar & Naqabi, 2008) as well as self-management (Demirel, Aşkın, & Yağcı, 2015). That is why metacognitive skills contribute to student academic success. The use of an external representation sys- tem regarding the use of diagrams in problem- solving showed that some teachers and stu- dents tended to separate diagram formulation steps from interpretation and variables identi- fication step so that they cannot attain suffi- cient results (Mansyur, 2015). Problem-solv- ing with different mental representation for- mats can be solved by students by compiling categories of mental representations (Ibrahim & Rebello, 2013) which can be stated exter- nally (Metallidou, 2009). External representa- tion functions to promote investigation, con- duct reflection of alternative perspectives, so- lutions, and criticism, as well as facilitating the development of knowledge (Van Brug- gen, Kirschner, & Jochems, 2002). Cox and Jones (2011) found that the external represen- tation and visual thinking has the potential to create ideas and make complex issues more accessible, make the organization of know- ledge and its synthesis easier when studying Basic Chemistry, Organic Chemistry and Bi- ochemistry. Furthermore, the reasoning of students was more varied in certain moments and that such variability shows a productive thought process as a representation in study- ing electric fields in physics (Cao & Brizuela, 2016). In line with this, Ningsih et al. (2013) found that physics teachers dominantly dis- played representation in verbal and algebraic formats, yet when presenting concepts, and they did not emphasize the importance of dia- grams. Even more generally found that repre- sentation in learning science not only im- proves problem-solving abilities, but can also increase retention of related knowledge and facilitate the integration of new knowledge with prior knowledge (Cook, 2006). There- fore, in this study, the emphasis was placed on the aspect of metacognitive skills and external representation of the chemistry education stu- dents. The study was carried out longitudi- nally in the context of problem-solving, be- cause this aspect is one of the indicators of the success of chemistry education students in problem solving and problem solving is needed in studying chemistry. The study will also provide an overview of increased knowledge retention and integration of new knowledge from previous knowledge pos- sessed by students at different times. 2. Method The study used a qualitative method fo- cusing on the investigation of metacognitive skills and external representations in the con- text of problem-solving of chemistry educa- tion students. We longitudinally investigated the development of the aspects in two consec- utive semesters. Description of metacognitive skills and external representation obtained from the students with different metacogni- tive skills was based on the results of the Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 196 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving screening through a questionnaire developed by Cooper and Sandi-Urena (2009). Data collection began with the screening of students registered in the chemistry educa- tion of Tadulako University, even semester of academic year 2018/2019, with different me- tacognitive skills. As many as 97 students were given the metacognitive skills asses- sment questionnaire (MCAI). Two students who were respondents in this stu-dy with a self-assessment category having high and in- termediate metacognitive skills. The two respondentes were coded R1KMT and R2KMS. These respondents were drawn from a number of students who rated themselves as carrying out metacogni- tive skills during problem-solving with a score of > 80% in the high category and between 60-79% in the intermediate category (adapted from Demirel, Aşkın & Yağcı, 2015). The respondents solved two problems in a one-on-one thinking-aloud setting. The se- cond problem was resolved one week to two weeks after the first problem was re- solved. Respondents who solved problems using the same method were then interviewed at different times. The same process was re- peated on the same respondents using a simi- lar problem. Problem-solving and interview activities were recorded using a video cam- era. Categorization and coding of the prob- lem-solving and interview results were done for two aspects. The first categorization and coding were for the use of metacognitive skills. This coding involves the respondent's activities, such as; planning, monitoring, and evaluating. The planning includes activities such as; reading, translating, setting goals, formulating problems, and making plans. The monitoring includes problem-solving activi- ties, namely; reviewing or rereading prob- lems, checking answers, and pausing while looking at the paper or the answer in front of the student in question. Evaluating includes the activity of following the correct way to discover the answer whether (right or wrong) and involves taking several types of steps to check the correctness of the answers (final or intermediate answers), and or the respondent decided immediately that the answer is wrong and started working again. This evaluation also includes reflection, namely whether the answers obtained makes sense or is as ex- pected. The second categorization and coding were for the used of the representation. The coding involved respondents' activities in making external representations during prob- lem-solving. The representations can be in the form of pictures, symbols, statements, formu- las, and mathematical language (Kohl & Finkelstein (2005). 3. Result and Discussion In the following we will present the results of one-on-one thinking aloud and interviews in the even and odd semester. a. R1KMT The results of the study of thinking- aloud R1KMT towards solving the problem in terms of metacognitive skills and external representation in different semesters are given in Table 1- Table 3. Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 197 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Table 1. Results of Thinking-aloud R1KMT during Odd and Even Semesters when Carrying Out Problem-solving Planning. Respondent Even semester Odd Semester R1KMT In the first minute the respondent read the problem by emphasizing certain words in the statement of the problem. In the first minute the respondent read the problem by empha- sizing certain words in the statement of the problem. (1) The data needed to solve this problem is the first atomic number of Sulfur 16 with Oxygen 8 atomic number of. (1) The first step to be taken is to collect all known data, na- mely 1) the compound molecular mass = 30.026 grams/mol; 2) this compound is made from burning methanol in a catalytic reactor; 3) using a silver catalyst; 4) the use of compounds as cosmetic preservatives, nail hardener, fly and insect repellent and materials for making urea fertilizer. (2) The first objective to be achie- ved is how the two atoms are bound and the process is formed and deter- mine the shape of the molecule. (2) Then step 2 looks for what was asked (the respon- dent reread the problem). Here what is asked is the name of the compound and its molecular shape. To find out the molecular shape of a compound, look for its constituent elements, which is where it is (reread problem) When the data in Table 1 is viewed from the metacognitive skills perspective and ex- ternal representations, it can be described that: (1) It seems that R1KMT at different times and a relatively long time has always been consistent in solving the problem, namely used metacognitive skills in making plans. A different matter occurred when the respon- dent translated the problem further statement thoroughly in the odd semester compared to the previous semester, because she wrote in detail all the data in the problem state- ment that supports the achievement of the goal. In addition, the respondent at this stage also began writing down her problem-solving strategies step by step; (2) In both periods she also remained Consistent in stating the objec- tives to be achieved in problem-solving but, in the odd semester of this respondent had be- gun to make plans for the achievement of the objectives and even had carried out monito- ring by re-reading the problem state- ment. This shows an increase of the respon- dent’s prudence in making a decision from time to time; and (3) The respondent did not perform external representation both in the even or odd semester during the planning stage. Based on these findings, it can be said that the habit of doing good planning will in- crease one's awareness of thinking to solve problems according to targeted goals so that in solving problem one will be more careful in planning. This finding is in line with other research findings, namely that someone who chooses the right strategy and uses known in- formation from the problem in doing the task will lead to the right performance for someone (Demirel, Aşkınb & Yağcı, 2015; Silwana, Subanji, Manyunu & Rashahan, 2021). That good planning and monitoring is very impor- tant to help students plan how to settle pro- blems, manage the problem-solving process and ensure that it is on the right track (Berci- ter, 2002). This finding also reinforces previ- ous findings that the planning strategies iden- tified in students in solving problems are planning, setting goals, and priorities (Yang & Bai, 2019). Data description in the problem-solv- ing and external representation in Table 2 are: (1) in the even semester the respondents began arranging problem-solving strategy Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 198 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving while in the odd semester the respondent con- tinued her problem-solving strategies because it had been started in the previous stage of thinking. An interesting thing here is that the respondent conducted assimilation thinking activities several times by stating that "com- bustion reaction means involving oxygen". This showed that respondent matched the knowledge of the information received with the knowledge she had. The next assimilation activity is to write down the formula for the methanol molecule (CH3OH) after pausing for a moment to recall the knowledge she had about methanol. Representation of methanol into the form of a chemical symbol was done by the respondent to make it easier to write reaction equation and predict the products pro- duced. Another thing that was shown by the respondent was the ability to represent the re- sults of her thoughts in the form of a reaction equation as described in Figure 1. Figure 1. R1KMT’s Answer Sheet For Reaction Equation The next assimilation process at this stage is the assertion that the silver catalyst used only accelerates the reaction and does not affect the outcome of the reaction. Table 2. Results of Thinking-Aloud by R1KMT in Even Semester and Odd Semester when Formulating a Strategy and Monitoring Respondent Even semester Odd semester R1KMT (1) The strategy used in solving this problem is that we first look for the valence electrons of each atom (the respondent writes) (1) The next step is to make a reaction that occurs where a combustion reaction occurs. The combustion reaction involving oxygen and in this matter the methanol or hydrocarbons combustion in a catalytic reactor. So the methanol (the respondent thought for a while then wrote) CH3OH; and its reaction. Oh... here there is the use of catalysts and here are the by products of water. Thus the amount of C and H on the left and right of the number is the same, namely the number C is 2 and H = 8 and O = 4. The silver catalyst here serves to ac- celerate the reaction rate and is not found in the reaction result. (2) To determine electron va- lence, the first thing to do is to make an electron configura- tion, the second, make the Le- wis structure. The sulfur has an atomic number 16 so its elec- trons are on the skin; KLM has an electron number of 2 8 6 each. Next O8 = 2 6 (silent respondent) So S16 is valence electron 6 and O is va- lence electron 6. So the Lewis sructure (2) After that, the next step, fourth (the respondent is silent and writes) After obtaining the reaction re- sults, the molecular shape can be determined. Let’s say this is compound data, the compound obtained was CH2O so 6 was the atom number of C, H was 1, and O was 8. So the configuration of the electron to C is 2 and 4 and O = 2 and 6. So it can be concluded where C is the central atom with H and O, which surrounds the central atom. So the Lewis structure: Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 199 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Respondent Even semester Odd semester (3) Determine PEB and PEI to determine molecular shape. So this is O to be stable it binds electrons from S and this S also binds electrons from O to stabi- lize it so this is the shape So this is (hum) the PEB, the pair of free electrons is zero while the bound pair of electrons is 3. So the molecular shape is AX3 (Trigonal Planar). (Respondent rereads written answers) So it can be seen that the compound intended for the problem is formaldehyde or commonly called formalin in the form of trigonal molecular planar. So according to the VSPER theory the molecular form above is Trigonal Planar with 2 pairs of free elec- trons on the atom that surrounds the central atom and electron pair with 4 bondings. Next, in the next stage the respondent used the concept of equalizing the reaction equation by counting the number of elements involved in the reaction as in Figure 2. Figure 2 R1KMT’s Answer Sheet for the Concept of Equalizing A Reaction In the interview session, the respondent ensured that this reaction product was cor- rect and one way to do that was to test the type and number of elements involved in the reac- tion. The activity shows that the respondent conducted regulation of cognition through monitoring while evaluating problem-solving to ensure the correctness of the results. (2) It seems that the process of thinking and repre- sentation carried out by the respondent are the same for even or odd semester. The respon- dents rearranged the plan with the same steps to write the Lewis molecular structure in which form was to be determined. The ac- tivity was started by making electron configu- rations in the form of images. A different manner was carried out by the respondent in the odd semester when she was going to make a molecular structure. There was a lack of confidence of the respondent when she was to describe the molecular structure of formal- dehyde where the respondent paused for a moment then crossed it out and made a new molecular structure. This shows that the res- pondent continuously monitored her problem- solving process; and (3) Different representa- tions were made in the even semester, namely from the representation of molecular shapes to the next picture form, then to a mathemati- cal equation form, and finally in the form of mathematical statements (Figure 3). Figure 3. R1KMT’s External Representation Se- quence in the Even Semester In contrast to the odd semester, represen- tation of the shape of the molecule was ex- pressed as a mathematical statement from a picture form as shown in Figure 4. Figure 4. R1KMT’s External Representation Se- quence in the Even Semester The respondent’s confidence toward this result shows in the interview result that if the geometric form was like this, then it is a Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 200 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving trigonal planar. This indicates that there was an increase in the respondent's ability to pre- dict molecular shapes based on the number of bonding electrons and free electron pairs of the compound in the Lewis structure. Thus, student with high metacognitive skills will in- creasingly develop their knowledge along with the development of cognitive regulation. In contrast to the findings of Yanti, Amin, and Sulaiman (2018) who found that students who have multiple intelligences, in presenting information, preparing plans, and implementing plans in problem-solving, use the same representation whereas when evaluating results, using various forms of re- presentation. Table 3. Results of Thinking-Aloud by R1KMT in Even Semester and Odd Semester when Evaluating Problem-solving Result Respondent Code Even semester Odd semester R1KMT So the molecule formed is AX3 be- cause the PEB of the central atom is missing. (read the questions then write) This answer can be strengthened by the information that is known in the problem, which is the use of the compound in question, namely as a preservative of cosmetic products, nail hardener, fly and insect re- pellent, as well as making urea fertilizer. Where all of the above statements refer to formaldehyde com- pounds and where the formaldehyde molecular mass is also known to be 30.026 grams/ mol. The data in Table 3 shows that the last step undertaken by respondent in the two pe- riods was evaluating results. Evaluation skills performed by the respondent in the odd se- mester shows a more developed thought pro- cess compared to the even semester. In the odd semester respondent evaluated results using logical thinking skills, namely connec- ting formaldehyde molecules formula with re- lative molecular mass and its proper- ties. This indicates that a person who is accus- tomed to organizing his/ her thinking to solve problems will increase his/ her awareness to convince himself/ herself in making decisi- ons, so that mistakes in solving problems will increasingly diminish. This is in line with the findings of Rahman et. al. (2010a) which stated that evaluation skills will allow stu- dents to reduce the mistakes made in the pro- cess of problem-solving. The same applies with the findings Veenam and Spaans (2005) which stated that metacognitive skills occur together, but it is not entirely dependent on the intellectual ability because metacognitive skills exceed intelligence as a predictor of le- arning performance. b. R2KMS The results of the study of thinking- aloud and interview in R2KMS toward sol- ving a given problem in terms of metacogni- tive skills perspective and external represen- tation in different semesters are given in Tab- le 4 - Table 6. Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 201 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Table 4. Results of Thinking-Aloud by R2KMS in Even and Odd Semester When Conducting Planning, and Monitoring Respondent Code Even semester Odd semester R2KMS Respondent read the problem slowly Respondents read the problem slowly (1) In solving the first problem, na- mely knowing the atomic number of iron is 26 and the atomic number of oxygen is 8. (1) The first compound formed is obtained by bur- ning methanol, so what we need to know is the mo- lecular formula of methanol, which is CH3OH. After that we make the combustion reaction of methanol using a silver catalyst or argentum. Now the reaction is like this, then equated before combustion and after combustion (2) The purpose of solving this problem determines the process of forming compounds from iron that bind with oxygen and determining the compound formula In this problem it is stated that in addition to the com- pounds produced, water was also produced (reread the problem) Well the combustion reaction of methanol, to me, produces carbon dioxide gas and water From the perspective of metacognitive skills, and external representations, the data in Table 4 can be described as follows: (1) it seems that R2KMS at different times and re- latively long periods of time is less consistent in solving problems. Respondent in the even semester did the planning by translating the problem and stating the objectives to be achi- eved systematically, while in the odd semes- ter the respondent translated the problem without stating the goal, yet she immediately formulated a strategy to solve the pro- blem. Respondent revealed the solution to his/her problem directly by stating that the compound obtained from burning methanol is carbon dioxide gas after writing down the reaction equation; (2) It appears that respon- dent in this odd semester carried out assimila- tion thinking activities by linking new infor- mation received with the knowledge they al- ready have. This is done by translating the word ‘combustion’ by presenting chemical re- action with the addition of oxygen. In addi- tion, decision making about the compounds produced is done through equalizing the reac- tion equation; (3) Other activities of respon- dent at this stage was monitoring when per- forming reaction equalization by repeatedly make an improvement through crossing out mistakes which is shown in Figure 5; and (4) Decision making by the respondent without an evaluation caused the respondent to has a chance of failing in solving problems. It is identified by the methanol combustion re- action results obtained, namely carbon dio- xide. Figure 5. R2KMS’s Answer Sheet in Monitoring on Reaction Equation Equalization in Odd Semester The result of the interview showed the respondent stated that all combustion reac- tions must be the result of carbon dioxide. This certainly cannot be denied when metha- nol is completely combusted allowing carbon dioxide to be produced. But in this problem, the compound in question has a relative mo- lecular mass of 30.026 g/mol and functions as a preservative, a fly repellent, and as a mate- rial for making urea fertilizer. If the respon- dent reread the problem statement and re-exa- mined the results, this would not have happe- ned because carbon dioxide does not have Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 202 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving these characteristics (a relative molecular mass of 44 g/ mol). Therefore, evaluation skills are needed in problem-solving. The data in Table 5 shows that; (1) Pro- blem-solving style that is done by respondents both in the even semester and odd semester both go through strategy development. In both of these time periods, the respondent showed a consistent problem-solving manner by showing the same representation, namely the picture representation into a statement, al- ternating between them; and (2) What was in- teresting about how students carried out pro- blem-solving in the Even Semester was that they carried out monitoring against the pro- blem-solving process that was still inconclu- sive, as shown in Figure 6. It was different in the odd semester, the respondent was very sure of the answer by gi- ving scientific explanations in each part of the solution. This finding is in line with the fin- dings of Fitrianna, Dinia, Mayasari, dan Nu- rhafifah (2018) that students with high mathe- matical abilities have been able to present data/information from one representation into diagrams, graphs, or tables and solve pro- blems using words or written text. The data in Table 6 shows that the re- sults evaluation technique carried out by R2KMS did not change from time to time. Respondent did not try to assess the problem statement when making a decision and did not re-monitor the problem-solving process, so they (she) did not realize that the molecular mass of CO2 was not 30.026 grams/mol. This was evident when the interview was conducted because the respondent did not try to think that the truth of the results could be tested with the relative molecular mass of car- bon dioxide or its properties. Therefore, in addition to getting used to thinking how to think about the task being carried out, insight or mastery of concepts related to the problem being solved was also required. These results indicate the need to follow up on the findings of Rahman, et. al. (2010b) stated that teachers need to be awere of the importance of promo- ting metacognitive development in the class- room. Ijirana (2017) stated that metacognitive skills are an important part of learning for che- mistry education students, because according to Setiawan, Arisanty, Hastuti, and Rahman, (2020) greatly contribute to improving lear- ning outcomes. Masari and Anghel (2012) found that it plays an important role in lear- ning success and is closely related to authentic assessment and holistic learning. Therefore, attention of teachers is requi- red in designing learning strategies by paying attention to students in the intermediate meta- cognitive skills category, because this cate- gory always dominates in the classroom (De- mirel, Aşkın, & Yağcı, 2015; Altındağ, 2008; Temel, Özgür, Şen & Yılmaz, 2012; Kiremitçi, 2013; Jaafar & Ayub, 2010; Duran, 2011). Figure 6. R2KMS’s Answer Sheet in Monitoring on Reaction Equation Equalization in Even Semester Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 203 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Table 5. Results of Thinking-aloud in R2KMS of Even Semester and Odd Semesterwhen doing Planning, Monitoring, and Troubleshooting Respondent Code Even semester Odd semester R2KMS (1) The first step to do is deter- mine the valence electrons of the two elements by using elec- tron configurations. Electron configurations of iron and oxygen. Iron or Fe with atomic number 26 and oxygen with atomic number 8 (Looking back at the answer written) Oxygen with electron shell configuration 2 6 requires 2 electrons to achieve stability. If these 2 elements bind, then Fe will release 3 electrons so that Fe becomes Fe +3, while O cap- tures 2 electrons to O -2. The load contained in each element is not the same number so it ne- eds to be equalized. (1) To determine the molecular shape of CO2 we first look for the valence electrons of each binding element and determine the atom used as its central atom. Valence electrons can be obtained from the electron configurations of each element: Carbon, the valence electron 4 while oxygen has a valence electron 6, which means that C is able to bind to 4 other atoms and valence electron from oxygen is 6 so that oxygen requires 2 electrons to achieve stability. So the results obtained are Fe2O3. Now for CO2 gas that is used as a central atom na- mely, C, or carbon with 4 valence electrons binding to O which has 6 valence electrons. The bonds between the CO2 has 2 pairs of bonding electrons or 2 domains. Before we determine the molecular shape of a com- pound we need to pay attention to its free electron pairs, bonding electron pairs and its electron pairs, so we can determine the molecular shape of a com- pound. Now here 2 PEI (with body movements), by ma- king Lewis structure from the CO2 we can Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 204 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Respondent Code Even semester Odd semester determine the shape of the molecule. In my opinion the molecular shape of the CO2 is linear. because the central atom does not have all the free electron pairs, all electron pairs have been used to bind to oxygen so that the resulting molecular form is li- near. Table 6. Results of Thinking-aloud R2KMS of Even Semester and Odd Semester when Evaluating Results. Respondent Code Even semester Odd semester R2KMS Yes, I'm sure my answer is correct be- cause oxygen requires (writes) 2 elec- trons and Fe releases 3 electrons so that Fe2O3 can form compounds that have stable valence electrons or the result of these two elements is equal to zero. The shape of the molecule produced is li- near because there is no repulsion between the free electron pair and the bonding elec- tron pair so, I am sure, sure that the molecu- lar shape of CO2 is linear. 4. Conclusion Chemistry education students who are considered to have high metacognitive skills when solving a problem carry out three acti- vities of thinking, namely planning, monito- ring, and evaluation, and also consistently created external representation in both sepa- rate semesters. Thinking activities in pro- blem-solving planning in the odd semester were more meticulous than the previous se- mester (even). In fact, even in this thinking activity, the students had been monitoring the planning process to reduce their mistakes. In addition, the knowledge possessed by stu- dents in this category had been growing over time as demonstrated by their ability to relate new information to the knowledge they have, to link between concepts, and supported by their ability to represent that knowledge exter- nally. The same applies to their ability to eva- luate results, the student was used to regulating their own thinking in solving pro- blems so as to increase her awareness to con- vince herself in decision making. Chemistry Education student who were categorized as having intermediate metacog- nitive skills when problem-solving only con- ducted thinking activities in planning and mo- nitoring consistently, but not in evaluating ac- tivities. His/her knowledge and ability to construct external representations also deve- loped, but did not sufficiently to guarantee her success in problem-solving. That is why student in this category has a great chance of failing in solving a problem. 5. References Altındağ, M. (2008). Hacettepe üniversitesi eğitim fakültesi öğrencilerinin yürütücü biliş becerileri. Yayınlanmamış yüksek lisans tezi, Hacettepe Üniversitesi, Sosyal Bilimler Enstitüsü, Ankara. Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 205 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Bereiter, C. (2002). Education and mind in the knowledge age. Mahwah, NJ: Lawrence Erlbaum Associates. Cao, Y. & Brizuela B.M. (2016). High school students’ representations and understandings of electric fields. Physical Review Physics Education Research, 12 (2), 020102(19). Cook, M.P. (2006) Visual Representations in Science Education: The Influence of Prior Knowledge and Cognitive Load Theory on Instructional Design Principles. Published online 20 June 2006 in Wiley InterScience (www.interscience.wiley.com), DOI 10.1002/sce.20164. Cooper, M. M. & Sandi-Urena, S. (2009). De- sign and Validation of Instrument To As- sess Metacognitive Skillfulness in Che- mistry Problem Solving. Journal of Che- mical Education, 86(2), 240-245. Cox, J.R. & Jones B. W. (2011). External Re- presentations in the Teaching and Lear- ning of Introductory Chemistry . Scientific Research. Creative Education. 2(5) pp. 461-465. DOI:10.4236/ce.2011.25067. Demirel, M., Aşkın, İ., & Yağcı, E. (2015). An investigation of teacher candidates’ metacognitive skills. Procedia-Social and Behavioral Sciences, 174, 1521- 1528). Duran, S. (2011). İlköğretim öğretmenlerinin bilişsel farkındalık düzeylerinin çeşitli değişkenlere göre incelenmesi. Yayınlan- mamış yüksek lisans tezi. Kafkas Üniver- sitesi, Sosyal Bilimler Enstitüsü, Kars. Fischer, A., & Neubert, J. C. (2015). The mul- tiple faces of complex problems: A mo- del of problem solving competency and its implications for training and asses- sment. Journal of Dynamic Decision Ma- king, 1(6). Fitrianna, A. Y., Dinia, S., Mayasari, M., & Nurhafifah, A.Y. (2018). Mathematical Representation Ability of Senior High School Students: An Evaluation from Students' Mathematical Disposi- tion. Journal of Research and Advances in Mathematics Education, 3(1), 46-56. Haidar, A. H. & Naqabi, A. K. (2008). Emira- tii High School Student Understanding of Stoichiometry and the Influence of Meta- cognition on their Understanding. Rese- arch and Science and Technological Education, 26(2), 215-237. Ibrahim, B. & Rebello, N.S. (2013). Role of Mental Representations in Problem Solving: Students’ Approaches to Nondirected Tasks. Physical Review Special Topics - Physics Education Research, 9, 020106 (2013). Ijirana & Supriadi. (2018). Metacognitive Skill Profiles Of Chemistry Education Students In Solving Problem At Low Ability Level. Jurnal Pendidikan IPA In- donesia, 7(2) pp. 239-245. DOI:10.15294/jpii.v7i2.14266. Ijirana. (2017). Development of concept un- derstanding and problem-solving ability of chemistry education students through metacognitive skills-based learning. Ad- vances in Social Science, Education and Humanities Research , 174, 75-79. Jaafar, W. M. W. & Ayub, A. F. M. (2010). Mathematics self-efficacy and meta-cog- nition among university students. Proce- dia Social and Behavioral Sciences, 8, 519-524. Kiremitçi, O. (2013). Beden eğitimi ve spor yüksekokulu öğrencilerinin üstbilişsel farkındalık düzeyleri üzerine bir ince- leme. Pamukkale Journal. of Sport Scien- ces, 4(3),29-40. Kohl, P. B., & Finkelstein, N. D. (2005). Stu- dent Representational Competence and Self-Assessment When Solving Physics Problems. Physical Review Special Top- ics-Physics Education Research, 1, 010104. Mansyur, J. (2015). Teachers’ and Students’ Preliminary Stages in Physics Problem Solving. International Education Studies, 8(6). Masari, G. A., & Anghel, O. (2012). Compa- rative study on developing metacognitive abilities of students from technical, voca- tional and human sciences. Procedia- Indonesian Journal on Learning and Advanced Education (IJOLAE)| p-ISSN 2655-920x, e-ISSN 2656-2804 Vol. 3 (3) (2021) 194-206 206 Longitudinal Study of Metacognitive Skills and External Representation of Students in the Context of Problem-Solving Social and Behavioral Sciences, 46, 4418-4422. Metallidou, P. (2009). Pre-service and In-ser- vice Teachers’ Metacognitive Knowledge About Problem-Solving Strategies. Tea- ching and Teacher Education, 25, 76-82. Ningsih, H. Y. S. T, Mansyur, J. , Kamaluddin & Darsikin. 2013). Perilaku Penggunaan Representasi Eksternal Guru Fisika dalam Aktivitas Pembelajaran. Prosiding Seminar Nasional Pendidikan, Palu 04 Mei 2013. Universitas Tadulako. Rahman, S., Mahmud, Z., Yassin, S. F. M., Amir, R., & Ilias, K. W. (2010b). The De- velopment of Expert Learners in the Classroom. Contemporary Issues in Edu- cation Research, 3(6), 1-8. Rahman, S., Yasin, R. M., Ariffin, S. R., Ha- yati, N., & Yusoff, S. (2010a). Metacog- nitive skills and the development of me- tacognition in the classroom. In Interna- tional Conference on Education and Educational Technologies, 47-351). Rickey, D. & Stacy A. M.. (2000). The Role of Metacognition in Learning Chemistry. Journal of Chemical Education. 77(7). Setiawan, F. A., Arisanty, D., Hastuti, K. P., & Rahman, A. M. (2020). The Effect of Metacognitive Ability on Learning Out- comes of Geography Education Stu- dents. Indonesian Journal on Learning and Advanced Education (IJO- LAE), 2(2), 82-90. Silwana, A., Subanji, S., Manyunu, M., & Rashahan, A. A. (2021). Students’ Res- ponses Leveling in Solving Mathematical Problem Based on SOLO Taxonomy Viewed from Multiple Intelligences. In- donesian Journal on Learning and Ad- vanced Education (IJOLAE), 3(1), 1-16. Shubber, A. H., Udin, A. B., & Minghat, A. B. (2015). Vickers test simulation to im- prove metacognitive skills. Procedia-So- cial and Behavioral Sciences, 204, 45- 53. Temel, Ş., Özgür, S. D., Şen, Ş. & Yılmaz, A. (2012). The examination of metacogni- tive skill levels and usage of learning strategies of preservice chemistry teachers. Procedia Social and Behavioral Sciences, 46, 1445-1449. Van Bruggen, J. M., Kirschner, P. A., & Jo- chems, W. (2002). External representa- tion of argumentation in CSCL and the management of cognitive load. Learning and instruction, 12(1), 121-138. Veenman, M. V. J, & Spaans, M. A. (2005). Relation between intellectual and meta- cognitive skills: Age and task differen- ces. Learning and individual differen- ces, 15(2), 159-176. Yang, C., & Bai, L. (2019). The use of meta- cognitive strategies by Chinese PhD stu- dents of social sciences in Australian uni- versities. International Journal of Edu- cational Research, 97, 43-52. Yanti, Y. R., Amin, S. M., & Sulaiman, R. (2018). Representation of Students in Solving Simultaneous Linear Equation Problems Based on Multiple Intelligence. In Journal of Physics: Conference Series, 947 (1), p. 012038). IOP Publishing. Yen, J. C., & Lee, C. Y. (2011). Exploring problem solving patterns and their impact on learning achievement in a blended le- arning environment. Computers & Edu- cation, 56(1), 138-145.