Analysis of student’s creative thinking ability in the computation physics course

Copyright


INTRODUCTION
Education plays a very critical role and factor in human life because it is far a vehicle for creating quality human resources in terms of expertise and abilities in order to have the ability to suppose critically, creatively, and openly.Thinking ability is one of the essential aspects to be developed.Thinking is a person's mental manner that is more than just remembering and understanding (Putra, 2012).Thinking causes a person to move to expand his mind beyond the information he hears.Thinking, for example, is someone's ability to discover new solutions to a problem.One of the higher-order thinking abilities to deal with problems in science and real life is the ability to think creatively.Creativity is a learning goal rarely considered (Agustinaningsih, 2020).Students show the importance of creativity in learning with high creativity, high problemsolving abilities, and better learning achievements (Andari, 2015).The ability to suppose creatively is one of the competencies that physics students must possess.
Creative thinking is the ability to offer a spread of possible answers primarily based on the information furnished, emphasizing the variety of numbers and suitability (Nur, 2014).Creative thinking is an aggregate of thinking of logic and divergent thinking primarily based on instinct but in awareness (Wechsler, 2018).Creative thinking is often also referred to as divergent thinking.Divergent thinking has three leading indicators: fluency, flexibility, and elaboration (Guildford, 1968).Creative thinking includes thinking fluently, flexibly, original, and elaboratively (Almeida, 2008).Expertise in creative thinking as a concept production manner that emphasizes factors of fluency, flexibility, novelty, and elaboration (Hoiriyah, 2019).There is also an indicator of creative thinking, particularly sensitivity to problems, that is, the ability to understand or ignore something much less relevant so that the real problem can be identified (Darwanto, 2019).Sensitivity is the ability to discover, understand, apprehend, and reply to an assertion, state of affairs, and problem (Nurlaela & Ismiyati, 2015).Stimulus is needed to encourage awareness of a problem in response to a situation.Creative thinking, indeed, cannot be separated from the word creativity.Creativity is usually described as the potential to create a new product.Creative thinking is the capability to easily and flexibly discover solutions to a problem quickly and flexibly (Simanjuntak, 2019).The student's ability to think creatively allows the student to obtain many ways or alternatives to a problem, although sometimes too many ways will make it difficult for students to arrive at the final result, with many choices of ways will allow students to achieve their goals compared to students who do not have a solution to the problem.Solve the problem.Therefore, creative thinking is one of the essential things for students.
Computational Physics is essential in the physics study program at Universitas Negeri Padang.Computational physics disciplines that combine physics, numerical analysis, and computer programming have made it easier to process large and non-linear experimental data.Simulation experiments and non-linear and non-symmetrical mathematical models in computational physics can be completed with the help of numerical methods in computer programs (Said, 2015).Students in Computational Physics learning should be able to formulate the basics of numerical analysis techniques to solve physics problems algorithmically (Akmam, 2018).So that the understanding of physics, in theory, experiment, and computation, must be comparable so that numerical solutions and visualization or modeling are obtained that are appropriate for understanding physics problems.However, a common problem faced by students in solving computationally is the lack of ability to describe complex problems and analyze the general differences and similarities of a problem so that in the end, students cannot generalize problem-solving patterns (Angraini, 2019).In this case, it is related to the ability to think computationally so that scholars can think gradually and sequentially, not just be capable of programming computers.Computational thinking involves students understanding large-scale modeling systems using appropriate levels of abstraction and popularity, using mechanical inference descriptions to analyze complex data and providing supplementary computations for physics experiments on science problems.So there needs to be an encouragement in college students that affect the progress of critical and creative thinking abilities, namely college students' traits in showing critical and superior understanding; creativity in solving problems.Thinking Computational should be supported by critical thinking and thinking of creative (Akmam, 2019).The creative factor can assist explain and interpreting summary ideas, consequently allowing college students to achieve more mastery.Therefore, creative thinking is one of the essential things for students to have in studying Computational Physics.
Gravity: Jurnal Ilmiah Penelitian dan Pembelajaran Fisika, 8(2), 2022, 80 Copyright © 2022, Gravity, ISSN  In essence, each student's creative thinking level is different, with high, medium, and low creative thinking abilities (Agustinaningsih, 2020).This is influenced by what factors affect students' ability to think creatively.Students with high creative thinking skills tend to be more active and sensitive, try to express opinions or ideas, ask and answer various questions.Students with moderate and low creative thinking skills do not fully answer and ask questions and are afraid to issue ideas.It is essential to realize and analyze the ability of creative thinking from college students within the Physics of Computational course to support student development, so creative thinking becomes an interesting issue among researchers.Given the significance of creative thinking abilities in supporting student development, especially in Physics learning, it is necessary to describe students' creative thinking abilities in the Computational Physics course so that knowing students' creative thinking abilities can provide alternatives in determining the correct learning method.Based on these issues, the title of this research is "Analysis of college students' creative thinking ability within the Computational Physics course."

RESEARCH METHODS
This type of research is descriptive research with quantitative methods.Research of descriptive aims to obtain quantitative data by describing a phenomenon in fact and as it is and has nothing to do with other variables (Sanjaya, 2013).The analysis in this study was carried out at the level of description, namely analyzing and presenting facts systematically.Therefore, this research was conducted to describe the creative thinking abilities of Universitas Negeri Padang Physics college students in the Computational Physics course.The populace on this examination were college students who took the Computational Physics course in January-June 2021.The sample in this examination amounted to 50 student respondents who took the Computational Physics course in January-June 2021.The instrument used inside the take a look at becomes inside the form of statements.A questionnaire is used to see the creative thinking ability of college students in the Physics of Computational course.In the questionnaire grid, which refers to the creative thinking aspect of college students, the use of a scale of Likert, the variables to be measured are translated into variable signs, and then those signs are used as benchmarks for compiling objects that may be in the form of statements.The questionnaire uses a Likert scale containing positive and negative statements adjusted to indicators of students' creative thinking abilities.For each item, four answer choices are provided with alternative answers, namely: Always, Often, Rarely, and Never.Questionnaires in this study were also given score levels for each alternative answer.The questionnaire instrument needs to be examined for validity to check the instrument's feasibility to be used in the study (Sugiyono, 2012).Content material validity analysis of the Aiken V coefficient was used to check the validity of the observation sheet device.Content validity using Aiken's V coefficient (Azwar, 2012:113;Aiken, 1985:133) is obtained by applying a formula.
Where r is the quantity given through the evaluator, lo is the bottom validity score range, c is the highest validity rating number, n is the number of experts & practitioners who carry out the Gravity: Jurnal Ilmiah Penelitian dan Pembelajaran Fisika, 8(2), 2022, 81 Copyright © 2022, Gravity, ISSN 2528-1976 assessment, and i is an integer from 1,2,3 to n.The validity coefficient to find out that the instrument is in the valid category is based on the criteria in Table 1 (Retnawati, 2016) (Hari, 2008).

RESULTS AND DISCUSSION
Inferential data analysis in this study used the SEM approach with the help of LISREL 8.8 software to measure how much influence each indicator had on the analyzed students' creative thinking skills.SEM is one of the multivariate statistical analyses (Rex, 2011).The SEM method contains two types of variables, namely the observed and latent variables, so research variables are arranged based on this concept.The variable of creative thinking ability (KBK) is the latent variable, and the observed variables (manifest variables) are the indicators of creative thinking ability, namely Problem Sensitivity (PS), Fluency (FU), Flexibility (FE), Originality (OR ), Elaboration (EL), and Evaluation (EV).The validity of the items that compose a construct can be seen from the t-value.Based on the evaluation results, all observed instrument indicators meet the requirements of the t-value ≥ 1.96.Then, to form a fit model, the indicator items are removed, then a re-estimation is carried out to see the new structural model.The path diagram of the Loading Factor re-estimation results is provided in Figure 1.
After getting the re-estimation effects, it is possible to assess the Goodness of fit criteria index.However, because the chi-square, p-value, and RMSEA values do not meet the criteria for the measure of fit or Goodness Of Fit (GOF), meaning that the model does not fit, it is necessary to do a model respecification (modification).Re-specification of the model is done if a proposed model is not a fit model (Rex, 2011).Re-specification of the model is carried out to increase the version's universal fit by decreasing the chi-square value.Re-specification of the model is done by The information contained in the modification index can be seen in Table 2.After making modifications based on this information, a path diagram is obtained, as provided in Figure 2.After modification, the chi-square value becomes more minor to 189.10 with a p-value that is more than desired criteria (0.0517>0.05), in order that the relatively fit of the model primarily Gravity: Jurnal Ilmiah Penelitian dan Pembelajaran Fisika, 8(2), 2022, 83 Copyright © 2022, Gravity, ISSN 2528-1976 based on the chi-square value has a good fit.The value of RMSEA becomes smaller by 0.062 (close fit<0.05≥goodfit≤0.08>badfit) with a p-value is 0.29>0.05,and the value of RMSEA is inside the 90% reliance interval for RMSEA (0.0; 0.094) in order that the RMSEA is in a reasonable degree of precision.Based totally on those outcomes, the RMSEA fit measure has a good fit.The outcomes of the close fit of the model may be visible in Table 3.Based on the SEM concept that a model can be corrected for its errors by utilizing a modification index (Rex, 2011).After the modification, the fit of the structural model was improved.The magnitude of the direct influence between latent and observed variables may be seen in the output or on the path diagram of the structural model.The coefficient of determination (R 2 ) measures how some distance the model's capability to explain variations inside the dependent variable is.A value near one way that the impartial variables offer the data had to expect the variation of the dependent variable.The direct effect analysis of the structural model of this research model is presented in Table 4.In this study, the variable that has the most significant influence on the KBK variable (Creative Thinking Ability) is the FE (flexibility) variable, which is 0.96, and the variable that has the most negligible impact on the KBK variable is the EV (evaluation) variable, which is 0.58.The coefficient of determination (R2) for the FE variable is 0.87, meaning that 87% of the FE variable affects the KBK variable.Moreover, the coefficient of determination (R2) for the EV variable is 0.25, meaning that 25% of the EV variable affects the KBK variable.The results of students' creative thinking abilities within the Computational Physics course are presented in Table 5.Based on Table 5 it saw the statistical distribution of creative thinking abilities as a whole.Based on the table, it may be described that the number of samples in the data is 50 respondents, where the maximum value is 133 and the minimum is 74.The median value is 97.50, the mode value is 104, the variance is 191.643, and the standard deviation is 13.844.Based on the average value obtained in the total data of 97.50, the percentage of student achievement levels on the creative thinking ability questionnaire within the Computational Physics course is 65%.

CONCLUSION
It is based totally on the results of the research and discussion which have been defined.It could be concluded that the creative thinking ability of college students within the Computational Physics course as an entire is 65%, the variable that has the most significant influence on the KBK variable (Creative Thinking Ability) is the FE (flexibility) variable, which is equal to 0.96 and the variable that has the slightest effect on the KBK variable is the EV (evaluation) variable, Gravity: Jurnal Ilmiah Penelitian dan Pembelajaran Fisika, 8(2), 2022, 85 Copyright © 2022, Gravity, ISSN 2528-1976 which is 0.58.The coefficient of determination (R2) for the FE variable is 0.87, meaning that 87% of the FE variable affects the KBK variable.Moreover, the coefficient of determination (R2) for the EV variable is 0.25, meaning that 25% of the EV variable affects the KBK variable.

Figure 1 .
Figure 1.Path diagram of loading factor re-estimation results

Table 1 .
Valid criteria ) Evaluation.The theoretical framework model that has been built is then transformed into a path diagram to describe the causal courting among latent variables and determining variables.The model fit test is concerned with testing the fit between the model and data.Several criteria for a measure of fit or Goodness of Fit (GOF) can be used to carry out this step.The assessment of the level of statistical conformity with the model is carried out through several stages, specifically: 1) complete model suitability test, 2) measurement model suitability and 3) structural model suitability

Table 2 .
Modification indices recommend featuring an errors covariance

Table 3 .
The goodness of fit measures (GOF) after specific GOF Target match rate Estimated results Match rate

Table 4 .
Direct effect structural model and coefficient of determination (R 2 ) There was a significant change in the size of the GOFI in the structural model.The estimation results obtained by the PS variable is 0.91, the FU variable is 0.94, the FE variable is 0.96, the OR variable is 0.64, the EL variable is 0.69, and the EV variable is 0.58.The coefficient of determination R2 shows how much influence the variables PS, FU, FE, OR, EL, and EV have on the KBK variable.The estimation results obtained by the PS variable are 0.85, which means Copyright © 2022, Gravity, ISSN 2528-1976 that the PS variable impacts the KBK latent variable by 85%, the FU variable is which means the FU variable impacts the KBK latent variable by 86%, the FE variable is which means the FU variable impacts the KBK variable of latent by 87 %, the OR variable is 0.46 which means the OR variable impacts the latent variable KBK by 46%, EL variable is 0.49, which means the EL variable impacts the variable of latent KBK by 49%, and the EV variable is 0.25 which means the EV variable impacts the latent variable KBK by 25%.This indicates that there may be an effective and extensive courting among the variables PS, FU, FE, OR, EL, and EV on the students' creative thinking ability (KBK) variables in the Computational Physics course.