Journal of The Korean Association For Science Education (한국과학교육학회지)

The Korean Association for Science Education (한국과학교육학회)
권호 표지
DOI QR코드

DOI QR Code

Analysis of Characteristics of Clusters of Middle School Students Using K-Means Cluster Analysis

K-평균 군집분석을 활용한 중학생의 군집화 및 특성 분석

  • Received : 2022.12.12
  • Accepted : 2022.12.26
  • Published : 2022.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study is to explore the possibility of applying big data analysis to provide appropriate feedback to students using evaluation data in science education at a time when interest in educational data mining has recently increased in education. In this study, we use the evaluation data of 2,576 students who took 24 questions of the national assessment of educational achievement. And we use K-means cluster analysis as a method of unsupervised machine learning for clustering. As a result of clustering, students were divided into six clusters. The middle-ranking students are divided into various clusters when compared to upper or lower ranks. According to the results of the cluster analysis, the most important factor influencing clusterization is academic achievement, and each cluster shows different characteristics in terms of content domains, subject competencies, and affective characteristics. Learning motivation is important among the affective domains in the lower-ranking achievement cluster, and scientific inquiry and problem-solving competency, as well as scientific communication competency have a major influence in terms of subject competencies. In the content domain, achievement of motion and energy and matter are important factors to distinguish the characteristics of the cluster. As a result, we can provide students with customized feedback for learning based on the characteristics of each cluster. We discuss implications of these results for science education, such as the possibility of using this study results, balanced learning by content domains, enhancement of subject competency, and improvement of scientific attitude.

최근 교육에서 교육 데이터마이닝에 관한 관심이 높아지고 있는 시점에 과학교육에서 평가 결과를 활용하여 학생들에게 적합한 피드백을 제공하기 위해 빅데이터 분석의 적용 가능성을 탐색해 보고자 하였다. 연구에서는 국가수준 학업성취도 평가의 24문항에 응시한 2,576명의 평가 자료를 활용하여 비지도 기계학습의 한 가지 방법인 K-평균 군집분석을 이용하여 학생들을 군집화하였다. 학업성취도 평가 자료를 활용한 군집화 결과, 학생들을 6개의 군집으로 나누어 볼수 있었다. 상위권이나 하위권에 비해 중위권 학생들이 다양하게 다른 군집으로 구분됨을 알 수 있다. 군집분석의 결과를 보면, 군집화에서 가장 중요하게 영향을 주는 요인은 학업 성취였으며, 군집별로는 교육과정의 내용 영역별, 교과 역량별, 정의적 특성 면에서 서로 다른 특성을 보이고 있었다. 하위 군집에서는 정의적 영역 중에서 학습의욕이 중요하게 영향을 주고, 교과 역량 면에서는 과학적 탐구 및 문제 해결력과 과학적 의사소통 능력이 중요하게 영향을 주고 있었다. 내용 영역 면에서는 운동과 에너지와 물질 영역에 대한 성취가 군집의 특성을 구분하는 중요한 요인으로 작용하고 있었다. 따라서 평가 자료를 활용해 학생을 군집화한 후, 이러한 군집별 특성을 바탕으로 학생들에게 학습을 위한 맞춤형 피드백을 제공할 수 있을 것으로 판단된다. 본 연구에서는 이러한 연구 결과를 바탕으로 군집분석 연구 결과 활용의 가능성, 내용 영역별 균형 있는 학습, 교과 역량 증진, 과학적 태도의 향상 등 과학교육의 시사점을 제안하였다.

Keywords

References

  1. Ahuja, R., Jha, A., Maurya, R., & Srivastava, R. (2019). Analysis of educational data mining. In: Yadav, N., Yadav, A., Bansal, J., Deep, K., Kim, J. (eds) Harmony Search and Nature Inspired Optimization Algorithms. Advances in Intelligent Systems and Computing, (pp. 897-907). Springer, Singapore. https://doi.org/10.1007/978-981-13-0761-4_85
  2. Asif, R., Merceron, A., Ali, S. A., & Haider, N. G. (2017). Analyzing undergraduate students' performance using educational data mining. Computers & Education, 113, 177-194. https://doi.org/10.1016/j.compedu.2017.05.007
  3. Baek, J. (2022). The characteristics of middle school students' understanding about chemical equations and balancing process of chemical equations. School Science, 15(5), 468-485.
  4. Baek, J., Lee, J., & Choi, W. (2022). Analysis on the characteristics of academic achievement of middle school students about 'composition of matter': focusing on the results of the national assessment of educational achievement(NAEA). Journal of the Korean Chemical Society, 66(2), 136-149.
  5. Cho, S., Ku, N., Kim, H., Lee, S., & Lee, I. (2019). OECD programme for international students assessment : an analysis of PISA 2018 results. Korea Institute for Curriculum and Evaluation(KICE) Research Report RRE 2019-11.
  6. Choi, J. & Chung, H. (2021). K-means clustering and random forest analysis of factors In fluencing group classification by job preference. Korean Journal of Educational Research, 59(8), 113-144.
  7. Dong, H., Kim, G., Kang, M., Jang, E-S., Sung, K., Yang, S., Kim, S., Lee, J., Ku, J., Park, S., Kim, S., Choi, W., Kim, Y., Lee, K., & Kim, H. (2018). A Study on the development of national assessment of educational achievement according to the 2015 revised national curriculum. Korea Institute for Curriculum and Evaluation(KICE) Research Report RRE 2018-4.
  8. Hartigan, J. A., & Wong, M. A. (1979). Algorithm AS 136: A k-means clustering algorithm. Journal of the royal statistical society. series c (applied statistics), 28(1), 100-108.
  9. Hyun, J. H., & Lee, J. W. (2015). A study on the effects of teachers' feedback on high school students writing performance. The Journal of Humanities Studies, 98, 431-455.
  10. Joo, Y. J., Chung, Y. L., & Lee, Y. K. (2011). The structural relationship and latent means analysis of gender among academic self-efficacy, interest, external motivation and science achievement for high school students. Journal of the Korean Association for Science Education, 31(6), 876-886. https://doi.org/10.14697/JKASE.2011.31.6.876
  11. Kim, H., & Jung. H. (2020). A case study of yielding NAEA assessment results of sub-domains in science applying the cognitive diagnosis model. The Journal of Research in Education, 33(4), 89-113.
  12. Kim, H., & Jung. H. (2021). Characterizing the academic achievement of middle school students in science subject based on latent profile analysis. The Journal of Curriculum and Evaluation, 24(2), 177-198. https://doi.org/10.29221/jce.2021.24.2.177
  13. Kim, K. (2017). A study on the functions of quality management of curriculum by national assessment of educational achievement. Journal of Curriculum Evaluation, 20(4), 145-173. https://doi.org/10.29221/jce.2017.20.4.145
  14. Kim, N. O., Park, M. A., Lee, B. N., & Sohn, W. S. (2018). The role of teacher characteristics and feedback in developing elementary students' affective and cognitive achievement. Journal of Curriculum and Evaluation, 21(2), 129-151.
  15. Kim, Y., & Kim, S. (2019). Finding an alternative model of the national assessment of educational achievement based on the new trend of academic ability and the 2015 revised curriculum. Journal of Education & Culture, 25(2), 149-173. https://doi.org/10.24159/joec.2019.25.2.149
  16. Kim, Y., & Lee, S. (2021). Analysis of middle school students' responses characteristics revealed in solving the problem related to blood circulation in national assessment of educational achievement. Biology Education, 49(4), 443-456.
  17. Kim, S.-S. (2009). Automated K-means clustering and R implementation. The Korean Journal of Applied Statistics, 22(4), 723-733. https://doi.org/10.5351/KJAS.2009.22.4.723
  18. Ko, S. (2018). Predicting learning achievement using big data cluster analysis - focusing on longitudinal study. Journal of Digital Contents Society, 19(9), 1769-1778. https://doi.org/10.9728/dcs.2018.19.9.1769
  19. Koo, D., & Jung, S. (2021). Analysis of academic achievement data using AI cluster algorithms. Journal of The Korean Association of Information Education, 25(6), 1005-1013. https://doi.org/10.14352/jkaie.2021.25.6.1005
  20. Ku, J., Kim, K. J., Kim, J-S., Park, S. B., Park, J., Sung, K., Lee, K. S., Lee, S., Lee, J., Jang, E-S., Kang, J., Choi., W., Shim, K-C., & Shin, M-K. (2019). The analysis of the characteristics and trends of academic achievement based on the results of the national assessment of educational achievement in the 2009 revised curriculum. Korea Institute for Curriculum and Evaluation(KICE) Research Report RRE 2019-7.
  21. Kwak, Y. (2017). Exploration of features of korean eighth grade students' attitudes toward science. Journal of the Korean Association for Science Education, 37(1), 135-142.
  22. Kwak, Y. (2018). Effects of educational context variables on science achievement and interest in TIMSS 2015. Journal of the Korean Association for Science Education, 38(2), 113-122.
  23. Lee, B. (2015). Analysis on the characteristics of national assessment of educational achievement(NAEA) items for the science subject: with a focus on classical mechanics. New Physics: Sae Mulli, 65(5), 485-495. https://doi.org/10.3938/NPSM.65.485
  24. Lee, B., & Lee, I. (2015). Analysis on the characteristics of national assessment of educational achievement(NAEA) items for science subject: with a focus on optics. Journal of the Korean Association for Science Education, 35(3), 465-475. https://doi.org/10.14697/jkase.2015.35.3.0465
  25. Lee, B., & Sohn, W. (2017). The effects of formative feedback on basic psychological needs and classroom engagement: teacher-student relationship as a moderator. Journal of Educational Evaluation, 30(1), 123-143.
  26. Lee, D. N., & Lim, C. (2018). Statistical methods for testing tumor heterogeneity. The Korean Journal of Applied Statistics, 32(3), 331-348. https://doi.org/10.5351/KJAS.2019.32.3.331
  27. Lee, J. (2019). Analysis of middle school students' vision concepts by achievement level. New Physics: Sae Mulli, 69(7), 727-736. https://doi.org/10.3938/npsm.69.727
  28. Lee, J. (2020). Analysis of middle school Students' characteristics of understanding motion and energy: focused on analysis of national assessment of evaluation achievement. The Journal of Learner-Centered Curriculum and Instruction, 20(17), 815-835.
  29. Lee, J., & Chung, H. (2019). The analysis of relations among PISA 2015 science achievement, teaching and learning on science, academic self-efficacy: using multilevel structural equation modeling. Journal of Research in Curriculum & Instruction, 23(2), 93-107.
  30. Lee, J., & Chung, Y. (2014). An analysis of structural relationship among the attitude toward science, science motivation, self-regulated learning strategy, and science achievement in middle school students. Journal of the Korean Association for Science Education, 34(5), 491-497. https://doi.org/10.14697/jkase.2014.34.5.0491
  31. Lee, J., & Ku, J. (2020). Analysis of the characteristics of middle school students' science academic achievement based on the results of NAEA in the 2009 revised curriculum. Journal of the Korean Association for Science Education, 40(4), 429-436.
  32. Lee, J., Ku, J. Choi, W, Shim, K. C., & Shin, M. K. (2020). Analysis of achievement characteristics by achievement standard of the middle school curriculum based on the national assessment of educational achievement. Journal of the Korean Association for Science Education, 40(5), 473-483. https://doi.org/10.14697/JKASE.2020.40.5.473
  33. Lee, J., & Shim, K. C. (2021). Analysis of the response characteristics on constructed response items of national assessment of educational achievement (NAEA) - life science domain -. Biology Education, 49(3), 431-441.  https://doi.org/10.15717/BIOEDU.2021.49.3.431
  34. Lee, M., Dong, H., Park, I., Kim, W., Seo, M., Jung, H., Kim, K., Kang, M., Jang, E, Sung, K., Rim, H., Kim, S., Pae, J., Kim, S., Lee, J., Park, J., Yang, K., Kang, T., & Shin, Y. (2017). A study on the enhancement of the NAEA system incorporating the aims of the 2015 revised national curriculum. Korea Institute for Curriculum and Evaluation(KICE) Research Report CRE 2017-8.
  35. Lim, S. A., & Lee, J. (2016). Affective factors as predictors of math achievement: comparison of OECD high performing 10 countries in math. Journal of Educational Evaluation, 29(2), 357-382.
  36. Liou, P. Y. (2021). Students' attitudes toward science and science achievement: An analysis of the differential effects of science instructional practices. Journal of Research in Science Teaching, 58(3), 310-334. https://doi.org/10.1002/tea.21643
  37. MacQueen, J. (1967). Classification and analysis of multivariate observations. In 5th Berkeley Symp. Math. Statist. Probability (pp. 281-297).
  38. Ministry of Education (MOE). (2022). 1st comprehensive plan for guaranteeing basic academic ability (2023-2027).
  39. Park, H., & Cho, K. (2005). K-means clustering of social indicator survey data. Journal of The Korean Data Analysis Society, 7(2), 465-476.
  40. Park, M., & Heo, G. (2021). Development of prediction model for middle school students' underachievement using machine learning classification techniques: a focus on DELS. Journal of Educational Technology, 37(3), 627-648.
  41. R Core Team. (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
  42. Sang, G., Kim, K., Park, S., Jeon, S., Park, M., Lee, J., & Min, Y. (2020). The trends in international mathematics and science study (TIMSS): finding from TIMSS 2019 for korea. Korea Institute for Curriculum and Evaluation(KICE) Research Report RRE 2020-10.
  43. Seo, M., Kang, M., Kim, G., Kim, M., Kim, S., Kim, S., Kim, H., Dong, H., Park, S., Sung, K., Yang, S., Lee, J., Jang, E-S, Jung, H., & Seo, E. (2018). Developing items for affective domain of national assessment of educational achievement(NAEA) and verifying their quality. Korea Institute for Curriculum and Evaluation(KICE) Research Report ORM 2018-39-2.
  44. Seo, M., Kim, K., Lee, J., Jeon, S., Kim, S., & Min, Y. (2021). Comprehensive analysis on the results of TIMSS 2019 and its trends. Korea Institute for Curriculum and Evaluation(KICE) Research Report RRE 2021-5.