Fulfilling the law of a single independent variable and improving the result of mathematical educational research

• Published in: Journal of physics. Conference series 2018-01, Vol.948 (1), p.12010
• Main Authors: Pardimin, H, Arcana, N
• Format: Article
• Language: English
• Subjects: Education; Factorial design; Independent variables; Physics; Qualitative analysis; Research methodology; Robustness (mathematics); Statistical analysis; Statistical methods; Teaching methods
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/948/1/012010

Many types of research in the field of mathematics education apply the Quasi-Experimental method and statistical analysis use t-test. Quasi-experiment has a weakness that is difficult to fulfil "the law of a single independent variable". T-test also has a weakness that is a generalization of the conclusions obtained is less powerful. This research aimed to find ways to reduce the weaknesses of the Quasi-experimental method and improved the generalization of the research results. The method applied in the research was a non-interactive qualitative method, and the type was concept analysis. Concepts analysed are the concept of statistics, research methods of education, and research reports. The result represented a way to overcome the weaknesses of quasi-Experiments and T-test. In addition, the way was to apply a combination of Factorial Design and Balanced Design, which the authors refer to as Factorial-Balanced Design. The advantages of this design are: (1) almost fulfilling "the low of single independent variable" so no need to test the similarity of the academic ability, (2) the sample size of the experimental group and the control group became larger and equal; so it becomes robust to deal with violations of the assumptions of the ANOVA test.