Open AccessGenerating Fuzzy Interval Data and Its Application to Find The Relation Between Math Anxiety with Self Efficacy Using Correlations Analysis
Author(s) -
Hari Susanto,
Tatik Sutarti,
Agustina Sri Hafidah
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1254/1/012084
Subject(s) - mathematics , fuzzy set , fuzzy logic , fuzzy mathematics , data mining , correlation , defuzzification , fuzzy number , interval (graph theory) , relation (database) , correlation coefficient , fuzzy classification , pearson product moment correlation coefficient , type 2 fuzzy sets and systems , artificial intelligence , statistics , machine learning , computer science , combinatorics , geometry
In this article, adding a dimension that has not yet been explored in published literature, we investigate the process of generating fuzzy interval data using a crisp data, applying the process to the students’ anxiety and self-efficacy crisp data in learning mathematics, and determining the two variable correlations using Cheng and Yang fuzzy correlations. We take Fuzzy correlation developed by Cheng and Yang because of being easy to understand, and its development adopts the formula of Pearson correlation. The data used are fuzzy interval data and trapezoid fuzzy. In this case, the crisp data taken using traditional questionnaires cannot be employed. The findings of this research reveal that (1) the crisp data fuzzification process to be interval fuzzy provides a new alternative in obtaining fuzzy interval data; (2) fuzzy correlations using fuzzification results provide the conception that the anxiety has a negative correlation at a moderate effect level. The correlation coefficient shows the same conclusions with Pearson correlation and fuzzy correlation using original fuzzy data obtained from data collection.