Title :
Can text mining technique be used as an alternative tool for qualitative research in education?
Author :
Lee, Chi-Kwan ; Chin-I Cheng ; Zeleke, Aklilu
Author_Institution :
Dept. of Math., Central Michigan Univ., Mount Pleasant, MI, USA
fDate :
June 30 2014-July 2 2014
Abstract :
Qualitative research methodology is commonly used for education research. The main advantage is that it is capable of gaining more rigorous insights of research objectives when compared with quantitative research. However, due to the extensive time required, the sample size for a qualitative research is usually very small, and can be as small as two subjects. As a consequence, the results are difficult to be inferred to the population. Text mining methodology is a modern technique to analyze text data. It has been successfully applied to a wide variety of business problems. However, it has not yet been applied as a tool for qualitative research in education. In this study, we attempt to apply text mining method to analyze students´ written reasons of a question related to the concept of variation. Four sections of an introductory statistics course participated in the study (n = 218). The results suggest text mining method can be an alternative, if applied appropriately. The strength and weakness of using text mining for qualitative research are investigated.
Keywords :
data analysis; data mining; educational computing; qualitative education research; text data analysis; text mining technique; Education; Filtering; Histograms; Sparse matrices; Standards; Synthetic aperture sonar; Text mining; Histogram; Misconception; Singular Value Decomposition; Text Cluster; Text Filtering; Variation;
Conference_Titel :
Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), 2014 15th IEEE/ACIS International Conference on
Conference_Location :
Las Vegas, NV
DOI :
10.1109/SNPD.2014.6888691