The Relationship between Academic Achievement and Experimental Ability in Chemical Field  

Oleh:

Ikeda, Ryosuke ; Sano, Natuki ; Kotani, Akira ; Hayashi, Yuzuru ; Suzuki, Tomomichi

Jenis: Article from Proceeding
Dalam koleksi: 12th ANQ Congress in Singapore, 5-8 Agustus 2014, page 1-7.
Topik: Outliers; Missing values; Attitude towards the experiment; Calculation ability; Robustness
Fulltext: QP2-5.6-P0280.pdf (987.61KB)

Isi artikel

In Japan, it is often said that students tend to drift away from science as a result of amendment of national educational guideline by the Ministry of Education, Culture, Sports, Science and Technology. One of the reasons is that students have less time on experiment, as reported by Sasaki et al. It is said that academic achievement has relation to experimental ability. However, there are few empirical study that considered the relationship between the two. The purpose of this study is to clearly demonstrate correlation and pure relationship between the academic achievement and the experimental ability by statistical approach. In this study, we dealt with achievement test and chemical experiment which were done by college students. Two problem were present in the in the raw data. The first problem was that there were many missing values and outliers. The second problem was that raw data weren’t indicators for evaluating experimental ability themselves. It was difficult to get the result using existing methods to those problems. Hereupon we categorized into two. Since there are many missing values and outliers in raw data, because it is desirable to treat a robustness to outliers of this study. As analysis of missing values, it was necessary that we standardize evaluation for missing values because frequencies that missing values occur are different. We use a few specific actions. It was found that there is necessity considered missing values and there are differences in academic achievement between kinds of missing values. As analysis of outliers, we could grasp how many outliers there are in raw data. Besides, it was possible to propose the effectiveness of using the statistics robust for raw data whose statistics influenced by outliers. It was found that compared the academic achievement by kind of outliers, also data related to calculate a significant influence on academic achievement. Eventually, it was possible to grasp a little correlation between experiment ability and academic achievement data

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali