# 使用统计工具

4.5.1.1使用统计工具

§假设检验(使用T-tests由于小样本大小),
§在于按斯皮尔曼等级相关的方法,
§倍数评分的决心。

4.5.1.1 Statistical Tools Used

Since this report has been structured around ordinal data and not cardinal data, traditional statistical measures associated with ordinal data such as mean, standard deviation could not be calculated. The following tools have been used for the purposes of analysis:

§ Hypothesis testing (using T-tests due to small sample size),
§ Co-efficient of Rank Correlation as per Spearman’s method,
§ Co-efficient of Determination.

Correlation is a measure of the degree of interdependence and association which exist amongst two variables (Arlene, 1995). Correlation co-efficient can be computed in various ways, but for analyzing ranked data in this research, Edward Spearman’s co-efficient of rank correlation as well as Kendall’s co-efficient of rank correlation (Dixon, 1992) may be used.

The rank correlation coefficient devised by Spearman has been widely used for analyzing ordinal data (Bryan, 1994). As per Vaughan (2003), it can be regarded as the non-parametric counterpart of Pearson’s correlation coefficient. Its advantage stems from the fact that it does not require any assumption of the data’s distribution pattern. The formula for calculating the same is:Correlation Co-efficient = 1 – [6 * ∑Diff2]/[No(No2 – 1 )]