Cox, Hand and Herzberg (2005, p 47) defines that the statistical methods for the data analysis can be categorized as probabilistic or as descriptive. The former method consists of graphical display and tabulation of great importance both in presentation of conclusion and in detailed analysis. Probability enters in statistical analysis in different ways. The most obvious is that one identifies explicitly that conclusions drawn from limited data are uncertain and that therefore it is a better thing to control and measure probabilities drawing the incorrect conclusions partly but by no means solely as a precaution against over interpretation. The statistical significance test is the most widely referred component based with such probabilities. In order to calculate whether some variations are significant or not statistical tests required to be applied. Statistical significance tests estimate some confidences interval around some estimated mean value. The confidence interval has been derived analytically for coherence and spectra. Statistical tests are just a way of working out the probability of obtaining the observed or even more extreme variations among samples or between expected and observed value if a specific hypothesis usually the null of no difference is true. Once the probability is known the experimenter can make a decision about the variation, using criteria that are understood and used uniformly (Klimesch, 2006, p 209; Mckillup, 2005).
考克斯，手和赫茨伯格（2005，p 47）定义了用于数据分析的统计方法可以分为概率或描述。前者包括图形显示和报告的结论，在详细分析重视制表。概率进入不同的统计分析。最明显的是，一个标识明确，从有限的数据中得出的结论是不确定的，因此它是一个更好的东西来控制和测量的概率部分但绝不仅仅作为一种防范过度解释图的不正确的结论。统计检验是最广泛称为基于组件的等概率。为了计算是否有些变化是重要的或不需要的应用统计试验。统计学意义试验估计一些信心区间估计的平均值。置信区间是连贯和光谱解析。统计测试只是工作的概率得到观察甚至更极端的变化之间的样品或预期值与实测值之间，如果一个特定的假设通常是无差异的零假设为真。一旦概率是已知的实验者可以决定使用的变化，是理解和使用统一的标准（klimesch，2006，P 209；mckillup，2005）。
