Читать книгу Experimental Design and Statistical Analysis for Pharmacology and the Biomedical Sciences - Paul J. Mitchell - Страница 35

Of the data distributions briefly described above, the majority are only of value to understand the theoretical basis of statistical analysis (the Chi square, t‐ and F‐distributions are important once we get into inferential statistics, but only if we wish to understand the process of how such tests work). In contrast, the most important distribution for the experimental pharmacologist is the Normal Distribution which shall be discussed in detail later (see Chapters 6 and 7).

Figure 4.9 The t‐distribution. The probability density function of the t‐distribution with 1 (dotted line), 2 (dashed line), and 4 (thin solid line) degrees of freedom compared with the Standard Normal Distribution (bold solid line; N(0,1)). For the t‐distribution data the X‐axis values indicate the value of t. For the Standard Normal Distribution, the X values indicate the mean of zero with standard deviations either side of the mean. Y‐axis indicates probability in all cases. Note as the degrees of freedom increase so the probability density function of the t‐distribution approximates towards the Standard Normal Distribution (see also Appendix A.5).

Figure 4.10 The F distribution. The probability density function of the F distribution with 1, 4 (bold solid line), 2, 8 (thin solid line), 4, 20 (dashed line), and 8, 32 (dotted line) degrees of freedom. X‐axis values indicate the F ratio, while Y‐axis values indicate probability density function (see also Appendix A.6)