Wine Chemistry and Biochemistry

13 Statistical Techniques for the Interpretation of Analytical Data 679

Table 13.1Results of the one-sample t test for 15 replicate values from a new analytical method
for a standard material with known concentration of 0.34 g (Massart et al. 1990)

Reference
Mean SD N constant (μ 0 ) t-value df P-value

0.316 0.056 15 0.34 –1.67 14 0. 117 ∗
t-value=value of the statistic tcal,df=degrees of freedom, P=associated probability
∗The two meansμandμ 0 are not different (P>0.05).

for 15 replicate values from a new analytical method for a standard material with
known concentration of 0.34 g (Massart et al. 1990), obtained witht-test for single

meanprocedure in theBasic Statistics and Tablesmodule of STATISTICA program

version 7.1 (StatSoft, Inc., http://www.statsoft.com/)..) Since theP-value associated

with the t-value is greater than 0.05, we cannot reject the null hypothesis at the 95%

confidence level, and there are no systematic errors in this analytical method for

this sample. The same conclusion is obtained from the 95% confidence interval for

μ(0.285, 0.347) that includes the reference value. Using theDescriptive statistics

procedure in the same module, the value of the W statistic in Shapiro-Wilk’s test is

nonsignificant, (W = 0.970, P = 0.86>0.05), and the normality could be accepted

(also from Fig. 13.1).

0.20 0.22 0.24 0.26 0.28 0.30 0.32 0.34 0.36 0.38 0.40 0.42

Value

–2.0

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

Expected Normal Value

Fig. 13.1Normal probability plot for data