## Table 3

The analysis of the t-test for means of DEF failures in 2010-2014 for 24 h+, 24 h−, 48 h+, 48 h−, 72 h+ and 72 h−, where W is the value of Shapiro – Wilk test, F the value of Levene test*.

Variable Normality Homogeneity of variance n Mean t-test
Mean 24 h+ W = 0.94 18 10.66
p = 0.33 F = 3.74 t = 3.42
Mean 24 h− W = 0.89 p = 0.0095 18 4.03 p = 0.002
p = 0.04
Mean 48 h+ W = 0.90 18 9.17
p = 0.054 F = 2.26 t = 2.21
Mean 48 h− W = 0.96 p = 0.10 18 5.91 p = 0.03
p = 0.63
Mean 72 h+ W = 0.92 14 7.23
p = 0.22 F = 2.22 t = 2.13
Mean 72 h− W = 0.92 p = 0.16 14 4.31 p = 0.04
p = 0.23
*

In the case of the assumption of normality of our distributions, we formulate hypothesis H0: the distribution of data is normal for mean 24 h+ as for mean 24 h−. Using the Shapiro-Wilk test (for a small sample) we observe that the assumption on normality is not fulfilled for mean 24 h−, because p = 0.04, < 0.05. Thus, we reject H0 and accept H1, which says that the distribution of data is not normal. Analogically, we check assumptions for the Levene test.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.