For each data set, the assumptions of parametric testing were checked using Kolmogorov Smirnov and Shapiro Wilk tests.
Normality of quantitative variables (Shapiro-Wilk test) was checked.
The threshold of p value in multiple tests was checked through manipulating the false discovery rate (FDR) value.
The test was checked after 5 min: if the two lines (control line and test line) appeared, the test was considered positive.
Significance of differences between samples fixed at the 0.05 probability level for all statistical tests was checked with the Mann–Whitney U test.
The positive conversion of tuberculin test was checked during treatment period after therapeutic abortion, therefore we thought those could be an indirect evidence of intact cellular immunity.
Furthermore, each growth positive tube in the tests was checked microscopically to confirm that the cell morphology corresponded with the cell morphology of the inoculum.
The proportionality assumption for the Cox regression and the log rank test was checked and verified by using the log-log plot.
Accuracy of the test was checked by parallel use of a quality control strain (Enterococcus faecalis ATCC 29212) as suggested by the Clinical and Laboratory Standards Institute (CLSI) [ 23].
"Our air pollution test was checking if you can see the mountains, which turns out to be a pretty good measure of fine particulate pollution actually".
