Sentence examples for sizes were tested using from inspiring English sources

The 95% CI on fitted effect sizes were derived and the differences between the effect sizes were tested using bootstrapping methods that are robust, conservative, and require fewer assumptions than classical methods[47].

The 95% confidence interval (CI) on fitted effects were derived and the differences between the effect sizes were tested using bootstrapping methods that are robust, conservative, and require few assumptions relative to classical meta-analytic methods [43].

Potential moderators of the observed effect sizes were tested using the Q b statistic (Table  2008).

Finally, to estimate the demographic processes of each population, declines/expansions of effective population sizes were tested using a coalescence-based method.

SVM-RFE is used to select relevant features during classifier training and various feature subset sizes are tested using a leave-one-out cross validation.

Multiple sets of four bearings (207H size) were tested using the least-of-four technique and sudden-death statistics to obtain a Weibull distribution based upon a bivariate model that accounted for load, time, and sample size.

Before employing a phylogenetically based comparative method, phylogenetic autocorrelation in relative brain size, body mass, and relative testicular size was tested using Phylogenetic Independence version 2.0 [17], [17].

The adequacy of sample size was tested using data from smoking cessation trials that showed: 30-40% smoking cessation at 6-months with similar protocols [ 27, 28].

The sample size calculations were tested using simulation, to check that they provided the desired power and α, or the desired confidence interval width.

In the postmortem cohort, given the small sample size, group differences were tested using nonparametric statistics: either Mann-Whitney tests for 2 groups, or Kruskal-Wallis test for 3 groups followed by post hoc Mann-Whitney tests to determine differences between subgroups.

The observed size distributions in suspension were tested using three statistical distributions, such as, log-normal, log-hyperbolic and log-skew-Laplace.