The phrase "a test size" is correct and usable in written English.
It can be used when referring to the dimensions or scale of a test, often in contexts like statistics, research, or product development.
Example: "To ensure the validity of our results, we need to determine an appropriate test size for our study."
Alternatives: "a sample size" or "a testing scale".
Without loss of generality, we assumed Θ0= 0. A test size, α, of 0.05 was used.
By contrast the DeLong test of the AUCs is exceptionally conservative, with a test size far below nominal.
Appropriate fixed matrices of regression coefficients, B, contrast matrices, C and U, and Θ0 were chosen to test a within-subject interaction for a test size, α, of 0.05.
To conduct a test, size-matched stimulus crayfish were placed in the pens at the ends and the focal crayfish placed in an opaque cylinder in the middle of the central arena to settle for 2 minutes.
6000 replications were performed giving estimates with an estimated precision (half width of two-sided 95% CI) on a 5% test size of ±0.6% and ±1.3% on a 50% misclassification rate.
In order to detect a difference of 20% between the two arms, with a two-sided test size of 5% and power 80%, a recruitment of 200 patients would be required for this study.
The sample size was estimated separately for each trial assuming a one-sided test size of 5% and a power of 80%.
Comparisons were made with a Fisher's Exact or a Mann Whitney U test (size of hospital).
Different cost and γ values were tested using a training set size of 5000 and a test set size of 50,000.
This will clarify such issues as "when to look" and "when to pay a price" (e.g., test size and power) and characterize such issues in the context of pharmaceutical industry drug development.
Five different training set sizes (500, 1000, 2000, 4000, 8000) were tested and evaluated with a test set size of 1000.
