A test statistic defines how closely your data’s distribution corresponds to the distribution predicted by the statistical test’s null hypothesis. The frequency with which each observation occurs is defined by the data distribution, which can be explained by its central tendency and variation based on the central tendency. Because various test statistics predict various distributions, selecting the appropriate statistical test for your theory is critical. Using the central tendency, deviation, sample size, and quantity of regressors in your statistical model, the test statistic summarises your measured data into a single number. In general, the test statistic is computed by dividing the sequence into your data (i.e., the relationship between variables or the difference between groups) by the variance of the data. (i.e., the standard deviation).
The Z-statistic, which has the standard distribution under the null hypothesis, is the test statistic for a Z-test. Assume you run a two-tailed Z-test with 0.05 and get a Z-statistic (also known as a Z-value) of 2.5 based on your data. This Z-value is equivalent to 0.0124. You certify statistically significant and reject the null hypothesis.
