The T-test is primarily used when the population standard deviation is unknown and must be estimated from the sample. While it is often used for small samples (n < 30), the defining requirement for using the T-distribution rather than the Z-distribution is the unknown population standard deviation.

The Law of Statistical Regularity states that a moderately large group of items chosen at random from a large population will, on average, possess the characteristics of the entire population. This principle forms the theoretical foundation for sampling theory, allowing researchers to make inferences about a population based on a representative sample.

Sampling error refers to the natural variation that occurs when a sample is used to estimate a population parameter. Because a sample is only a subset of the total population, the statistic calculated from it (such as the sample mean) will rarely perfectly match the true population parameter. This difference is not due to mistakes in data collection but is an inherent characteristic of the sampling process itself, which decreases as the sample size increases.

Sampling is preferred when the population is infinite (making a census impossible) or when only approximate results are needed, which saves time and cost. While heterogeneity (option 2) often requires larger samples, it does not inherently make sampling 'better' than a census, which would be more accurate for heterogeneous populations.

The Kruskal-Wallis test is a non-parametric method used to determine if there are statistically significant differences between two or more groups of an independent variable on a continuous or ordinal dependent variable. It requires one nominal variable (the grouping variable) and one measurement variable (the dependent variable) to perform the rank-based analysis.

For a chi-square test to be valid, the sample must be random, the total observed and expected frequencies must be equal, and each expected frequency should generally be at least five to ensure the approximation to the chi-square distribution is reliable.

The t-test is specifically designed for hypothesis testing when sample sizes are small (typically n < 30) and the population standard deviation is unknown. It uses the sample standard deviation as an estimate for the population parameter. In contrast, the Z-test is used for large samples or when population parameters are known.

Sampling error generally decreases as the sample size increases because the sample becomes more representative of the population. Conversely, non-sampling error, which includes human errors, measurement biases, and data processing mistakes, tends to increase with larger sample sizes due to the greater complexity and volume of data handling required.

The standard deviation of a sampling distribution, such as the distribution of sample means, is technically defined as the standard error. It quantifies the amount of sampling error inherent in a statistic, indicating how much the sample statistic is expected to vary from the true population parameter across different samples.

The null hypothesis (H0) is a statement that there is no effect or no difference. It serves as the default position that researchers attempt to test against, assuming that any observed differences are due to chance or sampling error.