Many parametric statistical tests rely on the assumption that the underlying data follows a normal distribution. This includes the t-test for independent means, the paired t-test for dependent means, and z-tests for proportions (when sample sizes are sufficiently large). These tests utilize the properties of the normal distribution to calculate p-values and confidence intervals, allowing researchers to make inferences about population parameters from sample data.

The significance level (alpha) represents the probability of rejecting the null hypothesis when it is actually true. Confidence level is calculated as 1 minus the significance level. Therefore, at a 5% significance level, the confidence level is 1 - 0.05 = 0.95, or 95%.

Non-sampling errors are systematic errors that arise from human mistakes, faulty design, or processing issues. These can occur during observation, data approximation, or the data entry and processing stages.

The degrees of freedom for a contingency table are calculated using the formula (r-1)(c-1), where r is the number of rows and c is the number of columns. Here, (2-1) * (4-1) = 1 * 3 = 3.

Stratified sampling is a method where the population is partitioned into non-overlapping groups called strata based on shared characteristics. Random samples are then drawn from each stratum to ensure representation across the entire population, which increases the precision of the statistical estimates compared to simple random sampling.

The t-test is commonly used to assess the statistical significance of coefficients, including those in regression or association analysis, especially when sample sizes are small and the population standard deviation is unknown. Note: While Chi-square is often used for categorical association, the provided answer key specifies the T-test.

In a t-test for the difference between two independent sample means, the degrees of freedom (df) are calculated as the sum of the sample sizes minus two (n1 + n2 - 2). This adjustment accounts for the estimation of the population variance from two separate samples, ensuring the t-distribution remains accurate for small sample sizes.

The standard error of the mean (a) is the standard deviation of the sampling distribution (4). The base of a point estimate (b) is the mean of the sampling distribution (3). A non-specific or alternative hypothesis (c) is represented by Ha: μ1 ≠ μ2 (1). A parameter value (d) is a descriptive statistic based on all population observations (2).

Non-probability sampling occurs when the selection of elements is based on subjective judgment rather than random selection, meaning the probability of inclusion for each element is not known or is zero.

Coverage error occurs when the sampling frame does not accurately represent the target population. If the selection process fails to include certain groups or over-represents others, the resulting sample is biased, leading to inaccurate inferences about the population.