7. Estimation and Inference

Sample Plan

Sample Plan – set of rules used to select a sample

Simple Random Sampling

Simple Random Sample – subset of population where each element has equal selection probability.

Systematic Sampling

Systematic Sampling – method used when unable to code/identify all population members; selects every kth member to reach desired sample size.

Sampling error

Sampling Error – difference between observed statistic and estimated value due to using population subsets.

Stratified Random Sampling

Stratified Random Sampling – method where population is divided into strata based on classification criteria;

simple random samples are drawn proportionally from each stratum and combined into one sample.
prob

Cluster Sampling

Cluster Sampling – method requiring division of population into clusters (subpopulation groups);
each cluster represents entire population;
certain clusters are selected as a whole via simple random sampling.
prob

Convenience Sampling

Convenience Sampling (по удобству) — method selecting elements based on researcher’s accessibility; samples may not represent entire population, limiting sampling accuracy.
non-prob

Judgmental Sampling

Judgmental Sampling — method of selectively choosing elements based on researcher’s knowledge and professional judgment. (best of non-prob. sampling)
non-prob

Central Limit Theorem

Standard Error of the Sample Mean

Она показывает, насколько точно выборочное среднее оценивает истинное среднее. Чем больше выборка, тем меньше разброс.

Bootstrap

Jackknife

Jackknife – method of selecting samples by removing one observation at a time from original data set without replacement.