โครงสร้างเกี่ยวกับแนวคิดของสัดส่วนประชากรและความน่าจะเป็นของการเกิดผลสำเร็จในการแจกแจงแบบเบอร์นูลี / สมจิตต์ บัวบาน = The conceptual framework of population proportion and probability of success in bernoulli distribution / Somjit Buaban
The objective of this study is to illustrate the basic concept of estimating population proportion and probability of success under Bernoulli distribution. Clear understanding on this matter leads to determination for appropriate or correct choice of statistical method. The study is divided into two parts, i.e. the analysis of basic concept on theory of statistics from examples in statistical textbooks, and a simulation study for investigate on properties of point and interval estimator. It was founded that many statistical textbooks lack of the clear explanation on the basic concept of the estimating population characteristic under Theory of Sample Survey and the estimating of parameter under Theory of Statistical Inference. In addition, the distribution of statistical “a” (number of interest in the sample) approaches normal distribution when “n” (sample size) and “N-n” (difference between of population size and sample size) are large. For interval estimation, the estimator given by M.E.Thompson (1997) has the closet coverage probability to the confidence level. When stratified random sampling is used, the weighted estimator has lower mean absolute error comparing to unweighted estimator.