Case introduction

 

Sampling survey to the total population of Heshun county in Shanxi province in year 1998

 

Description:

We need to take some samples from 326 administrative villages which is stratified by town to investigate (the minimum investigated unit is administrative village) the total population of Heshun county in Shanxi province in 1998, require that the total cost isn't greater than 100,000 yuan , the initial input is 5000 yuan, the investigation cost of the sample villages in each town can be approximately estimated, the administrative village size of each town is known . According to the previous investigated data, the standard deviation of the village population in each layer can be estimated .

 

Sampling box:

326 administrative villages of Heshun county

 

Layer:

10 towns

 

Sampling method:

Stratified random sampling

 

Condition:

  1. The total cost is 100,000 yuan;
  2. The initial input is 5,000 yuan ( This cost is the fixed cost that is irrelevant to the sample size, include organization 、 propaganda 、 design of the questionnaire and the necessary device, and so on) ;
  3. The average investigation cost of each sample village is 1,000 yuan;
  4. The average analysis cost of each sample village is 200 yuan (the investigated data are checked, arranged, etc).
According to the previous investigated results, the standard deviation of the village population in each layer (town) can be estimated (Standard deviation is the root of the mean of sum of squares of deviations, which reflects the dispersion degree of this layer) as follows:

In the table: CODE is the town number, NAME is the town name, NUM is the administrative village size of this town, STDEV is the standard deviation of the population of this town surveyed previously.

Request:

  1. Total optimal sample size and sample size of each layer (town) ;
  2. Distribute the samples;
  3. Estimate the total population and sampling accuracy of Heshun county according to the sampling sites and the investigated data.