Qnt 561 Week 2

Submitted by: Submitted by

Views: 873

Words: 303

Pages: 2

Category: Business and Industry

Date Submitted: 03/21/2011 09:33 PM

Report This Essay

21. What is sampling error? Could the value of the sampling error be zero? If it were zero,

what would this mean?

Answer:

A population accounts for absolutely everyone in consideration for observation whereas sample is only a part which is considered for observation to reach the conclusion on selected sample. The collected sample should correctly represent the population in unbiased manner for establishing the success of experiment from the result of observation. Sampling error is the result of difference between estimation of statistics from a sample and population.

Let us take an example of purchasing the IT products in a firm. Initially, Chief Technological Office assigns the responsibility to managers for selecting ten employees so that he collects their opinion and advice on the product. But he realized that managers can select their favorite subordinates which can make the sample biased. So he put the id and name of each employee in excel sheet and created one extra corresponding column for each employee id which is filled by executing the function of random number. Then, selecting the top ten results after sorting the random numbers gives him unbiased sample.

In above example, population could have been taken participation from each employee i.e. data is collected from each member of population. But the sample has been chosen as representative of population to draw the conclusion. Hence, the difference between results of two scenarios is evaluated by sampling error.

Sampling error can be zero in some of the cases. If we have to take the opinions of all the engineers in above case, and sample takes the true proportion of 100% engineer. It can be stated in other way as if the sampling error is zero then the population is uniform or perfect representative sample of population is taken for research purpose.