Statistical Methods Goals

We have asked to discuss the goals of two statistical methods. Please tell me if Im thinking right

The goal of a census the goal of a census is to determine type, diversity, and distribution of a population within a research area. A substantial sample size is necessary for accurate data analysis.

The goal of Existing sources ( I have not gotten to this one yet)

Survey sampling the principal goal of survey sampling is the Estimation of finite population parameters. The real issue, however, is not whether the population can be viewed as large, but whether the sample can. It is precisely because samples are often very large in survey sampling, that the apparently exclusive use of randomization methods dominated its practice for so long.

The goal of designed experiments is to determine which factors are important with respect to location and scale. A ranked list of the important factors is also often of interest


TS Contributor
The goal of a census, by definition, is to learn about a population by taking a complete count of every item / person in the population. There is no concept of a "sample" when one talks about a census. A census is a 100% sampling.

Existing sources are sometimes referred to as "secondary data" and usually come from government or industry/marketing statistical reports and databases. They often serve as guides for more specific research projects.

The goal of survey sampling is to learn about a particular population of people by drawing a random sample from that population and measuring characteristics or attributes or attitudes that you want to learn about. Sometimes simple random sampling is not the most efficient sampling method, and other procedures such as stratified sampling or cluster sampling are employed. With survey sampling, it is very important to try to learn as much about the population under study before drawing the sample so that the most economical approach can be used.

The goal of a designed experiment is to learn which factors influence a response variable, in a highly controlled environment, by systematically varying the factors (often called independent variables), and observing the resultant changes that occur in the dependent variable.