Populations, Samples and Censuses

A population is the set of individuals in a group. Often information needs to be collected about the population, Taking a census and taking a sample are methods of collecting data about the population.

A govenrnent for example needs to plan the supply of public services in advance. A hospital might cost £100m to build, and should not be built if the demand for it is not there. For this reason, the local population needs to be accurately forecast, along with it age distributionand other relevant quantities.

Every ten years the governments in many countries conduct censuses. Every individual completes a form to submit all relevant information and the government uses to to guide government policies and priorities. Conducting a census gathers complete information but is an expensive and time consuming process. A form must be delivered to every house, completed and returned. If it is not returned, often a representative must call on that house in person to chase up the form. The patience of government for this is limited. Many people do not compete a census form full stop – they are out of the country during the time of the census, they cannot read, they cannot be bothered, they don't believe in government anyway...- and there comes a point when the government stops chasing them for a response. Alternatively the entire census is conducted by personal interview, with government representatives knocking on doors until someone answers (or they give up).

A sample is easier, simpler, quicker and cheaper to conduct. Only a representative fraction of the population need be sampled, with men, women, old, young etc being sampled in the same proportion as they occur in the population.

A sample is much less accurate than a census and has a margin of error, often quoted. For example, political opinion polls are quoted asBeing much cheaper, many more of these can be conducted, and if the subject of the poll is voting intentions for example, trends may be tracked over time, with several such being conducted each month, often by private organisations – newspapers, politicals parties and broadcasters. These are published, agglomerated and then much analysed, especially in the period before elections.

Add comment

Security code
Refresh