Data collection methods

What is a census?

where data is collected about/from all members of a population

What is the advantage of a census?

Fully accurate results can be collected

What are the disadvantages of a census?

• Time-consuming AND expensive

• May involve destroying/using up all the members of a population (made up of consumables)

• Massive volume of data to process

What is sampling?

Where data is collected from a subset of the population?

What are the advantages of sampling?

• cheaper and quicker than a census

• less data needs to be analysed

What are the disadvantages of sampling?

It may not accurately represent the population

Can introduce bias

Data set may not cover a large enough group to represent smaller sub-groups

What is (simple) random sampling?

• A process of sampling

• every member of a population has an equal chance of being selected to be part of a sample

• Each item is numbered (given an identifying number)

• Random numbers selected via random number generator

What are the benefits of random sampling?

fully bias free

easy + cheap to implement

What are the disadvantages of random sampling?

Not suitable for large populations where not all members of the population can be listed/numbered

Must have a SAMPLING FRAME

When should you use simple random sampling?

Where there is a small population, or you need a small sample

Where you need to be fully bias-free

What is stratified sampling?

• the population is divided into strata

• random sample of the same proportion is taken from each stratum

• The proportion of a sample belonging to a stratum is equal to the proportion of the population belonging to the stratum

How is the number of members sampled from a stratum calculated?

When can stratified sampling be used?

When the population naturally divide into different strata, and the sample is large

What are the advantages of stratified sampling?

The sample will be representative of the population structure - guarantees PROPORTIONAL REPRESENTATION

The members are selected from each stratum randomly

What are the disadvantages of stratified sampling?

Population must be split into distinct strata

Selection within each stratum suffer from same disadvantages as random sampling