Stats - Data Collection

population

the whole set of items that are of interest

census

observes or measures every member of a population 

advantages

- should give a completely accurate result

disadvantages

- time consuming and expensive

- hard to process large quantity of data

- cannot be used when the testing process destroys the item

sample

a selection of observations taken from a subset of the population which is used to find out information about the whole population

advantages

- less time consuming and less expensive than a census

- few responses required

- less data to process than in a census

disadvantages

- data may not be as accurate

- sample may not be large enough to give information about small sub groups of the population

sampling units

individual units of a population

sampling frame

a list of individually named or numbered sampling units

simple random

every sample of size n has an equal chance of being selected

how to carry out

use a sampling frame. each person / thing on it is allocated a unique number and a calculator is used to generate different random numbers, ignoring repeats. the people corresponding to these numbers make up the sample. 

simple random advantages and disadvantages

advantages

- free of bias

- easy and cheap to implement for small populations and small samples 

- each sampling unit has a known and equal chance of selection 

disadvantages

- not suitable when the population size or the sample size is large 

- a sampling frame is needed

- the sample may not accurately reflect the proportions of the population from each group if simple random is only used on its own

systematic sampling

the required elements are chosen at regular intervals from an ordered list (sampling frame) 

the first person to be chosen should be chosen at random, using a random number between 1 and n

systematic sampling advantages and disadvantages

advantages

- simple and quick to use 

- suitable for large samples and large populations 

disadvantages

- a sampling frame is needed

- can introduce bias if the sampling frame is not random (patterns may occur when selecting every nth person)

stratified sampling

the population is divided into mutually exclusive strata according to characteristics of the whole population and a random sample is taken from each 

the proportion of each strata in the sample should match its proportion of the whole population

stratified sampling advantages and disadvantages

advantages

- sample accurately reflects population structure 

- guarantees proportional representation of groups within a population

disadvantages 

- population must be clearly classified into distinct strata

- selection within each stratum has the same disadvantages as simple random

quota sampling

the researcher selects a sample that reflects the characteristics of the whole population

how carry out

the population is then divided into groups according to a characteristics of whole population 

and the size of each group determines the proportion of the sample that should have that characteristic

the sample is then chosen by the researcher until all of the quotas have been filled, ignoring those who turn out to fit quota that has already been filled or if they refuse to be interviewed.

quota sampling advantages and disadvantages

advantages

- allows a small sample to still be representative of the population 

- no sampling frame required

- quick, easy and inexpensive 

- allows for easy comparison between different groups within a population

disadvantages 

- non random sampling can introduce bias

- populations must be divided into groups which can be costly or inaccurate 

- increasing scope of study increases number of groups, which adds time and expense

- non responses are not recorded as such

opportunity sampling

taking the sample from people who are available at the tome the study is carried out and who fit the criteria you are looking for

opportunity sampling advantages and disadvantages

advantages

• easy to carry out

• inexpensive

disadvantages 

• unlikely to provide a representative sample

• highly dependent on the individual researcher (introducing bias)

can be made more representative by:

• surveying people in different locations / other parts of town

• surveying people at different times or even random times