Sampling Flashcards
Inferential statistics
A set of methods that enable us to make probabilistic statements about the characterizations of populations on the basis of sample information
Population
Any entire collection of people, animals, plants or things from which we may collect data. A population can be finite or infinite in size (too large to ennumerate). It is often not possible to collect data from entire population, so we select sample.
Sample
A group of units selected from a larger group, it is invariably finite in size.
Parameters
Measurements that describe a population
Simple Random Sampling
All members of the population have an equal chance of being selected as part of the sample
Stratified Random Sampling
Divide the population into two or more homogenous subgroups and then take a simple random sample
Rationale
Divide by rational interest categories
Cluster sampling
A population is divided into clusters. A random sample of the clusters is is then taken, individuals are then randomly selected from these clusters
Systemic sampling
Makes use of regular intervals between the individuals selected eg. interview every 3rd male
Spatial sampling
Spatial sampling from maps or other spatial sampling frames may be: point samples, area quadrant samples, line traverse samples
Quadrant
A square, rectangle, circle or other shape (plot) that is used as a sample unit
Transect
A line along which measurements are made continuously or at regular intervals
Non-probability Sampling
Judgemental or purposive samples: Convenient/opportunistic and/or volunteers
Bias
A good sample is UNBIASED, so that estimates of population parameters are close to the true value
Accuracy
A good sample provides an estimate of a population parameter that is as close as possible to the true value
