07. Quantitative methods Flashcards
Primary data
Involves the collection or generation of data with a specific project or task in mind.
Secondary data
Collected for distribution and use by other interested parties.
Cross-section data
Observations that come from different individuals or groups at a single point in time.
Time-series data
A set of observations usually collected at discrete & equally spaced time intervals, e.g. annually.
A population
All members of a well-defined set or group.
(Can be finite or infinite).
A sample
Subset of a population
A parameter
A number describing a whole population e.g. the population mean.
A statistic
A number describing a sample e.g. the sample mean.
Why do we often have to work with samples?
Because it is not cost-effective to gather data on whole populations.
What is it that is desirable for a sample to appropriately represent?
The key features of a population.
For a sample to be unbiased, what must it be?
Representative of the population.
Random vs non-random sampling
- Random: every member of a population has a chance of being selected for the sample.
- Non-random: involves some element of judgement in selecting the sample.
What is the issue if a sample is too small?
It may bias the population estimate.
Sampling methods can be divided into probability & non-probability methods.
Probability methods?
Have a known probability for each member of the population to be selected.
Name 3 techniques included in sampling probability methods.
- Random
- Systematic
- Stratified
Random sampling
Each member of population has equal & known chance of being selected.
Systematic sampling
Every nth record is selected from a list of population members.
Stratified sampling
1st identify characteristics of population that are already known, then selecting a random sample to represent those characteristics in the correct proportion.
Why is stratified sampling often considered superior to random sampling?
Because it reduces sampling error - the possibility of selecting an unrepresentative sample.
Sampling methods can be divided into probability & non-probability methods.
Non-probability sampling?
Members are selected from the population in some non-random manner.
Name 4 examples of non-probability sampling.
- Convenience sampling
- Judgement sampling
- Quota sampling (the non-probability equivalent of stratified sampling)
- Snowball sampling (may be used if the desired sample characteristic is very rare)
Pro & con of snowball sampling
- Reduces search costs
- Introduces bias as it reduces the likelihood that the sample will represent a good cross-section from the overall population.
Continuous data
Can take any value in an interval on the line from minus infinity to plus infinity.
Discrete data
Can only take a finite number of values.