Variability and sampling 2 Flashcards
Mastery
- Systematic Sampling
* Another method of sampling from the population with a…
* Define the … item
* …= population size/sample size
* Pick every … item
Example:
You want to obtain a sample of 500 individuals from a town of 10,000
people using systematic sampling.
k= population size/sample size
k= 10,000/500
k= 20
e.g. 20 th, 40 th, 60 th, 80 th, ..
- Systematic Sampling
* Another method of sampling from the population with a defined sample size number
* Define the kth item
* k= population size/sample size
* Pick every kth item
- Stratified Random Sampling
* Population is too …
* Divide the population into …
* Size
* Colour
* Randomly select from each …
* Proportionally select… - Cluster Sampling
* Divide the population into …
* Randomly select …
- Stratified Random Sampling
* Population is too large
* Divide the population into “strata” Characteristics
* Size
* Colour
* Randomly select from each strata
* Proportionally select the number - Cluster Sampling
* Divide the population into “clusters”
* Randomly select the cluster to be used
Frequency Distributions
In statistics, a frequency distribution is a table that displays the frequency of various outcomes in a sample.
[1] Each entry in the table contains the frequency
or count of the occurrences of values within a particular group or interval, and in this way, the table summarizes the distribution of values in the sample.
Other types of distributions that are not normally
distributed
- … distribution (… law)
- Example of non-normal distribution
- Pareto distributions are common in:
- Creative fields
- Productivity
- Charitable donations
- Artistic works / talent
- Wealth / Income..
Other types of distributions that are not normally
distributed
* Pareto distribution (80 /20 law)
* Example of non-normal distribution
* Pareto distributions are common in:
* Creative fields
* Productivity
* Charitable donations
* Artistic works / talent
* Wealth / Income
Data types influence the way your data will be
graphed!
why and when
changing it from discrete to continous
Bias: Definition -
negatively and positively skewed bias
Bias: Definition - To unevenly weight something
You might desire negatively skewed data when analyzing exam scores where most students perform well, or in situations where you expect a majority of values to cluster around a higher end, with a few outliers at the lower end.
Confounding variability
* Makes trends or differences in the data…
* Signal to noise ratio (… vs …)
* Confounding variability increases with…
* Increases in … groups
* Is generally … to control for
- Examples of potential confounding variability in a
cardiovascular study: - Time of last meal, sleep quality, nutritional status, outside stress, inflammation, minor infections… etc.
Confounding variability
* Makes trends or differences in the data more difficult to see
* Signal to noise ratio (valuable information vs information)
* Confounding variability increases with diversity of subject pool
* Increases in heterogenous groups
* Is generally very difficult to control for
- Examples of potential confounding variability in a
cardiovascular study: - Time of last meal, sleep quality, nutritional status, outside stress,
inflammation, minor infections… etc.
What does confounding variability look like?
platykurtic
leptokurtic