Stats Flashcards
Advantages and disadvantages of mode? And what type of measurement is it?
Measures Location.
Advantages: Gives most common value and not usually affected by outliers.
Useful for numerical data.
Always is a value from the data set
Disadvantages: Does not make full use of data.
May not be representative as low frequency.
Could potentially be more than one.
Advantages and disadvantages of median, and what type of measurement is it?
Measure of location
Advantages: Not affected by outliers and gives middle piece of data
Disadvantages: Can’t be used for non- numerical data.
Not always a value of the dataset
Doesn’t make use of all the data
Advantages and disadvantages of the mean, and what type of measurement is it?
Measure of location Advantages: Shows the average value. Makes use of all the data. Disadvantages: Affected by outliers and anomalies (for small datasets). Not always a value from the data set. Can't be used in non-numerical data.
Advantages and disadvantages of range, and what type of measurement is it?
Measure of spread
Advantages: Shows the quantity of data.
Reflects the full dataset.
Disadvantages: Affected by outliers
Advantages and disadvantages of inter-quartile range, and what type of measurement is it?
Measure of spread
Advantages: Not affected by outliers
Disadvantages: Doesn’t make use of all the data
Advantages and disadvantages of standard deviation, and what type of measurement is it?
Measures spread
Advantages: Makes use of all data, and for large datasets, outliers have a small input.
Disadvantage: For small datasets, outliers have a big impact.
Define population
The set of things you are interested in
What is a Census?
Observes/measures every member of a population
What is a sample?
Subset of population. Used to find info for the whole population. (Bigger=better)
Define statistic? And what are they used for?
Value calculated from a sample, e.g mean or standard deviation.
Used to estimate parameters, e.g mean for the sample is an estimate for the mean of a population.
(Sample–> Statistic)
What is a parameter?
Number/Value that describes and entire population, such as the mean or standard deviation of the population
(Population–> use parameters)
What is a sampling unit?
Individual unit from population,
E.g particular person in UK
What is a Sampling Frame?
List of all sampling units in a population
Advantages and disadvantages of a sample over a census
Advantages: Quicker and Cheaper.
Less time consuming to process data.
Disadvantages: Less accurate (data may not represent whole population).
Can be biased by choice of sample.
Advantages and disadvantages of a census over a sample
Advantages: More accurate, completely unbiased and represents entire population.
Disadvantages: Slower, more expensive and takes longer to process data (lots of units).
Destroy all your population in testing, e.g opening food products to see contents.
Describe Simple Random Sampling
- Get a list of all members in population
- Number all items in population
- Use random numbers to select sample and ignore repeats
Describe Systematic Sampling
- Number all items in population
- Let n=population size/sample size
- Use a random number from 1 to n to select first item and choose every nth item after that until you have the required sample size
Describe Stratified Sampling
-Divide population into groups (e.g eye colour, age etc)
-Decide how many to sample from each group with ((number in group)/(number in population)) x sample size
- Use simple random sampling to select items from each group
(When question asks for stratified, describe Simple Random process as well!)
What are the Random Sampling Methods?
Simple random
Systematic
Stratified
What are the non-random sampling methods?
Opportunity and Quota
Describe Opportunity Sampling
Sample consists of any items that are available to be sampled. (This method can also be referred to as convenience sampling)
Describe Quota Sampling
Population divided into groups and the same proportion of items from each group are included in the sample just like stratified.
Then use opportunity sampling
For linear interpolation, for median, what is the equation?
Q2-LB/UB-LB = n/2 th value - LF/UF-LF
How much data lies within one standard deviations of the mean for a normally distributed variable?
68% approx
