Year 1 Statistics Flashcards
Census
Gathering the required information from everyone within the population
Simple random sampling
Takes a random selection from the population to use as a sample
Stratified sampling
Proportionally represents each sub group in a population. Each sub group is randomly sampled
Quota sampling
Non randomly people are selected from a quota of sub groups they are not necessarily proportionally represented
Systematic sampling
A population is ordered and every nth item is put in the sample
Opportunity (convenience) sampling
Using samples that are easily accessible
Cluster sampling
Population is divided into sub groups and samples only come from a few of those sub groups
Self selected sampling
Individuals have chosen to be in the sample
Variables
Properties being measured
Frequency
Number of occurrences of a particular result
Categorical data
Each item has a category rather than a numerical significance (qualitative)
Discrete numerical data
Not always an integer but either on number or another
Ranked data
The data is positioned within a group of
Continuous numerical data
Variables could take any value
Quantitative data
Numerical data
Qualitative data
Wordy data calculations can not be used
Measures of central tendency
Mean
Median
Mode
Midrange
Measures of spread
Range
Maximum and minimum values
Quartiles
Deviation and variance
Sxx
Sum of the squared deviation
Sigma (x-mean)^2
s^2
Variance
Sxx / (n-1)
s
Standard deviation
Square root (Sxx / (n-1))
Ways to calculate outliers
Anything 1.5 times larger than interquartile range above/below the upper/lower quartile
Anything more than 2 standard deviations from the mean
Univariate data
One variable
Bivariate data
Two variables
Multivariate data
Many variables
Correlation coefficients
Between -1 and 1 where -1 is a perfect negative correlation and 1 is a perfect positive correlation
Trial
An experiment with only two possible outcomes/events
X~B(n,p)
P(X=r)
nCr p^r(1-p)^(n-r)
When to use the binomial distribution model
There are n independent trials
Two possible outcomes with fixed probabilities
Find the expectation of a binomial distribution
E(x) = np
Binomial hypothesis testing
P-value > significance level
Do not reject H0
P-value < significance level
Reject H0
