STATISTICS - Definitions Flashcards
(30 cards)
CENSUS
Observes every member of the populations
A - Completely accurate
D - Time consuming and expensive
SAMPLE
A selection of observations taken from a subset of the population.
A - Less data to process
D - Less accurate than a sentence
SIMPLE RANDOM SAMPLING METHOD
Every member of the population has an equal chance of being selected
Every unit is allocated a number within the sample frame
Numbers are called by random number generator or lottery sampling
Repeats ignored
Take the data corresponding to each number
SYSTEMATIC SAMPLING METHOD
Elements are chosen at regular interviews from an ordered list - every fifth person
Sample size / total population - gives the interval
Randomly generate a number in the first interval to be the starting point
QUOTA SAMPLING
An interviewer selects a sample that reflects the characteristics of the whole population
The whole population is divided into groups according to a given characteristic
The size of each group determines the proportion of the sample that should have that characteristics
Interviewees are allocated to a group, continue until quota is full
Ignore non-answers or answers from full quotas
OPPORTUNITY SAMPLING
Taking a sample from people who are available at the time and place the survey is being carried out and fit the criteria you are looking for.
SIMPLE RANDOM SAMPLING A/D
A - Free of bias + easy to implement on small scales
D - Need a sampling frame. Not suitable for large sample sizes
SYSTEMATIC SAMPLING A/D
A - Simple and quick. Suitable for large samples
D - Need a sampling frame. May not catch systematic errors
STRATIFIED SAMPLING A/D
A- Proportional Representation + accurate reflection of population
D - Population must be classified into distinct strata. Stratas have the same d as random sampling.
QUOTA SAMPLING A/D
No sampling frame required
Easy to compare between groups
Non-random - subject to bias
Filled quotas are not recorded
OPPORTUNITY SAMPLING
A -
Easy to carry out and inexpensive
D -
Unlikely to be representative and highly dependent on the researcher
QUALITATIVE
Non-numerical data - Eye colour
QUANTITATIVE
Numerical data - Shoe size
CONTINUOUS
Can take any value in a given range - Data you can measure - Time
DISCRETE
Data you count - can only take specific values
No. of people
CLEANING THE DATA SET
Removing outliers
POINTS FOR COMPARING DATA
-Skew
-Mean / Median - Average
-IQR or standard deviation - Variance
SAY ON AVERAGE
SAMPLE SPACE
The set of all the possible outcomes
UNION
Or AUB
INTERSECTION
And AnB
EVENT
A subset of the sample space associated with a certain outcome
P(AuB) =
P(A) + P(B) - P(AnB)
CONDITIONS FOR BINOMIAL
- Fixed number of trials, n
-Trial is either success or failure
-Trials are independent
-Probability of success, p, is constant
BINOMIAL FORMULA
nCx (P)^x X (1-P)^n-x
