Sampling Techniques :spiral_note_pad: (Key Terms :bookmark: (Census:…
Sampling Techniques :spiral_note_pad:
Non-random Sampling Methods
Quota Sampling :bar_chart: :ballot_box_with_check:
Method: Deliberately choose a sample which reflects characteristics of the population until that quota is filled. For example, you can create different age ranges and interview anyone within that range until the quota is filled.
Advantages :check:: Allows a small sample to still be representative of the whole population, no sampling frame is required, allows for easy comparison between different groups within a population and it is quick, easy and inexpensive. More cost effective and easier to choose a sample.
Disadvantages :green_cross:: non random sampling can introduce bias, it is costly and inaccurate since populations have to be divided into groups, the bigger the study the more groups there must be which is more expensive and non responses are not recorded. It is less random.
Opportunity Sampling :bookmark_tabs: :man-woman-boy:
Method: Choose the first people that you encounter. For example, stand outside a busy street and interview the first 20 people you encounter.
Advantages :check:: quick and easy as you can just stand in the street and survey the first 20 people. It is also inexpensive.
Disadvantages :green_cross:: It is unlikely to provide a representative sample and could be bias depending on the location, time of day and how many people you interview (so is highly dependant on the individual researcher).
Stratified Sampling :face_with_cowboy_hat:
Method: the sample is split into 'strata' with distinct characteristics that don't overlap. e.g gender or eye colour.
Advantages :check:: The sample accurately represents the population structure and it guarantees a proportional representation of groups within a population.
Disadvantage :green_cross:: It is difficult since the population have to be clearly classified into different strata and only a selection within each stratum suffers from the same disadvantage as simple random sampling.
The best way to make a sampled strata is: (the number of people in that strata / the number in whole population) x how many people you want to collect (the sample size)
Simple Random Sampling :tophat: :slot_machine:
Method: assign a number to each population and pick a random number. For example, pull some names out of a hat.
Advantages :check:: There is no bias, is easy and cheap for small populations and each sample unit has a known and equal chance of selection.
Disadvantages :green_cross:: Not suitable for a large population (this can be very confusing and expensive) and a sampling frame is needed.
Systematic Sampling :robot_face:
Put a population in order (eg alphabetical) and choose every n'th person. Remember to choose the first item randomly to give an unbiased sample.
Advantages :check:: this is simple and quick to use and us suitable for large samples and large populations.
Disadvantages :green_cross:: Results could generate a pattern, this method could mean you miss faults, a sampling frame is needed and if the sampling frame is not random it could introduce bias.
Key Terms :bookmark:
: Questioning everybody in the population.
: The whole set of items.
: a selection of the population
: a list of everyone in the population.
: numerical observations (weight, time, height)
: a word/description (eye colour)
: something that is measured and can be rounded. Can take any value in a given range (time, height, weight)
: Can only take specific values that our counted (the number of students in a class and the results of rolling a dice).
: Data is split up into class intervals so specific individual values are not known.