Sampling Flashcards
(22 cards)
What are the differences between census and sampling
What sampling techniques do you need to know
When should each sampling technique be used
Width is 3 to 6 minutes is 4 not 3. For inequalities if one does not have an equal to sign then it doesn’t count hence, the width would be the two numbers in the inequality subtracted
What reason is used to justify use of a histogram
Time is continuous
When and how should you use simple random sampling and give an example
When every member of the population has an equal chance of being selected.
The population is small and accessible.
You want to avoid bias.
How:
Number every member of the population.
Use a random number generator or draw lots.
Example:
You have a list of 100 students and want to select 10. Assign numbers 1–100 and use a random number generator to pick 10 numbers
When and how should you use systematic sampling
When to use:
When the population is ordered in some way.
You need a quick, easy method and assume the list has no pattern that could bias results.
How:
Number the population.
Choose every nth member (after selecting a random start).
Example:
You want to survey customers entering a store. If 1,000 enter daily and you want 50, select every 20th customer after picking a random number between 1 and 20.
When and how should you use stratisfied sampling
When to use:
When the population has distinct subgroups (strata) and you want representation from each.
You want your sample to reflect the population proportions.
How:
Divide the population into strata.
Calculate the number to sample from each group:
Use simple random sampling within each stratum.
Example:
A school has 300 girls and 200 boys, and you want a sample of 50 students.
→ 30 girls and 20 boys should be randomly selected
When and how should you use Quota sampling
When to use:
When time/resources are limited.
You need quick results but still want population representation.
How:
Split population into groups.
Interviewer is given a quota (e.g., 10 males aged 18–25) and selects individuals meeting the criteria, typically through convenience.
Example:
A market researcher needs 40 responses: 20 men and 20 women. They approach people on the street until each quota is filled
When and how should you use opportunity (convenience) sampling
When to use:
When time and cost are very limited.
You don’t need statistical representativeness (often used in preliminary studies).
How:
Use the most easily available people.
Example:
You survey the first 10 people you meet at the shopping centre
When to use census vs sampling
A census involves collecting data from every single member of the population.
When to use:
When the population is small, or
When you need complete accuracy (no estimation).
Advantages:
More accurate (covers everyone).
No risk of sampling bias.
Disadvantages:
Time-consuming.
Expensive.
Not practical for large populations.
Definition:
Sampling means collecting data from a smaller group (a sample) selected from the population.
When to use:
When the population is large.
When there are time, cost, or resource constraints.
When an estimate is acceptable.
Advantages:
Faster and cheaper.
Less data to handle.
Still useful if well-designed.
Disadvantages:
Less accurate.
Risk of sampling bias if not done carefully.
for all sampling methods, what is the sampling unit and sampling frame
unit : individuals of a population
frame : list of sampling units, e.g number of students in a school
