sampling Flashcards
What is a sample?
a representative subset of the population
What is a population?
the entire set of participants of interest
What is generalizability?
the ability to infer population characteristics based on the sample (external validity)
What are the general types of sampling?
A. Probability sampling:
- The likelihood of any member of the population being selected is known
(recommended method)
B. Non-probability sampling:
- The likelihood of any member of the population being selected is unknown ( or not equal)
(This is not ideal but is often used)
What is simple random sampling?
Each individual has an equal chance of being selected
a. Achieved by random number table or random number generator
b. Ensure individuals in the population have a lot in common!!!
How does a random number table work?
the sample is first defined
i. Participants are ordered from 1 to N
ii. A starting point is picked using appropriate numbers
This process is followed until the sample is complete
How is a random number generator different than a random number table?
a. The sample is placed in a random sequence (not alphabetically)
b. Excel is used the generate the number
(using ‘randbetween (1,N)’ )
What is systematic sampling?
- Population participants are placed in a sequence (no set order)
i. Divide population by sample size (N/n)
ii. Randomly select a starting number
Using systematic sampling how would you select a desired sample of 200 students from a population of 4000 students?
a. First divide: 4000/200 = 20
i. This means every 20th person will be selected
b. Then pick a random starting point: lets guess 1267
i. This means the first person is participant 1267 and then every 20th person after them is chosen.
1267, 1287, 1307, 1327 etc
what is stratified random sampling?
stratified random sampling is when the population is broken down according to the variables of interest
a. This is more difficult to achieve as you need to know a lot about the population and when dealing with many variables can be an added challenge
i. However, given its complexities this method provides a high degree of representation
define cluster sampling:
selecting groups rather than just individuals
- Randomly select from within identified groups
(This is an effective and easy method. However, you must ensure the group members are similar)
What are the 2 non probability techniques? Explain
a. **Convenience sampling: **
i. Using people you have easy access to
ii. Not random
iii. No representation
Pro: convenient and inexpensive
Con: degree of generalizability is questionable
b. **Quota sampling: **relies on the non-random selection of a predetermined number or proportion of individuals.
i. Similar to stratified as you first divide the population into subgroups and then recruit sample individuals until you reach your quota.
1. Group ratios identified
2. individuals are selected on a convenient basis
pro: ensures some degree of presentiveness of all the strata in the population
con: degree of generalizability is questionable
Can we generalize convenience sampling?
no, convenience sampling is not generalizable because it may produce biased data. Since your conduction the survey with a group of easily accessibly people, your results may not be representative of the entire population.
a. Generalization would only be possible if the sample was randomly drawn from the population
Can we generalize quota sampling?
no
What is the difference between convenience and quota sampling?
- in convenience sampling, you continue to sample units until you reach the required sample size.
- in quota sampling you first need to divide your population of interest into subgroups (strata) and estimate their proportions (quota) in the population.
