About non-probabilistic sampling
Probabilistic sampling allows a representative sample, and the result of a research from a probabilistic sampling can be extended to the whole population.
Sometimes we do not have enough resources to do a probabilistic sampling. A simpler sampling method is non-probabilistic sampling. We may choose a non-probability sampling due to the following reasons.
- Research budget
- Time limit
- Accepted error rate
About available sampling
Collection of sampling unit that is readily available, and can be used directly for a research.
For example, taking samples of people in a mall for every person as we go, without randomization.
This sampling method is not recommended as it causes bias.
This sampling method is usually used as a preliminary/starting point to quickly gain an insight about a population, before deciding to use a probability sampling to the population.
About purposive sampling
Sampling method where the sampling unit are chosen because of a specific characteristic or criteria, and ignoring sampling units that does not match this criteria.
The criteria for choosing is determined entirely by the researcher.
About quota sampling
Quota sampling is a non-probability sampling technique used in research where a researcher selects a sample that reflects certain characteristics of the population.
It involves dividing the population into subgroups (called quotas) based on specific traits such as age, gender, income, or education level.
Once the quotas are determined, participants are selected from each subgroup until the desired number of participants is reached for each category.
Steps:
-
Identify Key Characteristics (Subgroups) The researcher identifies key characteristics relevant to the study (e.g., age, gender, occupation, etc.) and divides the population into subgroups (quotas).
-
Determine the Quota for Each Subgroup The researcher determines how many participants are needed from each subgroup to match the proportion of the characteristics in the broader population. For example, if 60% of the population is female, the sample should also include 60% females.
-
Select Participants within Subgroups Once the quotas are set, the researcher selects participants from each subgroup (often in a non-random way, depending on availability or convenience) until the required number of participants for each group is reached.
-
Stop When Quotas Are Met The process continues until the desired number of participants for each subgroup is reached.
About snowball sampling
Researcher takes a sampling unit, and selects other sampling unit based on/from the current sampling unit.
This sampling method is used when the sampling unit is hidden, and when other sampling units can only be found from the current sampling unit.
For example, taking sample of someone who contracts AIDS, and finding other sampling units by asking the current sampling unit for other people who conctracts AIDS too.
This sampling method may be very biased, because the researcher may only contact sampling units from a specific group.
Reference
- Pasted image 20250319163216.png
- Pasted image 20250319163241.png
- Pasted image 20250319163458.png
- Pasted image 20250319163744.png
- Pasted image 20250319163837.png
- Pasted image 20250319164130.png
- Pasted image 20250319164450.png
- Pasted image 20250319164721.png
- Pasted image 20250319164854.png
- Pasted image 20250319165234.png
- Pasted image 20250319165754.png
