Sampling Method Description Advantages Disadvantages

1) Simple Random Each member of the population has an equal ☺ Easy to perform  Difficulty in identifying all population members
Sampling chance of being selected, usually through ☺ High probability of a representative sample  Challenging to contact selected sample
random draws or software. ☺ Complies with many statistical procedures members

2) Stratified Random The population is divided into subgroups ☺ Enhanced representation of subgroups  Complexity in defining strata
Sampling (strata) based on specific characteristics. A ☺ Flexible for proportional or disproportionate  Resource-intensive
random sample is taken from each stratum. samples  Challenges with small sample sizes
☺ Improved subgroup analysis
3) Systematic Sampling Researchers select every nth individual from a ☺ Simplicity and ease of use  Potential periodic patterns in the population
list or sequence after randomly choosing a ☺ Less time-consuming than random sampling affecting results
starting point.

4) Cluster Sampling The population is divided into clusters, and a ☺ Suitable for large and dispersed  Potential cluster-level biases
random sample of clusters is selected. All populations  Reduced precision compared to simple random
individuals within selected clusters are ☺ Cost-effective and time-saving sampling
included.

5) Convenience Researchers select participants based on their ☺ Easy and quick to implement  Likely to introduce selection bias
Sampling convenience and accessibility, rather than ☺ Suitable for pilot studies or preliminary  Limited representativeness and generalizability
random selection. research

6) Snowball Sampling Initial participants refer additional ☺ Useful for hard-to-reach or hidden  Non-random and may lead to biased samples
participants, creating a chain or "snowball" populations  Difficulty in estimating population parameters
effect. ☺ Cost-effective for certain research contexts

7) Purposive Sampling Researchers intentionally select specific ☺ Suitable for specialized research questions  Limited generalizability and potential bias
individuals or groups based on their expertise, ☺ Efficient for studying unique populations  Not suitable for estimating population parameters
characteristics, or relevance to the study.
Aspect Quota Sampling Stratified Sampling
1. Probability Non-probability/non-random sampling method. Probability/random sampling method.

2. Division of Population is divided into subgroups (quotas) based on Population is divided into homogeneous and mutually exclusive
Population specific characteristics or demographics. strata based on specific characteristics.

3. Selection Individuals within each quota are selected non- Random sampling is performed independently within each stratum.
Within Groups randomly to meet predetermined quotas.

4. Representation May not be as statistically representative of the Ensures that each stratum is adequately represented in the sample,
population because of non-random selection within leading to more statistically reliable results.
quotas.

5. Bias Prone to selection bias if not carefully managed. Less susceptible to selection bias compared to quota sampling.

6. Generalizability Limited generalisability to the entire population due to Generally allows for more accurate generalisation to the entire
potential bias. population.

7. Precision Typically less precise and may have wider confidence Typically more precise and narrower confidence intervals.
intervals.

8. Use Cases Commonly used when random sampling is difficult or not Used when it's important to ensure that specific subgroups are well-
feasible, such as in convenience sampling. represented in the sample.

9. Statistical Requires more advanced statistical techniques to Requires less correction and adjustment in statistical analysis.
Analysis correct for bias and improve reliability.
 Aspect
Sampling Errors Sampling Bias
1. Definition Errors that occur due to random fluctuations in the sample selection Systematic errors that occur when the sample is not representative of
process. the population.

2. Nature Unpredictable and random in nature. Predictable and systematic in nature.

3. Types - Random Sampling Error: Occurs by chance and can vary from - Selection Bias: Occurs when the method of selecting the sample
one sample to another. favours certain groups or excludes others, leading to an
unrepresentative sample.
- Non-Response Error: Occurs when some sampled individuals do
not respond or participate in the survey. - Measurement Bias: Occurs when the measurement tools or techniques
used in data collection are systematically flawed or biased.
4. Causes Due to the inherent variability in random sampling. Due to flaws or issues in the sampling process itself, leading to a
systematic distortion of results.

5. Impact on Can lead to fluctuations in sample estimates, but these errors tend Can significantly skew results in one direction, leading to incorrect or
Results to cancel out when aggregating data from multiple random misleading conclusions.
samples.
6. Manageability Sampling errors can be reduced by increasing sample size, but they Sampling bias can be reduced through careful design of the sampling
can never be completely eliminated. method and data collection process, but it requires deliberate efforts to
minimize.
7. Example If you randomly select 100 people from a population and calculate If you only survey individuals who attend a health clinic and do not
the average income, you might get a result that is slightly different include people who do not seek medical care, your data may be
from the population's true average due to random sampling error. biased toward individuals with health issues, leading to selection bias.
8. Variability in Sampling errors lead to variability in estimates, but the variability Sampling bias can systematically push estimates in one direction,
Estimates tends to follow a known statistical distribution. consistently overestimating or underestimating certain characteristics.

9. Addressing Addressed through statistical techniques such as confidence Addressed through careful sample design, randomization, and
Concerns intervals to quantify uncertainty. minimizing sources of bias during data collection.