Grounds for Probability and Non-Probability Sampling: UU-PhD-801

This report delves into the core concepts of probability and non-probability sampling methods, crucial for effective research design. It explores the fundamental differences between the two approaches, highlighting their respective strengths and weaknesses. Probability sampling, including simple random, stratified, and cluster sampling, is presented as a method that offers a representative sample of the entire population. It also discusses the importance of randomization to avoid bias and the application of these methods in quantitative research. In contrast, the report examines non-probability sampling, such as purposive and convenience sampling, and their suitability for qualitative research. The report further provides real-world examples to illustrate the application of each sampling method, emphasizing the importance of selecting the appropriate technique based on research objectives and the nature of the target population. The report concludes by highlighting the significance of sampling in ensuring research validity and reliability and stresses the importance of avoiding sampling bias.

Running head: Probability and Non-Probability Sampling 1
Grounds for Probability and Non-Probability sampling
Course code:
Student name:
Tutor:

Probability and Non-Probability Sampling 2
Introduction
Sampling is the art of selecting a part of a population to represent the entire population in
a research. The process of sampling is an important aspect of research as it lays a foundation of
how the entire data collection process and results will be. A wrong sampling process may be
detrimental to the results obtained from a research study. Wrong sampling may lead to
conclusion of results which might not be a representative of the whole population hence
misleading the purpose of research. For example, a research topic on the mean amount of money
an international student spends on accommodation will require a researcher to have his target
population to be students in an institution of learning. Conducting this research with a sample
randomly selected from a city center would not yield proper and valid results.
To add on, the outcome of the research will also depend on the sampling method used.
However, the sampling method will depend on the nature of you target population and the
approach of the specific research. Depending on the approach, a study can either employ
probability sampling or non-probability sampling.
Probability sampling
Probability sampling gives the researcher the opportunity to select a sample that
accurately represents the interest of the whole population.
Probability sampling such as simple random sampling which is also known as equal
probability sampling is employed in a research study where the researcher’s or the investigator’s
interest is not from influencing the choice of a sample (Weyer, 2014). In this method, every
member of the population should have an equal chance of being selected into the sample. In this

Probability and Non-Probability Sampling 3
method every member is selected. This therefore eliminates biasness that might arise due to a
researcher choosing a sample just because he or she is interested in the sample.
Probability sampling is also used when the samples need to be taken from different parts
of the targeted population. This need then requires the research to use stratified sampling
procedure to obtain the sample (Schreuder, Gregoire, & Wood., 2001). For example if a research
need to be conducted on an entire county, for the research not to be bias and select people from
one area, the province is stratified or divided into districts (strata). It is from these strata that the
research will go backwards and now apply random sampling to select a given number from each
stratum (district). The idea behind stratification is that the geographical area is large yet the
sample has to be picked from every part of that area.
When the size of the population is too large, probability sampling method that comes in
handy is the cluster sampling. It applies where the population is too large for simple random
sampling to be done (Duncan, 2018). For example, when a researcher wants to conduct a
research about feeding patterns in Africa, the continent may be divided into for example
countries. A given number of countries are then chosen using simple random sampling where
each country has got an equal opportunity of being selected. It is now from the selected countries
that samples are collected. The advantage of cluster sampling is that it is a very easy method to
use, however, its disadvantage comes in when the collected sample is not homogeneous. This
will result into inaccuracies of the data.
In large and more advanced researches, there is always need to combine two or more
probability sampling methods just to ensure randomization which leads to valid results. This

Probability and Non-Probability Sampling 4
combination of different sampling methods and using them at different stages of a research is
called multi-stage sampling.
Non-probability sampling
Just like in probability sampling, there are cases in a research where non-probability
sampling methods need to be used. Purposive sampling is one of these methods. When a
researcher needs to choose a sample according to the interest of his research, he will employ
purposive sampling method (Kalton, 2013). For example if a researcher wanted to conduct a
study about hypertension among the obese people between the age of 18 and 40, he or she will
use his judgment to identify the obese people who are of the specified age to select into the
sample for the study (Tansey, 2013).
The other non-probability sampling that comes in handy is convenient sampling.
Convenient sampling is applied in a situation where the sample to be selected is available in
terms of proximity or easiness of getting the sample or any other convenience (Brick, 2014).
Non-probability sampling methods are always appropriate for qualitative researches where
interviews are the main methods of collecting the data from the respondents.

Probability and Non-Probability Sampling 5
References
Brick, J. (2014). “The Future Of Survey Sampling.” Public Opinion Quarterly" (Vol. 5).
Duncan, G. (2018). “When to Promote and When to Avoid, a Population Perspective.”
Demography (Vol. 4 ).
Kalton, G. (2013). “Practical Methods for Sampling Rare and Mobile Populations.” Statistics in
Transition (Vol. 6).
Schreuder, H. T., Gregoire, G., & Wood., B. (2001). Sampling methods for multiresource forest
inventory. New York: JohnWiley and Sons.
Tansey, O. (2013). Process tracing and elite interviewing: PS: Political Science & Politics,
40(04), (Vol. 4).
Weyer, J. P. (2014). For what applications can probability and nonprobability sampling be
used? Environmental Monitoring and Assessment (Vol. 3).