Probability sampling includes any sampling method, which uses random selection forms. In random selection, a process must be set up to give each unit in the population an equal chance or probability to be selected. Non-probability sampling, on the other hand, is a technique of sampling that gathers samples in a process where each unit in the population is not given an equal chance of selection. It does not utilize random sampling methods. Its subjects are normally selected based on the researcher’s personal judgment or their accessibility.
Examples of probability sampling methods include:
- Simple random sampling: every unit in the population has an equal chance of selection. For example, if a study involves a population of 1,000 units, and a sample of 50 units is required, the sample would be selected randomly, probably using random number tables.
- Systematic random sampling: it is similar to simple random sampling. However, the samples are selected in an orderly manner, e.g., every 10th unit.
- Stratified random sampling: the units in the population are classified into groups e.g., boys versus girls. The units within each group have an equal chance of being included in the sample
- Cluster random sampling: the population is divided into clusters and then units from each cluster are selected through simple or systematic random sampling.
- Multi-stage sampling: it is a form of cluster sampling and involves the study of a large population. The first stage is cluster construction followed by selection of units within each cluster.
Examples of non-probability sampling methods include:
- Convenience sampling: also known as opportunity, accidental or haphazard sampling. The samples are chosen due to their accessibility to the researcher.
- Consecutive sampling: similar to convenience sampling. However, all the accessible subjects are included as part of the sample.
- Purposive sampling: this method is used to study a particular group of people. Those who do not fit the particular profile required are not included in the sample.
- Modal instance sampling: a form of purposive sampling. It is used to study the thoughts of a particular set of people and the researcher fears that data from this group may get lost in general study.
- Expert sampling: used when the study needs the assessment of highly skilled or knowledgeable people.
Research is conducted by experts to explore, describe, explain and draw conclusions on the topic of study. Exploration involves the desire of the researcher to become familiar with the topic. It enables the researcher to achieve better understanding of the topic and develop methods for subsequent studies. Research is also done to describe certain events. Observations are done from which descriptions are made. An example is the calculation of crime rates in different areas. Research studies are conducted to explain certain occurrences. Explanatory studies answer the question ‘why?’ For example, identifying the variables, which are responsible for high rates of crime.
In research, a unit of entity is the major entity analyzed in a study. They are individuals, artifacts, groups, geographical units and social interactions. Individual human beings are commonly used in social research. Individual characteristics and behavior are described and analyzed. Examples of individuals may be college students, factory workers or single parents. A social group can also be studied in a social research. A group is considered as a single entity by the researcher and share similar characteristics. Examples are families and criminal gangs.
Longitudinal studies involve the repeated collection of data from the same sample and over a long period of time. In cross-sectional studies, data is collected only on one occasion and at a particular point in time. Cross-sectional studies are subject to bias and errors
Social Research Methods. (2008, October 20). Research Methods Knowledge Base. Retrieved from Social Research Methods: http://www.socialresearchmethods.net/kb/sampprob.php