In many cases, the nature of a study or the resources available require mixing several types of sampling, which is also known as mixed sampling .
In this article we will introduce you to what this method consists of, its characteristics and what are its advantages and disadvantages.
What is mixed sampling?
Mixed sampling consists of creatively combining well-established qualitative and quantitative techniques to answer the questions posed by the research design .
The strategies of this sampling involve the selection of units or cases for a study using both probability sampling (to increase external validity) and intentional sampling strategies (to increase transferability).
In a mixed-sample study, there are usually several samples, the size of which can vary depending on the line and the research question, from a small number of cases to a large number of units.
Sampling methods used in mixed sampling
Mixed sampling requires the use of different sampling methods, such as:
Probability sampling
Probability sampling techniques are used primarily in quantitative research and consist of randomly selecting a relatively large number of units from a population in which the probability of inclusion of each member of the population is determinable.
The goal of probability samples is to achieve representativeness, which is the degree to which the sample accurately represents the entire population.
Intentional sampling
Purposeful sampling techniques are used primarily in qualitative research and can be defined as the selection of units based on specific purposes associated with answering questions in a research study.
It is a type of sampling in which specific environments, people or events are deliberately selected for the important information they can provide and that cannot be obtained directly through other options.
Mixed sampling strategies
Mixed sampling strategies can employ all probabilistic and intentional techniques. In fact, the researcher’s ability to creatively combine these techniques to answer study questions is one of its defining characteristics.
The researcher sometimes chooses procedures that focus on the generation of representative samples, especially when addressing a quantitative aspect of a study.
On the other hand, when a qualitative chapter of a study is approached, the researcher usually uses sampling techniques that generate cases rich in information.
The combination of both orientations allows the mixed sampling researcher to generate complementary databases that include information that has both depth and breadth in relation to the phenomenon studied.
Characteristics of mixed sampling techniques
Some of the elements that characterize the mixed sampling technique are:
- It is designed to generate a sample that addresses the research questions.
- Includes the use of multiple research samples. The samples vary in size depending on the line of investigation and the question.
- They focus on the depth and breadth of information that can be collected for the study.
- Most sampling decisions are made before the study begins, but quality-oriented questions can cause them to arise from other samples during the study.
- Sampling decisions focus on expert judgment, especially since they are interrelated. Some aspects oriented to quantitative studies may require the application of mathematical sampling formulas.
- Both formal and informal settings are used.
- Both numerical and narrative data are normally generated.
- Occasionally, mixed methods sampling strategies can produce only narrative or only numerical data.
Mixed sampling types
There are various types of mixed sampling that you can implement in your research, among which the following stand out:
Basic mixed sampling
A well-known basic mixed sampling strategy is stratified intentional sampling, also known as quota sampling .
The stratified character of this procedure is characteristic of probability sampling, while the small number of cases that is usually generated through it is characteristic of intentional sampling.
In this technique, the researcher first divides the group of interests into strata (for example, students above the mean, by the mean, below the mean) and then selects a small number of cases for intensive study within each stratum. relying on intentional sampling techniques.
This allows the researcher to discover and describe in detail the characteristics that are similar or different in the strata or subgroups.
Sequential mixed sampling
There are examples of quantitative-qualitative and qualitative-quantitative sampling procedures in all the social and behavioral sciences.
Normally, the methodology and results of the first stream inform the methodology used in the second stream. In many of these cases, the final sample used in the QUAN slope is used as the sampling frame for the subsequent QUAL slope.
Concurrent mixed sampling
Concurrent mobility management designs allow researchers to triangulate the results of the quantitative and qualitative components of their research, allowing them to confirm, validate, or corroborate the results of a single study.
Multilevel mixed sampling
Multi-level mobility management sampling strategies are very common in research examining organizations in which different units of analysis are nested within each other.
In studies of these nested organizations, researchers are often interested in answering questions related to two or more levels or units of analysis.
Advantages
Here are some of the benefits of mixing different types of sampling :
- Greater sample richness: By mixing different sampling methods, approaches are improved.
- Greater fidelity of the instrument. It helps to certify that the methods used are adequate and useful.
- Greater integrity in the treatment or intervention : Ensures the validity and reliability of the research .
- Better perspective of the data: It allows to consolidate interpretations and the usefulness of the results obtained.
conclusion
As you have seen, studies show that various types of sampling can be combined to adapt them to each situation or project through mixed sampling that can be very useful to answer complex research questions that involve data of a qualitative and quantitative nature .
