Purposive Sampling

SHARE THE ARTICLE ON

CONSECUTIVE SAMPLING1
Table of Contents

What is purposive sampling?

Purposive sampling refers to the researcher’s own judgment over the selection of a sample needed to experiment. The researcher will depend on his/her own opinion and judgment when choosing who will participate in the study. Purposive sampling is also called judgment, selective or subjective sampling as the sample is drawn based on one person’s decision. 

Purposive sample is considered as a non-probability sampling method because the decision whether to include a person into the sample group is taken by the researcher itself. They believe that judging the population and drawing out samples is a good way to have a representative sample while saving time and money. Hence, purposive sampling needs personal judgment to select cases that will answer the research questions and achieve research goals. 

See Voxco survey platform in action

Take a tour of our platform with on-demand survey demos

How to conduct purposive sampling?

Depending on the cases, purposive sampling can be categorised as:

  • Typical case – it explains cases that are average and normal. 
  • Extreme or deviant case – it refers to deriving samples from cases that are perceived as unusual or rare.  
  • Critical case – it focuses on specific cases that are dramatic or very important. 
  • Heterogeneous or maximum variation sampling – it relies on researcher’s judgment to select participants with diverse characteristics. It makes sure that there is the presence of maximum variability within the primary data. 
  • Homogeneous sampling – it focuses on one particular subgroup in which all the sample members are similar, such as a particular profession or level in an organization’s hierarchy
  • Theoretical samplingis a special case of purposive sampling that is based on an inductive method of Grounded Theory. 

Depending on your use case, you can adopt to any of the above purposive sampling methods. You need to specify the qualifying conditions or criteria for your research sample. Later, you can go ahead and reject any variable that doesn’t satisfy these conditions. 

Voxco offers easy-to-use online survey software

Advantages of purposive sampling

  • Qualitative research designs 

When focusing on purposive sampling, researchers have access to a wide range of qualitative research designs. To acquire the necessary data to make a conclusion, these designs frequently demand a different type of sampling method and technique. The numerous strategies made feasible by the purposive approach allow research designs to be more adaptable, allowing for the use of specific methodologies as needed to achieve the desired result. 

  • Creating generations from data 

Although it is impossible to extrapolate information from a chosen group to draw broad generalizations about the entire population, the various purposive sampling procedures do allow researchers to justify making broad generalizations based on their sample. To be valid, these attempts must be logical, analytic, or theoretical in nature. Each of the seven strategies takes a somewhat different approach to this procedure, so it’s up to the project’s researchers to figure out how things should go. 

  • Multiple phases  

Purposive sampling might include not just several phases for researchers, but each phase can also be built on the previous one. Even while this normally necessitates a different technique at the start of each phase, this process is beneficial since it provides a researcher with a larger pool of non-probability sampling options from which to choose. A classic illustration of this benefit is that a critical sample can be beneficial in establishing the value of a study, whereas expert sampling allows for a more in-depth review of the data available.

Browse through all that Voxco surveys have to offer!

  • Saving time and money in data collection 

Purposive sampling’s versatility allows researchers to save time and money while gathering data. It provides an adaptive process as circumstances change, even if they do so unexpectedly. You may cater to a variety of requirements and interests while still preserving a single focal point. As a result, it is possible to generate a logical final result that is representative of a certain population. You’re using a non-random strategy to get outcomes that can subsequently be used to help you make better decisions in the future. 

  • Demographics to obtain data points 

When researchers employ the homogeneous purposive sampling strategy to gather data, they are choosing people who share a set of traits. This resemblance could be based on emotional reactions, physical traits, or even household income levels. When researchers sought to discover how white people felt about the concepts of white privilege and racism, they asked white individuals. You might use the same procedures to research people who identify with a given gender, work for the same company, or have any other shared attribute. 

  • Maximum variation 

When researchers employ the homogeneous purposive sampling strategy to gather data, they are choosing people who share a set of traits. This resemblance could be based on emotional reactions, physical traits, or even household income levels. When researchers sought to discover how white people felt about the concepts of white privilege and racism, they asked white individuals. You might use the same procedures to research people who identify with a given gender, work for the same company, or have any other shared attribute. 

  • Including various extremes of population  

Purposive sampling can be used to look at averages, but it can also be used to identify the extreme viewpoints that exist in each population group. In every endeavor like this, there are always outliers to consider, and their perspectives might be just as important as what the “middle” person contributes to the end result. This benefit allows for a better knowledge of behavior patterns within a certain group, and it does not always have to be a negative one. 

Researchers may purposefully choose all of the individuals who achieve the highest levels of achievement while disregarding everyone else if they wanted to see why a specific group of students always received good grades while others did not.

  • Low margin of error 

When researchers conduct a random survey of a population group, the margin of error on their conclusions might be large. Take a look at the political polls that news media broadcast on a regular basis. The majority of them have a margin of error ranging from 3% to 6%, and in some cases even more. If your statistics show that those who answer “yes” make up 48% of the population, while people who say “no” make up 52%, the margin of error can contradict whatever conclusion you were hoping for. 

Purposive sampling allows researchers to obtain a narrower margin of error because the data they collect comes directly from the source. Each individual has distinguishing qualities that identify them in the same category.

Disadvantages of purposive sampling

  • Provides invalid inferential statistical procedures 

When you employ purposive sampling to obtain data, you’ll see that this structure contains a large number of inferential statistical processes. These figures are no longer valid. They allow you to extrapolate from small samples to a broader population, allowing you to make claims regarding the validity or accuracy of your findings. The only inference possibilities apply to the specific group that you are researching since the data is more complex than what you would get from a random sample. 

  • Prone to researcher bias  

Purposive sampling, regardless of the method employed to collect data, is very susceptible to research bias. The idea of creating a sample in the first place is based on the researcher’s opinion and personal interpretation of the facts. This difficulty becomes a substantial disadvantage when the judgements are either poorly considered or ill-conceived, and it can create bottlenecks in the path of a final solution. This problem is lessened when elicitation, agreed criteria, or a theoretical framework are in place. 

  • Challenging for representative nature of samples 

Researchers must show that the judgment used to choose the various units or individuals for purposive sampling was appropriate for the process. In practically every circumstance, the high degrees of subjectivity cast an unavoidable cloud on the outcomes. Even when theoretical, logical, or analytical structures are present, readers will always be skeptical of the generalizations reached until there is a method to defend the overall representative structures that were employed to obtain findings.

Voxco’s Visual Analytics Dashboards provide insights that aid decision-making

  • Data manipulation by participants 

When people are informed that they have been chosen for a research project, their conduct may change. They may choose to act in a way that helps researchers to reach the findings they want to see, or they may choose to act in a way that prevents researchers from reaching the conclusions they want to see. Some participants may opt to lie in order to achieve an unfavorable result because they have a personal prejudice that they want to make known. Only the researchers’ competence can establish whether the data acquired is valid, which indicates that the outcome being researched may be more unpredictable than expected at times. 

  • No way to evaluate reliability 

There are some exceptions to this disadvantage, but there is usually no way to assess the authority or the purposive sample experts’ credibility. As a result, determining whether or not there is a sampling error in the data presented by researchers might be nearly difficult. Even when the most knowledgeable members of the industry under investigation submit the data, the interpretation of the findings can be questioned. As a result, there are situations when purposive sampling is the least desirable alternative. 

  • Purposive sampling can still produce inaccurate results 

Purposive sampling does not use randomization when analyzing the overall sample process because that would defeat the goal of the method in the first place. When researchers want to learn about twenty-something entrepreneurs managing the gig economy, they shouldn’t chat with 40-year veterans of labor. This information will always have a bias. Because members of the population under study do not always have equal chances of being chosen, even the most logical sampling technique can produce erroneous results. 

Although the margin of error is narrower than it would be in a randomized procedure, it still exists. If researchers utilize a large enough sample for their research, it may also be greater than a random sample.

Explore all the survey question types
possible on Voxco

Read more

Hindol Basu 
GM, Voxco Intelligence

Webinar

How to Derive the ROI of a Customer Churn Model

30th November
11:00 AM ET