Sampling Techniques


Techniques of Collecting Data: || Stages in Sampling Techniques:



Introduction

In order to answer the study questions, it is doubtful that the researcher will be able to collect data from all circumstances. As a result, you'll need to pick a sample. Regardless of the method utilized, it is vital that the people chosen are representative of the overall population if a sample is to be used. This may imply concentrating on difficult-to-reach demographics.

Sampling can be used to make generalizations about a population or to make inferences about a theory. In essence, it is determined by the sample technique used.

In general, there are two types of sampling techniques:

  • Probability Sampling
  • Non-probability Sampling

  • What are your thoughts on this topic?


    Probability Sampling

    Probability sampling is a sampling approach in which a researcher chooses members of a population at random based on a set of criteria. With this selection criteria, all members have an equal chance of being included in the sample. A researcher could use a random number generating computer program to select a sample from the sampling frame after first constructing a sampling frame (Zikmund, 2002).

    Here are the types of probability sampling:

  • Simple Random Sampling-One of the most effective probability sampling approaches for reducing time and resource use. It's a trustworthy method of gathering data in which each and every member of a population is chosen at random and solely by chance. Each person has the same chance of being chosen to participate in a sample. Simple random sampling, like all probability sampling methods, allows the sampling error to be calculated and reduces selection bias. It has the distinct advantage of being the simplest method of probability sampling. One problem of simple random sampling is that you may not be able to find enough people who have your desired feature, especially if it is uncommon.
  • You can use random number generators or other techniques that are totally based on chance to accomplish this type of sampling.

    In this blog, I advocate using Python to produce appropriate sample/elements from the entire population. For example, If 10 samples are chosen from a population of 200 in the sampling frame, then type the following command:

    import random

    population = 200

    sampledata = range (population)

    print (random.sample(sampledata,10)

    [73, 94, 64, 79, 92, 26, 130, 97, 148, 152] - arbitrary outputs.

  • Stratified Random Sampling - This strategy divides the population into smaller groups that do not overlap but still reflect the full population. These groups can be sorted before sampling, and then a sample drawn from each category independently.
  • Example:

    There are 600 female employees and 300 male employees at the company. Assume you need to hire 150 people from a population of which 90 are women and 60 are men. You select the population into two strata based on gender to ensure that the sample reflects the company's gender balance. Then you select 90 women and 60 men at random from each group, giving you a representative sample of 150 employees.

  • Systematic Random Sampling - The sampling approach is used by researchers to select sample members of a community at regular periods. It necessitates the choice of a sample starting point and sample size that can be repeated at regular intervals. Because this type of sampling method has a predetermined range, it takes the least amount of time.
  • For a sample of size n, we divide our population of size N into k subgroups. The first of our k elements is picked at random from the first subgroup.

    Example:

    In a population of 4000 people, a researcher plans to take a systematic sample of 400 persons. He or she will number each individual in the population from 1-4000 and select every 10th person to be included in the sample (Total population/Sample Size = 4000/400 = 10).

  • Cluster Random Sampling - When the entire area of the research is too large, the researcher should divide it into smaller parts of the same or equal size, and then choose at random from the smaller units. It is predicted that the overall population will be divided into a smaller number of units, still from clusters of smaller units, and that some of these cluster units will be randomly picked for inclusion in the general sampling. One of the benefits of cluster sampling is that it is cost effective because it focuses on the specified clusters; nevertheless, it provides less precision than plain random sampling.
  • Example:

    Consider a scenario in which a company wants to assess the performance of cellphones across the Philippines. They can divide the population of the entire country into cities (clusters), select the most populous towns, and filter those who use mobile devices.

  • Multi-stage Random Sampling - Typically, a multi-stage sample approach is used in large geographical area inquiries for the entire country. Multistage sampling entails combining several probability sampling methods in the most effective and efficient way possible. The population is classified into clusters, which are then segmented and grouped into a variety of subgroups (strata) based on their similarities. One or more clusters can be picked at random for each stratum. This procedure is done until the cluster cannot be separated any longer.
  • Example:

    A country can be divided into regions, provinces, urban, and rural areas, with all regions with similar characteristics forming a stratum.

    Non-probability Sampling

    The non-probability approach is a sampling technique that involves gathering feedback based on a researcher's or statistician's sample selection abilities rather than a pre-determined selection process. In most cases, the results of a survey done with a non-probable sample are skewed and may not accurately reflect the intended population. However, there are times when non-probability sampling is far more valuable than the other type, such as during the basic stages of study or while performing research on a budget.

    The following are a list of non-probability sampling:

  • Convenience sampling - Because of the simplicity with which the researcher can conduct it and contact the subjects, it is commonly referred to as convenience sampling. Researchers have almost no power over the sample items they choose, and they do so only on the basis of proximity rather than representativeness. When gathering feedback, this non-probability sampling method is employed because of time and expense constraints. Convenience sampling is utilized in instances where resources are limited, such as the early phases of research.
  • Example:

    You're conducting research into student support services at your university, so you invite your classmates to fill a survey on the subject after each of your classes. This is a convenient way to collect data, but the sample is not representative of all students at your university because you only questioned students who were taking the same classes as you at the same level.

  • Judgmental or purposive sampling - The sampling design is based on the researcher's assessment of who will provide the most useful information for the study's goals. The individual performing the research should concentrate on persons who share the same viewpoint in order to obtain the necessary knowledge and be willing to share it.
  • It's frequently employed in qualitative research, especially when the researcher wishes to learn more about a certain phenomenon rather than making statistical inferences, or when the population is small and specific. In order to be effective, a purposive sample must have explicit inclusion criteria and justification.

    Example:

    You want to learn more about handicapped students' perspectives and experiences at your university, so you purposely choose a group of students with varying assistance requirements in order to collect a diverse set of data on their interactions with student services.

  • Quota Sampling - This sampling technique involves the selection of members based on a pre-determined standard. Because a sample is established based on certain criteria in this situation, the created sample will have the same characteristics as the entire population. It is a quick way to collect samples.
  • Example:

    If we want to investigate a population with 40% females and 60% males, we should divide the population into two groups. We'll require a sample size of 100, and the selection won't stop until the target is reached.

  • Chain Referral or Snowball Sampling - When it's tough to track down subjects, researchers use this sampling strategy. Is a selected design procedure that is typically carried out through the use of networks. It's handy when the researcher doesn't know much about the group or organization he wants to study; contact with a few people will lead him to another group. The study sample selection will be beneficial in terms of communication, decision-making, and knowledge dissemination to people.
  • Example:

    Let's pretend there's a poll about covid patients. If we continue to question people about their covid optimism, we can expect that the majority of them will remain silent. The majority of them will be unable to discuss it freely. We contact their family, volunteers, doctors, or anybody else who can assist us gather information to find out the actual figures.

    End Note:

    We studied about several sampling strategies and how they are utilized in this blog. Finally, we must remember that sample strategies should be applied according to the case at hand; in this case, we must employ the appropriate sampling approaches.

    Thank's for Reading



    "I'd love to hear your ideas on the types of sampling techniques in the comment section below."


    References:

  • https://www.questionpro.com/blog/types-of-sampling-for-social-research/
  • https://towardsdatascience.com/sampling-techniques-a4e34111d808
  • https://hal.archives-ouvertes.fr/hal-02546796/document
  • https://headachejournal.onlinelibrary.wiley.com/doi/abs/10.1111/head.13707
  • https://www.qualtrics.com/experience-management/research/sampling-methods/
  • https://uca.edu/psychology/files/2013/08/Ch7-Sampling-Techniques.pdf

  • Is this article useful to you?



    Techniques of Collecting Data: | | Stages in Sampling Techniques:

    Share this blog:

    Post a Comment

    1 Comments

    1. Thank you for reading! Any thoughts about sampling techniques. I'd love to hear your comment.

      ReplyDelete