Sampling methods and its types DR SURYA.pptx

This PowerPoint presentation (PPT) on Sampling Methods in Research provides a detailed overview of the different sampling techniques used in clinical research, clinical trials, epidemiology, and statistical studies. The presentation explains the types of sampling methods, their applications, advantages, disadvantages, and selection criteria in research studies. It covers both Probability Sampling (Simple Random, Stratified, Systematic, Cluster) and Non-Probability Sampling (Convenience, Purposive, Snowball, Quota) methods. Additionally, the PPT highlights the importance of sample size calculation, minimizing bias, and ensuring data reliability in clinical trials or research studies.

This presentation is highly beneficial for medical students, research scholars, clinical trial professionals, data analysts, and public health researchers who are working on research methodology, data collection, and clinical trial designs

1. Sampling Methods Dr. Surya Junior Resident 3 rd year Department of Pharmacology and Therapeutics KGMU Lucknow E-mail: [email protected]

2. Contents Definition Types of sampling Methods used in Probability sampling Methods used in Non - Probability sampling Summary References

3. Sampling Procedure by which some members of the population are selected as representatives of the entire population The sample needs to be representative of the population in terms of time, Place, Person Sampling frame : List of all sampling units in the population Sampling scheme : Methods used to select sampling units from the sampling frame

4. PROBABILITY SAMPLING NON-PROBABILITY SAMPLING DEFINITION Subjects of the population get equal opportunity to be selected as a representative sample Subjects of the population not get equal opportunity to be selected as a representative sample SELECTION Random sampling Non randomly ( Arbitrarily ) RESULTS Unbiases ( less sampling error ) Bias( more sampling error ) REQUIREMENT RESOURCES More in terms of time, cost, efforts less

5. Convenient sampling Convenient sampling is a matter of taking what you can get Accidental sample, is considered easiest, cheapest and less time consuming Principle : Provides a quick and easy way to gather data Example : To collect feedback from Junior Residents (JRs) who participated in the ROLEPLAY IN ECOPHARMACOVIGILANCE

6. Advantages of Convenient sampling : Easy and quick to administer Useful for preliminary or exploratory research Disadvantages of Convenient sampling : Highly prone to bias

7. Subjective / Judgmental Sampling Definition : Samples are selected based on researcher knowledge rather than specific criteria Example : Selecting Junior Residents based on recommendations from Senior Residents who are familiar with them Advantages: Easy to implement Disadvantages: Highly prone to researcher bias

8. Purposive Sampling : Focuses on selecting participants that specifically meet the purpose of the study criteria Example : Selecting JRs who have published research papers in the past year for a study on research Advantages : Cost-effective Disadvantages : Bias , limited scope ( small sample size)

9. Quota sampling Researcher ensures that specific characteristics (such as age, gender, education level, etc.) of the population are represented proportionally within the sample Dividing the population into subgroups (quotas) and then selecting a specific number of individuals from each subgroup Advantages: Ensures representation of specific characteristics Disadvantages: Potential for bias

10. Example of quota sampling : In a study on the effects of a new drug, the researcher ensures that the sample includes a specific proportion of male and female participants, as well as different age groups (18 to 35 yrs & 36 to 59 yrs), based on their proportions in the general population

11. Snowball sampling Used to identify hidden/ hard to reach populations in the absence of a sampling frame Identifies an initial participant based on certain characteristics of interest Initial participants then recruit other participants from their same criteria This process continues, expanding like a snowball rolling downhill

12. To be contd., Example : Investigating experiences of patients with a rare genetic disorder. Begin with a few patients found through clinics who refer others with the same condition Advantages: Useful for studying hard-to-reach or hidden populations Disadvantages: Prone to bias as it depends on initial subjects to recruit others

13. Simple random sampling Equal chance of selection for each sampling unit from target population Number all units (Randomly) Advantages: Unbiased and representative, if sample size is large enough Disadvantages : Time-consuming, requires complete sampling frame Example : If you have a list of all JR1s (26 names), you randomly pick numbers such as 9, 18, and 22. These correspond to the names selected as your sample

14. Systematic sampling A unit drawn every k th unit, to obtain a representative sample Draw a random number for starting , draw every k units from first unit Advantages: Less time consuming & Cost-Effective Disadvantages: Requires a Complete list

15. Example of systemic sampling: If researcher have 100 students for sampling Determine the Sample Size : You want to sample 26 students Calculate the Sampling Interval : The sampling interval (k) is calculated by dividing the population size (100) by the sample size (26): k = 100/26 ≈ 3.85 ( 4 ) Select the Sample : Choose a random starting point between 1 and 4 Define the Population : You have 100 students.

16. Suppose the random starting point is 2, you would select every 4th student from the list starting at the 2nd student So, you would survey students 2, 6, 10, 14, 18, 22, 26, and so on, until you reach your desired sample size of 26 students

17. Stratified sampling Classify population into homogenous subgroups ( strata) The population is divided into subgroups (strata) based on certain characteristics, and samples are taken from each subgroup ,draw sample in each stratum Advantages: Ensures representation of all subgroups (strata) within the population Disadvantages: More complex and time-consuming

18. Example of stratified sampling : Junior Residents in the Pharmacology Department at KGMU Identify Strata : You have three distinct strata JR1: 26 residents , JR2: 26 residents, JR3: 15 residents = 67 residents Determine Sample Size : Decide how many residents you want to sample from the entire group For example , let's say you want to desired sample 30 residents in total(67)

19. Proportionate Sampling & Apply Proportions to Sample Size : Calculate the proportion of the total sample from each stratum Multiply these proportions by your desired sample size (30) - JR1: 26/67 ×30≈11.64 (12) - JR2: 26/67 ×30≈11.64 (12) - JR3: 15/67 ×30≈6.7 (7)

20. Random Sampling within Strata : Randomly select the calculated number of residents from each stratum: JR1: 12 residents JR2: 12 residents JR3: 7 residents Combine Samples : Combine the samples from all strata to form your final sample

21. Cluster sampling Cluster sampling involves dividing the population into clusters (groups) and then randomly selecting entire clusters to study, including all individuals within those clusters Advantages: Cost-effective and convenient, especially for large and dispersed populations Disadvantages: Higher chance of sampling error

22. Example of Cluster sampling : For Junior Residents (JRs) working in the non-clinical departments at KGMU 1) Identify Clusters : Divide all JRs into clusters based on their respective departments Clusters : Pharmacology, Pathology, Forensic Medicine, Anatomy, Biochemistry, Microbiology , Physiology 2) Randomly select a few clusters. For Example : Pharmacology, Forensic medicine 3) Sample within Clusters : Include all JRs from the selected departments in the study

23. Multistage sampling A type of sampling that involves selecting samples in multiple stages, often using different types of sampling methods at each stage ( Several chained samples, several statistical units) Example : First selecting states randomly, then districts, then medical colleges within those districts, and finally PG residents within those medical colleges

24. TO BE CONTD., Advantages: Cost-Effective, Manageable , Flexibility Disadvantages: Time-Consuming Requires expertise

25. Summary Good quality & quality assurance ensures validity and precision results Probability samples are the only one that allow use of statistics Probability: Reduces bias, generalizable Non-Probability: Easier, cost-effective USES : 1) Probability: Research, surveys 2) Non-Probability: Exploratory studies

26. QUESTIONS Difference between probability and non-probability sampling? How does multistage sampling differ from cluster sampling? Types of Non probability methods ? Define convenience sampling and discuss one potential drawback of using this method? Explain the concept of sampling error and how it can be minimized ?

27. References Indian Council of Medical Research – National Institute of Epidemiology. Basic Course in Biomedical Research Course Material. Chennai: ICMR-NIE, https://nptel.ac.in/courses/127/106/127106134/ Mehta, T. Analytical study designs . Basic Course in Biomedical Research Handbook. 1st edition 2021. Pg.180 - 197

28. THANK YOU