sampling and its types.pptx presented by Preeti Kulshrestha M.Sc. nursing obs. and gync.

this ppt includes various terms regarding sampling , definition, purpose of sampling, characteristics of good sample, sampling process, factors influencing sampling process, list of types of sampling technique-probability sampling technique with examples = simple random sampling technique and its merits and demerits, stratified random sampling technique and its merits and demerits, systemic random sampling technique and its merits and demerits, cluster sampling and its merits and demerits, sequential sampling and its merits and demerits, nonprobability sampling technique with examples its uses = purposive sampling and its merits and demerits, convenience sampling and its merits and demerits, consecutive sampling and its merits and demerits, Quota sampling step and uses and its merits and demerits, snowball sampling and its types and its merits and demerits, sampling error and its reasons, types, how to minimize sampling error , problem of sampling.

1. SAMPLING PRESENTED BY- PREETI KULSHRESTHA

2. TERMINOLOGY POPULATION: - Population is the total of all the units in which a researcher is interested. E.g. Postgraduate nurses of India. TARGET POPULATION: - A target population is a total no. of people which meet the designated set of criteria ACCESSIBLE POPULATION: - people are also accessible as subjects for a study. E.g. A researcher is conducting a studying on the registered nurses working in Dayanand medical college and hospital Ludhiana. In this case, population for this study is all RN working in DMCH but some of them may be on leave and may not be accessible for research study. SAMPLING: - Sampling is the process of selecting a representative segment of the population under study.

3. TERMINOLOGY SAMPLE: - sample may be defined as representative unit of a target population. ELEMENT: - the individual entities that comprise the samples and population are known as elements. An element is also known as a subject in research. SAMPLING ERROR: - fluctuations in the values of the statistics of characteristics from one sample to another. SAMPLING PLAN: - Plan of a sampling method, a sample size, and the procedure. population target population accessible sample population subjects

4. SAMPLING DEFINITION Sampling is the process of selecting a representative unit of the target population under study.

5. CONTD. PURPOSES ECONOMICAL: - researcher can save lots of time, money, and resources. IMPROVED QUALITY OF DATA: - it is easier to maintain the quality of research work which would not be possible in case the entire population was involved. QUICK STUDY RESULT: - studying an entire population will take a lot of time . ACCURACY OF DATA: - It is always easy to establish better rapport and to collect more accurate data.

6. CONTD.

7. process

8. FACTORS INFUENCING SAMPLING PROCESS

9. CONTD.

10. CONTD.

11. PROBABILITY SAMPLING TECHNIQUE Stratified random sampling Sequential sampling Simple random sampling Systematic random sampling Cluster sampling Purposive sampling Snowball sampling Convenience sampling Quota sampling Consecutive sampling NON-PROBABILITY SAMPLING TECHNIQUE Types of sampling techniques

12. PROBABILITY SAMPLING TECHNIQUE It is based on the theory of probability. It involves random selection of the elements / members of the population . Every subject has equal chance to be selected as a sample. Chances of bias are relatively less.

13. Contd. SIMPLE RANDOM SAMPLING This is the most pure and basic probability sampling design. Every population member has a similar chance of being picked subject. Two essential prerequisites are- homogeneous, sampling frame. Methods: -lottery method (used for small size group) -use of table of random numbers (used for average size group) - use of computer (used for largest size group)

14. Contd. Merits – Most reliable and unbiased method Requires minimum knowledge of study population Free from sampling errors/bias

15. Contd. Demerits – Needs up to date complete list of all the members of the population Expensive Time consuming

16. Contd. STRATIFIED RANDOM SAMPLING This method is used for heterogenous population . Researcher divides the entire population into different homogenous subgroups or strata and then randomly selects the final subjects. Strata are divided according to selected traits such as age, gender, religion, socio-economic status, diagnosis, education, geographical region, nursing area specialization site of care etc. Categories : Proportionate stratified random sampling Disproportionate stratified random sampling

17. Contd. E.g.– Proportionate stratified random sampling Stratum A B C Population size 100 200 300 Sampling fraction ½ ½ ½ Final sample size 50 100 150

18. Contd. E.g.– Disproportionate stratified random sampling Stratum A B C Population size 100 200 300 Sampling fraction ½ ¼ 1/6 Final sample size 50 50 50

19. Contd. Merits – Ensures representative sample in heterogenous population Comparison is possible in two groups Save much time, money, and effort

20. Contd. Demerits – Requires complete and accurate information of population Large population must be available Possibility of faulty classification

21. Contd. SYSTEMIC RANDOM SAMPLING It involves the selection of every Kth case from a list of groups such as every 10 th person or every 100 th person. the first subject is chosen by the help of random number table. K=N/n Or K= Number of subjects in target population (N) Size of sample(n) E.g. a researcher wants to choose about 100 subjects from a total target population of 500 people. Therefore, K=N/n , 500/100 = 5. every 5 th person will be selected.

22. Contd. Merits – Simple to carry out Less time consuming Cheaper than simple random sampling Provide better representative sample

23. Contd. Demerits – If first subject is not randomly selected, then it becomes a nonrandom sampling technique. If sampling frame has nonrandomly distributed, this sampling technique may not be appropriate.

24. Contd. CLUSTER SAMPLING When simple random sampling is not possible due to size of the population, cluster sampling is carried out. E.g. entire population of Asia. This method is used in cases where the population elements are scattered over a wide area, and it is impossible to obtain a list of all the elements.

25. Contd. Merits – Cheap, quick and easy method for a large population Require only list of the members Same cluster can be used again for study

26. Contd. Demerits – Least representative of the population Possibility of high sampling error In Small homogenous population this technique is not at all useful.

27. Contd. SEQUENTIAL SAMPLING It is also called double sampling / multiple phase. This method is slightly different from other methods. The sample size is not fixed. Select small sample and tries out to make inferences, if not able to draw results, then adds more subjects until clear – cut inferences can be drawn.

28. Contd. E.g.– a researcher is studying association between smoking and lung cancer. no. of subjects smokers(A) non-smokers(B) having lung cancer (A) (B) 20 08 12 02 01 30 10 20 05 03 50 28 22 10 04

29. Contd. Merits – Study on best possible smallest sample Helping in finding the inferences of the study

30. Contd. Demerits – Time consuming Not possible to study a phenomenon, which needs to be studied one point of time Requires the repeated entry into the field to collect sample

31. NON-PROBABILITY SAMPLING TECHNIQUE Every subject does not have equal chance to be selected because elements are chosen by choice not by chance through nonrandom sampling methods. This technique can also be used in pilot study.

32. CONTD. USES : This type of sampling can be utilized in any particular trait. When random sampling is impossible When researcher has got limited budget, time, and workforce This technique can also be used in a pilot study.

33. Contd. PURPOSIVE SAMPLING It is also called ‘ judgmental ’ or ‘ authoritative sampling .’ Subjects are chosen with a specific purpose in mind. The researcher believes that some subjects are fit for research compared to other individuals. This is the reason why they are purposively chosen as subjects. Samples are chosen by choice not by chance, through a judgement based on his or her knowledge about the population.

34. Contd. Merits – Simple to draw and useful in explorative studies. Save resources, requires less fieldwork.

35. Contd. Demerits – Requires considerable knowledge about population Not always reliable sample Two important weakness – authority, sampling process

36. Contd. CONVENIENCE SAMPLING It is also called ‘ accidental sampling .’ It is based on convenient accessibility to the researcher. It is most common of all sampling techniques because it is fast, inexpensive, easy, and the subjects are readily available. May be used large population. E.g. a researcher wants to conduct study on the older people residing in Ludhiana

37. Contd. Merits – Easiest, cheapest, and least time consuming Help in saving time, money and resources.

38. Contd. Demerits – Sampling bias Findings can not be generalized Does not provide the representative sample from the population of the study.

39. Contd. CONSECUTIVE SAMPLING It is also called ‘ total enumerative sampling .’ It is very similar to convenience sampling. It can be considered as the best of all non-probability samples because it includes all accessible subject. May be used small size population. E.g. a researcher wants to conduct study the activity pattern of post-kidney-transplant patients who are admitted in post-transplant ward.

40. Contd. Merits – Not expensive Not time consuming Not intensive workforce or very less effort More representativeness of sample

41. Contd. Demerits – Can not be used to create conclusions

42. Contd. QUOTA SAMPLING Quota means no. of fixed proportion item from population. Researcher ensures equal representation of subjects It is based on the quota i.e. age, gender, education, race, religion, socio-economic status. This technique appears like stratified random sampling technique but in this technique, sample is selected without random process.

43. Contd. Contd. E.g. if the basis of quota is college level and the researcher needs equal representation, with a sample size of 100, select 25 first yr. students, another 25 second yr. students, 25 third yr. students, and 25 fourth yr. students.

44. Contd. Steps and uses: Dividing the population into subgroups Researcher should recognize the proportions of these subgroups in the population. Researcher choses subjects from various subgroups Sample is representative of the entire population 100 25 25 25 25 10 10 10 10

45. Contd. Merits – Economically cheap No need to approach all the candidates Suitable for studies like market and public opinion.

46. Contd. Demerits – Not sometime representative Bias is possible

47. Contd. SNOWBALL SAMPLING It is also known as chain referral sampling. This technique work like chain referral. It is used to identify potential subjects in studies where subjects are hard to locate, such as commercial sex workers, drug abusers etc,. It is used for very small subgroups and rare population. Researcher asks for assistance from the subject to identify other subject after observing the initial subject. initial subject

48. Contd. Types – LINEAR SNOWBALL SAMPLING - In this, each selected sample is asked to provide reference of only one similar subject.

49. Contd. EXPONENTIAL NONDISCRIMINATIVE SNOWBALL SAMPLING - In this, each sample member is asked to provide reference of at least two similar subject and large sample size can be achieved.

50. Contd. EXPONENTIAL DISCRIMINATIVE SNOWBALL SAMPLING - In this, one sample is asked for two references of similar subjects, out of which at least one subject must be active to provide further references and another could be non active in providing references.

51. Contd. Merits – Chain referral process Simple, cheap, and cost-efficient Require workforce

52. Contd. Demerits – Sample representative is not guaranteed. Little control of researcher over the sampling method. Chances of poor coverage of entire population.

53. SAMPLING ERROR DEFINITION Sampling error is the deviation of the selected sample from the true characteristics, traits, behaviors, qualities, or figures of the entire population.

54. CONTD. REASONS Error occurs because researchers draw different subjects from the same population, but the subjects have individual differences. Biased sampling procedure. Another causes of error is chance Two basic reasons are – chance error, - sampling bias

55. CONTD . Chance error : the error occurs by chance. E.g. a comparative study on malnutrition in under 5 years children in two cities A & B. unfortunately city B had a large no. of slum dwellers. So, it comprised a large no. of malnourished children thus skewing the result. Sampling bias : sampling bias is a tendency to favour a selection of sample units. E.g. a study is done on nursing student’s satisfaction with staying at a hostel or as a paying guest. This study is biased towards those students who are hostellers or paying guests, but it excluded the students who come from their own homes.

56. CONTD. TYPES SELF SELECTION BIAS : it happen when participants in the study have some kind of control over the study to participate or not. E.g. if a study is conducted on the number of people who can carry a load of 10 kg for 20 min, then only well- built people will have a preference to participate. EXCLUSION BIAS : it happens when some people of the group are eliminated from the study. HEALTHY USER BIAS : if the sample is selected from students who take the food from the mess and not the canteen where junk food is not served, thud there will be chances of healthy user bias.

57. CONTD. MINIMIZE SAMPLING ERROR Avoid convenience or judgmental sampling. Target population is well defined The sample frame should match

58. Problems of SAMPLING