Sampling Methods PDF

Learning Outcomes Explain the importance of doing sampling Identify the types of sampling...

Learning Outcomes Explain the importance of doing sampling Identify the types of sampling Understand sampling – probability and non-probability sampling Dr. Rohani Ismail, Phd Explain the procedure for each 09 May 2024 sampling technique Introduction Concept of sampling Number of subjects within the population After determining the sample size, a being studied may be so large cannot be researcher will decide on an accurately identified appropriate sampling method. It is difficult to obtain accurate measurements and calculations when it is Sampling method is a method where impossible to study the behaviour of every samples (participants or subjects) are subject within a population chosen from a population. Impossible to include everybody in a population-time, money, energy We do sampling-important to ensure the sample chosen will represent the population 4 being studied  Sampling breakdown Population is defined as the whole number of people or inhabitants in a country/region/ specific geographical area or a large group of individuals as a subject of study. Sample is a subset of the population which is selected to represent the population. Why sample? In research, most of the time, it is impossible for the researcher to include all individuals in the population, it will increase the cost, time, and energy to study all of them. Why we need a proper sampling method? Elements – are the selected respondents or The finding from a sample is generalised on objects from the defined target population the whole population. from which the researcher will study by That is why the researcher needs to apply the collecting the required data and information proper sampling method to obtain sample so (School teachers) that it is a representative of a population. Sampling unit – is a group of elements on There are many sampling method available to which observations is made (Schools) help researchers obtain sample from a Sampling frame – list of eligible sampling population. units and elements in the population from where the sample can be selected example: (List of schools, list of teachers in each school)  Target population Sampling method is divided into : The results (characteristics identified through the study on the sample) are generalised to its population I Research sample ↓ Sample (a small group of subjects) randomly selected from a specific population TYPES OF SAMPLING METHODS ☆ Schue , Non- Probability Probability sampling Sampling Simple random sampling Convenience sampling Probability Systematic sampling Quota sampling Sampling Stratified random sampling Purposive sampling Cluster sampling Snowball sampling 4(a)-i  Simple random sampling (SRS) Simple random sampling A simple random sampling is a sampling method where Prepare/get the list of population subjects. every item/individual in the population has the same Randomly select the respondents based on your probability of being chosen. calculated sample size. SRS needs a sampling frame. by using random number table (These are commonly found at the back of statistics/research method textbooks), lottery method (A researcher randomly picks numbers from a hat/ a bowl) or computer program (SPSS) https://www.youtube.com/watch?v=Uy4kGUhOuqA Systematic random sampling Systematic random sampling Is a method of selecting samples based on a Eg : select 100 medical record using a system of interval. A sample in which every kth systematic out of 1200 patients’ record in case is selected (usually with a random start), record unit, Hospital ABC where k is any constant. Sampling frame – 1200 patients’ record a) List the population subjects using numbers Sampling interval (1200/100) = 12 b) Determining the sampling interval: n = (total Choose randomly a number between 1 to 12 number of subject in a population (N)/ total Let say the first selected number is 6, then the number of respondents needed (n) next number would be 18, 30, 42,54, 66, until c) Selecting respondents you get 12 respondents.  Stratified random sampling Stratified random sampling Example: A probability sampling procedure in which the A researcher wants to determine the job stress level among population is first divided into strata (e.g. primary school teachers. Steps: gender, age group, geographical area : rural or 1) Calculate your sample size (n=50) urban). The strata is depending on the - 2) Divide the school into urban dan rural (strata): population characteristic or the elements that 2 schools urban, 2 schools rural researcher want to study. 3) Determine the total number of teachers from each school (Sc A: 20, Sc B: 30, Sc C:20, Sc D 40) N= 110 4) Calculate the sampling fraction: n/N : 0.45 5) Multiply number of teacher from each school with sampling fraction ↳ (Sc A: 9, Sc B:14, Sc C9, Sc D: 18), total number = sample size (9 + beg 14 + 9 + 18=50) : onion school ↓ Muban £↳ sbbnq , / rural area &/ taubrqpa. " ang ki ohtesqegg pom, While Cluster sample Sampling conducted in several stages. In the first stage a cluster of cases are sampled; then from these initial clusters, the actual sample of cases are selected: → shhnva Kena mask Single-stage cluster: several housing areas are selected as Non-probability cluster; each household is randomly sampled. Multistage cluster: several housing areas are selected, then sampling no some blocks are identified. From these block (flats), each floor is selected. Then each household is sampled. ¥. ¥¥;÷. ↓ %%% each , frog " " %, " " % bag § , 31 %. Stig %  -1%7%+2 s to 8 3 , ~> choose First First 10 @me %%g , → Choose area Convenience Purposive sampling → 100 people to be selected eg : school A non-probability sample in which the most convenient A non-probability sampling procedure in which the (most readily available) cases are chosen for the study researcher uses his or her judgment to select those Subject are chosen because they can be accessed easily respondents that best meet—judgment sampling and conveniently and has availability at a given time. The researcher decides what needs to be known and Convenience samples are sometimes regarded as sets out to find people who can and are willing to ‘accidental samples’ because elements may be selected in provide the information by knowledge or experience. the sample simply as they just happen to be situated (The This involves identification and selection of first 100 patients to enter pharmaceutical department were - individuals or groups of individuals that are proficient selected) and well-informed with a phenomenon of interest %:& ↓ ¥§ HER. :;É¥% % Is 9- % % a % 3- ✗ " ↳ 4 E. 9 Bay sq?'✗g,g%g %% got strata Aby gralitiue Quota sampling → already qualitative study ≠ quantitative → describe Snowball sampling people the world ~, questionnaire Sampling method where quotas are set up A snowball sampling is known as a network or according to some specified characteristics. chain referral sampling. ( other people ) meet Ex: Interviewers were asked to interview 50 Researcher will identify a small number of people living in a village. He must interview respondents who possess a specific 25 farmers, 15 housewives and 10 children. - - characteristic of interest. = With these quotas, interviewer is free to select the people to be interviewed. Then, this respondents will provide others who have the same characteristics too. ↳ ask This sampling method is useful where → and asner in qualitative sampling is difficult (drug addicts). b- Kena ask drug addicts to know addicts other thy home ? details www.tatioe - Weston  Sampling Sampling method method Advantages and disadvantages of sampling methods Advantages and disadvantages of sampling methods Probability samples Non-probability samples Sampling Advantage Disadvantage Sampling method Advantages Disadvantages method Simple random Gives all units in the target population equal Requires a large sample size. Laborious process Convenience Inexpensive, and the sample is generally The sample may not be representative chance of selected. which is very time-consuming. easy to access. of the target population. Difficult to get complete listing of every No chance of introducing researcher bias. element the population Inexpensive, does not require - Quota _ Not a probability sample and as such sophisticated skill to achieve a sample. the findings cannot be generalized to Can be expensive for large populations. the sampling population. Systematic Achieves a high degree of representativeness Requires considerable time and effort because s - the next 10 of the population. ~B , nek 20 can generalize the sample must be drawn from multiple enumerated listings. High risk of conscious sample bias, Affords a high degree of precision. Purposive Inexpensive, easy to collect the sample unless the results should be judged with Achieves an adequate representation of sub- circumspection. Stratified It is essential that the characteristics of the - layer of people groups, which could not be achieved by simple target population are known in order that random or systematic samples. stratification can be achieved. [ group of hosehold a Cluster Cluster sampling is economical when the Cluster sampling tends to contain population is large and widely dispersed. more Sampling errors than Snowball Useful when a sampling frame is not This method offers the researcher at - simple or stratified random available, and the sample is difficult to risk of sample bias. mmmm sampling. access via conventional means. 4(a)-ii minichiello for research methods in health sciences. 4(a)-iii CONCLUSION Two types of sampling Probability – simple random, systematic, stratified, cluster Non probability sampling – convenience, quota, purposive, snowball Both types of sampling have advantages and disadvantages