W13-L11 (DSA 4580) Sampling Techniques (PDF)

Research Methodology (DSA 4580) Step 3: Designing a Study and Collecting Data Sampling Techniques Academic Year: 2023-2024 Fifth Year DSA 4580 WEEK 13 LECTURE 11 Dr Inderjit M G, BDS, MDS in Public Health Dentistry [email protected] Thursday, 16 November 2023 Copyright © 2023 by PSAU, Abdulh...

Research Methodology (DSA 4580) Step 3: Designing a Study and Collecting Data Sampling Techniques Academic Year: 2023-2024 Fifth Year DSA 4580 WEEK 13 LECTURE 11 Dr Inderjit M G, BDS, MDS in Public Health Dentistry [email protected] Thursday, 16 November 2023 Copyright © 2023 by PSAU, Abdulhamid Al Ghwainem Disclaimer DISCLAIMER The information presented in this lecture is offered for educational and informational purposes and should not be construed as medical, dental, or research advice. While the amount of information in this handout is vast, and I make every effort to be as current and thorough as possible, the information cannot be taken as a reference manual or textbook. Please note that you should read the required textbooks as specified in the course curriculum and lecture references. 2 Notice WARNING Materials used in connection with this course or lecture may be subject to copyright protection. This material has been reproduced and communicated to you by or on behalf of Prince Sattam Bin Abdulaziz University. Materials may include, but are not limited to: documents, slides, images, audio, and video. Materials in this course are only for the use of students enrolled in this course, for purposes associated with this course, and may not be retained for longer than the class term. Unauthorized retention, duplication, distribution, or modification of copyrighted materials may be the subject of copyright protection by law. For more information, visit the Saudi Authority for Intellectual Property. Do not remove this notice 3 Lecture Objectives By the end of this session, you should be able to: • Differentiate between the population and the sample. • Describe the difference between homogenous and heterogeneous samples. • Differentiate between probabilistic and nonprobabilistic sampling and their types. • Explain what is meant by representativeness and generalizability. • Discuss sampling error and differentiate between a random sampling error and a system sampling error. • Explain the importance of knowing the who, the how, and the why for the purpose of sampling for different research designs. 4 Research Process/Steps Research: answering questions in logical and systematic ways Question Answer Research methodology: how to get from Question to Answer Identify study question Select study approach Design study and collect data 5 Analyse data Report findings Research Process/Steps Research: answering questions in logical and systematic ways Identify study question Select study approach Design study and collect data • Research Ethics • Sampling Techniques 6 Analyse data Report findings Sampling: overview § All research projects involve gathering specific data from specific sources in specific places at specific times. § Also known as sampling, the necessity of sampling occurs because we simply cannot gather all data from all sources at all places and all times. § In other words, we must make choices when we design our research projects. § This session on sampling techniques as another level of choice to be made by the researcher. 7 Objectives of Studying Sampling § To obtain the maximum information about the population without examining each and every unit of the population. § Finding the reliability of the estimates derived from the sample can be done by computing the standard error of the statistic 8 Populations and Samples § Studying populations is too expensive and time-consuming, and thus impractical § If a sample is representative of the population, then by observing the sample we can learn something about the population § And thus, by looking at the characteristics of the sample (statistics), we may learn something about the characteristics of the population (parameters). 9 10 METHOD OF SAMPLE SELECTION Sampling Population Sample Target Population Target Population è Sampling Population è Sample 11 Types of Research Populations Target population > source population > sample population > study population 12 Target and Source Populations § A well-defined study question identifies a target population to which the results of the study should apply –Unless a study goal is extremely narrow in scope, it is usually not possible to invite all members of a target population to participate in a study. –Instead, a more specific source population should be identified. § Ideally, the sampling frame consists of an enumerated list of population members. 13 Types of Research Populations At least four types of populations must be considered when preparing to collect data: 1. Target population: the broad population to which the results of a study should be applicable. 2. Source population (sampling frame): a well-defined subset of individuals from the target population from which potential study participants will be sampled. 3. Sample population: the individuals from a source population who are invited to participate in the research project. 4. Study population: the eligible members of the sample population who consent to participate in the study and complete required study activities. 14 SAMPLE § A group of sampling units that forms part of a population generally selected so as to be representative of the population whose variables are under study. § The smallest part of the population is called as a UNIT/INDIVIDUAL. Any FINITE part of the population is SAMPLE. A sample is a subset of the target population. 15 SAMPLING § Sampling may be defined as the selection of some part of an aggregate or totality on the basis of which a judgment or inference about the aggregate or totality is made. Or § It is the process of obtaining information about an entire population by examining only a part of it. 16 SAMPLING FRAME § A list containing all sampling units is known as sampling frame. Or § The sampling frame consists of a list of items from which the sample is to be drawn. 17 SAMPLING DESIGN (SAMPLING TECHNIQUE) § A sampling design is a definite plan for obtaining a sample from the sampling frame. Or § It is the procedure the researcher would adopt in selecting some sampling population is drawn. 18 Sample Populations § When a source population is much larger than the sample size required for a study, a subset of the source population may serve as a sample population. § Sampling bias occurs when the individuals sampled for a study systematically are not representative of the source population as a whole. –Non-random-sampling bias occurs when each individual in the source population does not have an equal chance of being selected for the sample population. 19 SELECTION BIAS § Occurs when there is a systematic difference between the characteristics of the people selected for a study & the characteristics of those who are not. MEASUREMENT BIAS § Occurs when the individual measurements or classifications of disease or exposure are inaccurate (i.e., they do not measure correctly what they are supposed to measure). 20 § A sample has to be taken from a population § A sample should be representative of the population § Value calculated from sample is called as STATISTIC § Value calculated from population is called as PARAMETER Statistics are NOT constants whereas parameters are constants. 21 The main features of a representative sample are 1. PRECISION 2. UNBIASED CHARACTER The main Objectives of Sampling are : 1. Estimation of population PARAMETERS from sample STATISTICS. 2. To TEST THE HYPOTHESIS about population from which the sample or samples are drawn. 22 Fr a me SAMPLING ERRORS r o r e c r n a E Ch e s on p s e R Error Error Response Sample Population 23 Sampling frame The total error in a study is the sum of the sampling and non-sampling errors T.E = S.E + N.S.E. Sampling error is the sum of the frame error, chance error & response error S.E. = F.E + C.E + R.E 24 SAMPLING TECHNIQUES There are two types of sampling techniques. 1. Probability sampling – § A variety of probability-based sampling methods can be used to ensure that all members of a source population have an equal likelihood of being invited to participate in a research study. 2. Non probability sampling – § Sometimes a nonprobability-based sample is appropriate. Nonprobability samples may also be used but are weaker methodologically since they are not likely to be representative of the population of interest. 25 SAMPLING METHODS PROBABILITY SAMPLING: § Simple random sampling NON PROBABILITY SAMPLING: § Judgment sampling § Stratified random sampling § Quota sampling § Systematic random sampling § Cluster random sampling § Convenience sampling § Multistage random sampling § Snowball sampling § Multiphase random sampling 26 JUDGEMENT SAMPLING § Selection of samples is left to the Judgment of investigator. § In this sampling accuracy of results depends upon investigator. ADVANTAGES : 1. Employed mainly when population is small. 2. Employed to conduct pilot study. LIMITATIONS : 1. Accuracy of results depends upon the knowledge of the investigator. 2. If investigator is biased it affects the acceptance or rejection of an hypothesis 27 QUOTA SAMPLING § Each investigator is allotted quota of persons which are to be interviewed. § Investigators are given instructions to interview persons within the quota with some specified characteristics. Example : Persons within the quota of 10 students, 6 professionals. - 20 students from each year of a 3-year course. 28 Convenience / Chunk Sampling / incidental sampling § Chunk is a fraction of population which is selected for investigator because it is conveniently available. Example : - In order to estimate oral hygiene status in the city the investigator may select a few areas near by his house. - Passers-by or visitors in a hospital canteen who just happen to be present on the day of the survey. § Results of this sampling are rarely representative because they are generally biased. 29 § A convenience population is a nonprobability-based source population selected due to ease of access to those individuals, schools, workplaces, organizations, or communities. § Convenience sampling must always be used with caution, since convenient sample populations are often systematically different from the target and source populations they are intended to represent. 30 Snowball sampling § Where a participant is asked to pass information to others who might be interested. This is not used very often in surveys but is useful for some groups, such as patients with rare diseases, who may know others with the same condition. 31 SIMPLE RANDOM SAMPLING (unrestricted random sampling ) § A procedure of selecting a sample in which every item in a population has an equal chance of being included in a sample. § Applicable when population is very small, homogeneous and readily available. To ensure randomness – lottery method. ADVANTAGES : No harm due to personal bias. DISADVANTAGES : Selection of sample costly and time consuming. 32 33 34 Systematic Random sampling It is applied to field studies when the population is large, scattered & homogenous. K=N/n K - sample interval or sample ratio N - population size n - Sample size. Ex : If 150 patients are to be included in the sample from a population of 3000. I.e. K=3000/150=20. ADVANTAGES : - Systematic design is simple, convenient to adopt. - The time & labor in collection of sample is relatively small. - It gives accurate results when population is large 35 36 Stratified random sampling - If population is heterogeneous, simple random sampling is not useful. - Purpose of this sampling is to increase the efficiency of sampling by dividing heterogeneous into homogenous. - These homogenous groups are termed as strata. Ex: Areas, classes, ages, sexes etc., ADVANTAGES : - There is a greater precision of results. - It gives better results when population is scattered. DISADVANTAGES : - It is too technical method - Time consuming. 37 Cluster sampling - When population is vast & scattered over a wide area, cluster sampling is applicable. - In this sampling the population is divided into groups, then a required no of groups or clusters are selected by simple random sampling. Ex : - City is divided into Sectors, required no of sectors are selected randomly. Sectors ---- I stage Blocks ---- II stage. 38 39 Multistage sampling - As the name implies this method refers to the sampling procedures carried out in several stages using random sampling technique. Country Province Governance Centers Villages Wards 1st stage 2nd stage 3rd stage 4th stage 5th stage 40 Multiphase sampling - In this method part of the information is collected from the whole sample & part from the subsample. Ex : T.B. Patients. Physical examination or Mantoux test. X-ray Sputum examination. Number of subsample in 2nd & 3rd phase will become successively smaller and smaller ADVANTAGES : - Less costly & more purposeful. - Less laborious. 41 Sample Populations 42 Sample Populations § No matter which sampling method is used, the goal is to end up with a sample population that is representative of the source population. § Some errors occur by chance and cannot be resolved by randomization, but many forms of bias can be mitigated with careful planning, rigorous methods, and sufficiently large sample sizes –Sampling bias can be avoided by selecting more appropriate source populations and applying inclusion and exclusion criteria consistently 43 Sample Populations § Berkson’s bias can occur when cases and controls for a study are recruited from hospitals and therefore are more likely than the general population to have comorbid conditions. § Healthy worker bias can occur when participants are recruited from occupational populations and therefore are systematically healthier than the general population. § Exclusion bias occurs when different eligibility criteria are applied to cases and controls, such as when controls with health conditions related to an exposure are excluded but cases with those comorbidities are not excluded. 44 Errors ØSampling errors: faulty sample design small sample size ØNon sampling errors: coverage error observational error processing error 45 Study Populations § Participation rate: the percentage of members of a sample population who are included in the study population. § In an ideal situation all of the sampled individuals agree to participate in the study, but a 100% participation rate is extremely rare. § A high participation rate helps prevent the selection bias that occurs when the members of the study population are not representative of the source population from which they were drawn. § A low response rate may result in nonresponse bias if the members of a sample population who agree to participate in a study are systematically different from nonparticipants. 46 Sampling for Cross-Sectional Surveys § The study population for a cross-sectional study should adequately represent the target population. § A population-based study uses a random sampling method to generate a sample population that is representative of a well-defined larger population § The most rigorous population-based studies use probability-based sampling methods to generate sample populations and then confirm that study populations are reasonably representative of the source populations from which they were drawn. § Convenience populations are not suitable for most cross-sectional studies, because they are not population-based. 47 Sampling for Cross-Sectional Surveys 48 Sampling for Case-Control Studies § When identifying possible participants for a case–control study, the first step is to find an appropriate and available source of cases. § After a source of cases is identified, a valid control group must be selected. § All cases and all controls in a case–control study must meet the same eligibility criteria, except for the ones relating to disease status. –The eligibility criteria for a study comprise the inclusion criteria that must be present for an individual to be allowed to participate in a study and the exclusion criteria that require an individual to be removed from the study population. 49 Sampling for Case-Control Studies 50 Sampling for Cohort Studies § The sampling methods used for a cohort study must align with the particular type of study design that will be applied. § For longitudinal cohort studies, the process of identifying representative source and sample populations is similar to the process for identifying these populations for a cross-sectional study. § For prospective cohort studies that seek to compare exposed and unexposed populations, identifying exposed and unexposed participants is similar to the steps for identifying cases and controls for a case–control study. 51 Sampling for Cohort Studies 52 Sampling for Experimental Studies § Sampling methods for experimental studies focus on the validity of the study and the safety of participants. § The risk of harm can be reduced by selecting an appropriate source population and defining strict inclusion and exclusion criteria. 53 Sampling for Experimental Studies 54 Sampling for Qualitative Studies § Qualitative data collection is not a detached, structured process based on a random sample of individuals. § Instead, researchers typically have intense contact with a selected group of informants –Key informants are individuals selected to participate in a qualitative study because they have expertise relevant to the study question –Purposive sampling is a nonprobability-based sampling method that recruits participants for a qualitative study based on the special insights they can provide 55 Sampling for Qualitative Studies § Qualitative studies often do not start with a set number of participants that are supposed to be recruited. § The goal is to reach data saturation, a time in the research process in which no new information about a particular theory is emerging from additional data collection because variations across population members have already been captured. –Some studies with homogeneous populations might reach saturation after 15 or 20 interviews, but some require larger numbers of participants. 56 Importance of Sample Size § When determining how many participants are needed for a quantitative or qualitative study to be meaningful, the goal is to recruit just the right number of participants, not too many and not too few –Recruiting too many participants wastes resources. –Recruiting too few participants makes the study invalid. 57 Sample Size § In statistics, sample size is the number of observations in a data set. –In the health sciences, the sample size is usually the number of individual humans in the study population. § The desired sample size for a quantitative study is based on statistical estimations about how many data points are required in order to answer the study question with a specified level of certainty. 58 Sample Size 59 Sample Size § A confidence interval (CI) is a statistical estimate of the range of likely values of a parameter in a source population based on the value of that statistic in a study population. § A narrow CI indicates more certainty about the value of the statistic than a wide CI. § Large data sets have confidence intervals that are narrower than the confidence intervals generated by smaller sample sizes. 60 Sample Size 61 Sample Size § A sample size calculator is a tool used to identify an appropriate number of participants to recruit for a quantitative study. –The range of suggested sample sizes is based on a series of assumptions (= guesses) about the expected characteristics of the sample population. –Sample size calculators are freely available online and are bundled with most statistical software programs. § When the level of certainty about inputs is low, it is better to err on the side of a larger sample size. 62 Sample Size 63 Sample Size 64 Type 1 and 2 Errors § In statistics, an error is a difference between the value obtained from a study population and the true value in the larger population from which the study participants were drawn that occurs by chance rather than as a result of systematic bias. § A type 1 error (α) occurs when a study population yields a statistically-significant test result even though a significant difference or association does not actually exist in the source population. –When α is set at 5%, about 1 in 20 statistical tests will result in a type 1 error. § A type 2 error (β) occurs when a statistical test of data from a study population finds no significant result even though a significant difference or association actually exists in the source population. –The best way to minimize the likelihood of type 2 errors is to have a large sample size. 65 Type 1 and 2 Errors 66 Power Calculation § Power is the ability of a test to detect significant differences in a population when differences really do exist. –Power is defined as 1 – β, so a 20% likelihood of a type 2 error corresponds to a power of 80%. § The power of statistical tests is increased when the number of participants included in the analysis is large. 67 Power Calculation 68 Power Calculation § Researchers must be prepared to rethink the study approach if their estimated number of available participants will not yield sufficient power. § The sample size estimates generated by sample size and power calculators refer to the study population (the actual number of participants), not the sample population (the number of individuals invited to participate in the study). § The number of people sampled for a study needs to be larger than the required number of participants, because the participation rate is unlikely to be 100%. 69 References Required: Neale, J., 2020. Research methods for health and social care. Bloomsbury Publishing. Jacobsen, K.H., 2020. Introduction to health research methods: A practical guide. Jones & Bartlett Publishers. Additional: Bowling, A., 2014. Research methods in health: Investigating health and health services. Maidenhead, United Kingdom: Open University Press. Creswell, J.W. and Creswell, J.D., 2018. Research design: Qualitative, quantitative, and mixed methods approaches. Sage publications. World Health Organization. Regional Office for the Western Pacific, 2001. Health research methodology : a guide for training in research methods. 2nd ed.. WHO Regional Office for the Western Pacific. https://apps.who.int/iris/handle/10665/206929 World Health Organization. Regional Office for the Eastern Mediterranean. (2004). A practical guide for health researchers. https://apps.who.int/iris/handle/10665/119703 Next Lecture: Step 3: Designing a Study and Collecting Data Questionnaire Development Identify study question Select study approach Design study and collect data Thank you! Any questions [email protected] Analyse data Report findings