Sampling techniques in research
Download as PPT, PDF29 likes5,276 views
This document discusses various methods for sampling populations and collecting data, including: - Probability and non-probability sampling techniques like simple random sampling, stratified sampling, and cluster sampling. - Data collection methods like questionnaires, literature reviews, observation, and interviews. It provides details on constructing questionnaires, conducting observations, and potential sources of error.
![Census
• When one wants to study the entire
population, then he is dealing with a census
• Under the census or complete enumeration
survey method, data are collected for each
and every unit [person, household, field,
shop, factory etc.] of the population or
universe
• A census has several advantages
05/07/14 5](https://image.slidesharecdn.com/samplingjulie-140507090528-phpapp01/85/Sampling-techniques-in-research-5-320.jpg)
Sampling techniques in research
- 1. Sampling Methods and Data collection methods ATWEBEMBEIRE JULIET
- 2. SESSION OBJECTIVES The session will cover the following areas • population • sampling and sampling techniques • Data collection instruments • Data collection procedures
- 3. Population and sample • A population is a complete collection (or universe) of all the elements (units) that are of interest in a particular investigation. • A Sample: is a collection of some (a subset) the elements of a population.. 05/07/14 3
- 4. Population and Sample • Target population- this is the population to which the research ultimately wants to generalize his results • Sampled population- this is the accessible population. It the population from which the sample will be drawn • A sample is the elements within the sampled population that take part in the study 05/07/14 4
- 5. Census • When one wants to study the entire population, then he is dealing with a census • Under the census or complete enumeration survey method, data are collected for each and every unit [person, household, field, shop, factory etc.] of the population or universe • A census has several advantages 05/07/14 5
- 6. Advantages & disadvantages of a census • Data obtained from each and every unit of the population • The results obtained are likely to be more representative, accurate and reliable Costly Time consuming Sometimes not necessary That’s why normally samples are used
- 7. What is Sampling? • Sampling: is the process of selecting elements from a population in such a way that the sample elements selected represent the population 05/07/14 7
- 8. Sampling Advantages of sampling • Reduced time and cost needed to collect data • More comprehensive data is obtained than in a census 05/07/14 8
- 9. sampling Disadvantages of sampling • The selected units may not be representative of the population especially when the sample size is small. 05/07/14 9
- 10. Sampling process • Sampling process involves five steps • Defining the population • Listing the elements of the population(sample frame) • Determine the appropriate sampling methodology • Decide on the appropriate sample size • Select a representative sample
- 11. Types of samples • Representative sample also known as probabilistic sampling • A representative sample is similar to the population in all important ways. • It is selecting a sample in a way that gives every element in the population a chance to be selected
- 12. Non-representative samples • Also called Non probabilistic sampling • Not truly representative • based on subjective judgment of the researcher. • Here the researcher decides on the elements to be included in the sample • Less desirable than probability samples.
- 13. Classification of Sampling Methods Sampling Methods Probability Samples Simple Random Cluster Systematic Stratified Non- probability QuotaJudgment Convenience Snowball
- 14. Categories of Representative /probabilistic Samples Random sample • Each individual in the population of interest has an equal likelihood and or chance of selection. • Each possible sample of a given size has a known and equal probability of being the sample actually selected. • Every element is selected independently of every other elements • It can be done by use of lottery method or random method
- 15. Systematic Random sampling • The sample is chosen by selecting a random starting point and then picking every i th element in succession from the sampling frame. • The sampling interval, i, is determined by dividing the population size by the sample size and rounding to the nearest if it not a full number e.g you want to study 20 house holds out of a hundred 100/20= 5. five becomes your interval. Every fifth house hold will be selected 05/07/14 15
- 16. Stratified Samples • A stratified sample is a mini-reproduction of the population. • Before sampling, the population is divided into characteristics of importance for the research e.g Males and Females, married and not married, Blacks, Whites and Colored e.t.c
- 17. cotd • The strata should be mutually exclusive and collectively exhaustive. In that every population element should be assigned to one and only one stratum and no population elements should be omitted • Elements are then selected from each stratum by a random procedure. 05/07/14 17
- 18. Types of Stratified random sampling • Proportionate stratified random sampling; taking proportionate samples from the population irrespective of inequality in number e.g out 100 DLTM participants 40 are ladies 60 gentlemen and you want a sample of 20. 40/100 X20 =8 ladies, 60/100 X20 =12 gentlemen • Disproportionate stratified random sampling. you take samples that not proportionate e.g. 10 ladies and 10 gentlemen of more ladies if interested in female related issues. OR you may pick 15 ladies and 5 gentlemen if you think ladies have the information you need 05/07/14 18
- 19. Cluster random sampling • This is a sampling methodology in which elements of a population are grouped into clusters and simple random sampling then performed on clusters. • Clustering is similar to stratification in that both involve partitioning the population into subgroups. • It is different in that the sampling cluster is heterogeneous. 05/07/14 19
- 20. Cluster sampling • Cluster sampling can be done in stages and this is called multistage sampling. E.g you may want to study causes of poverty in uganda. You multistage cluster uganda into several clusters; • Regions • Districts • Sub counties • E.t.c 05/07/14 20
- 21. Types of Non-representative/non probabilistic samples • Quota Sample • In this type of sampling the researcher is given definite instructions about the section of the public he is to question, • However the final choice of the actual persons is left to his own convenience and is not predetermined
- 22. Judgmental sampling Also referred to as Purposive sampling • Non-representative subset of some larger population • Constructed to serve a very specific need or purpose. • The researcher chooses subjects who in his opinion are likely to supply information relevant to the research problem
- 23. Snowball Sample • A snowball sample is achieved by asking a participant to suggest someone willing or appropriate for the study. • Snowball samples are particularly useful in hard-to-track populations
- 24. Convenience or Accidental Sample • A convenience sample is a matter of taking what you can get. • In this method respondents are selected because they happen to be in the right place at the right time
- 25. In Summary; Step 1: Define the the target population Step 2: Select The Sampling Frame Step 3: Probability OR Non-probability? Step 4: Plan Selection of sampling units Step 5: Determine Sample Size Step 6: Select Sampling units Step 7: Conduct Fieldwork
- 26. Sampling Error sampling errors: a measure of the difference between a statistic and the parameter it is estimating • These arise when the sample of the population surveyed is not representative of the population from which it is drawn.e.g sampling internet users in kampala and your sample has only university students
- 27. Non Sampling errors • Wrong concepts • Coverage errors-omission of a unit in a sampling frame • Errors due to misunderstanding the questions • Response errors-wrong responses • Non response errors • Questionnaires not understood 05/07/14 27
- 28. Data Collection methods • These are methods of collecting information from the field. • The methods are determined by type of data to be collected – Primary vs secondary data – Qualitative vs quantitative data • There are four ways of gathering data: • Questionnaires • Literature study • Observation • interviews
- 29. questionaire • It is a carefully constructed instrument that consists of a set of questions to which the subject responds to in writing • Questionnaires can be close ended or open ended
- 30. Using Open-Ended Questions Common when the researcher cannot anticipate people’s responses or If listing all possible major responses is impractical. Effective for stimulating thought, probing for complex questions
- 31. cotd • Quite valuable in exploratory research. • Very demanding for respondents. • Responses to such question can be difficult to classify and code.
- 32. Closed Ended questions • Responses are forced into fixed categories • Potentially very demanding on side of researcher. • Major complaint…not all relevant options are stated.
- 33. Questionnaire Construction Questions must give you enough information to accomplish your objectives Questions must stated to enable respondents to provide accurate answers.
- 34. Question Wording Keep wording simple and straightforward… If sampling certain groups (occupational groups), complex wording (professional jargon) may be necessary to facilitate communication & credibility.
- 35. cotd –Wording for more sensitive questions should be carefully designed. –Be extra careful in developing questions designed to obtain information on attitudes and beliefs.
- 36. Question Wording Aim for Clarity: vague questions produce vague answers. Avoid biased or leading questions. Avoid double-barrelled questions.
- 37. Literature study /document analysis • Method of data collection that involves analysis of documents. • These could be text books, magazines, financial reports, attendance registers e.t.c
- 38. observation • Observation Means seeing or viewing. • Observation can be of different types. • Participant observation where the observer is part of the phenomenon or group, which is being observed, and he acts as both an observer and a participant. • Non-participant observation where the observer stands apart and does not participate in the phenomenon being observed.
- 39. Advantages – It enables the researcher to study behavior as it occurs and this makes the data more reliable and free from respondent’s bias – It is the only method, which can be used when the participants cannot express themselves meaningfully e.g. children, the very sick, the mob, those in riots etc – If the individual is unwilling to be interviewed or fill in questionnaires
- 40. advantages – The researcher can adjust the goals and objectives of the study as the data is being collected. – Observation provides a richer and more direct account of the phenomenon and a study as a whole since behavior is taking place in its natural environment
- 41. Disadvantages • The researcher has to wait until the phenomenon or event takes place • It is difficult to have a random or representative sample, which is needed in the sample • Compared to interviewing it might be less expensive in terms of money and equipment but it costs more man-hours spent in the field and requires personal adoption to the conditions of the field
- 42. Disadv cotd • It cannot be used to study attitudes and opinions • Observed data is difficult to compress, quantify or reduce to codes which can be tabulated or processed by a computer • Observer fatigue could easily set in and distort our observation
- 43. interviews • In the interview, the researcher talks to the respondent and obtains information directly.
- 44. Advantages • Flexible. • In-depth. • Situation can be adapted. • Reasons for answers can be sought. • Clues can be followed up. • Yields a higher percentage of answering.
- 45. disadvantages • Time. • Costs. • Difficult to analyze responses. • Subjectivity.
- 46. Focus Group discussions • Focus Group: An interview conducted among a small number of individuals simultaneously; the interview relies more on group discussion than on directed questions to generate data. • Characteristics of Focus Groups • Typically 8 – 12 people • Homogeneous within group • 1.5 to 2 hours in length • Sessions recorded and transcribed
- 47. Further reading • Research is never conclusive, please read more • Amin, E. M. (2005). Social Science research: conception, methodology and analysis. Makerere University printery. Kampala • Neuman, W.L. (2003). Social Research Methods: Qualitative and Quantitative Approaches. Fifth Edition Pearson Education; Inc. 2003 Boston, USA