Methods.pdf
- 1. Research Methods & Materials 1 Zebenay Workneh (MSc, MPH, Assistant Professor)
- 2. Research methods 2 Methods the specific techniques, tools or procedures applied to achieve a given objective The techniques researchers use in performing research operations All those methods used by aresearcher during the course of studying his research problem
- 3. Main Components of research methods 3 Study area and period Study design Study Population Sample size and sampling procedure Variables and Operational definitions Data collectionTools and methods Data processing and analysis Ethical issues in research
- 4. •Study area and period: •Study area – Description of the background information including: Population, geography, institution as relevant, use map if possible. •Study period- data collection period.
- 5. • Choosing study population: • Sampling: • take a sub group from a large population & the sub group used as a basis for making inferences regarding the larger population. • Because; • Difficulty to get information from everyone in the population, • To obtain representative sample of a population, • Feasibility, reduced cost, greater accuracy and greater speed, • Population is dynamic (so that data should be collected in short time). • However, the operation needs rigid control; sampling error, & smallness in number render study to be suspected.
- 6. • The whole population is studied; • For enumeration (census), • when there is small population, and • where there are extensive resources • Otherwise, sampling. • The issues of representative sampling technique and adequate sample size are important for correct estimation of the parameter based on the statistic. • Target/source/reference population:The population to which inference is made. • Ideally the source and study population should overlap.
- 7. Important terms: • Study population:The part of the source population that‟s given the chance to be included in the study. • Sample: Part of the study population that‟s actually studied. •Sampling unit: Is the unit of selection in the sampling process. • Eg. Household • Study unit:The unit from which information is collected. • Eg. Study subjects (eg. Mother, under-five children, etc.)
- 8. Sampling techniques: • Probability sampling: Every unit in the population has a known, non-zero probability, of being sampled and the process involves random selection. • Non- probability sampling: is any sampling method where some elements of the population have no chance of selection or where the probability of selection can't be accurately determined.
- 9. Classification of Sampling Techniques 9 Sampling Techniques Nonprobability Sampling Techniques Probability Sampling Techniques Convenience Sampling Judgmental Sampling Quota Sampling Snowball Sampling Cluster Sampling Stratified Sampling Simple Random Sampling Systematic Sampling Multistage sampling
- 10. The Sampling Design Process 10 i. Define the Target Population ii. Determine the Sampling Frame iii. Select a Sampling Technique iv. Determine the Sample Size v. Execute the Sampling Process
- 11. 11
- 12. 12 Population: the material of the study whether it is human subjects, animals or inanimate objects. Reference population (target population): the population of interest, to which the investigators would like to generalize the results of the study. Study population: accessible population Population from which the sample actually was drawn and about which a conclusion can be made. the subset of the target population from which a sample will be drawn and conclusion is made.
- 13. 13 Sampling frame: the list of all the units in the reference population, from which a sample is to be picked Sample population: the actual group in which the study is conducted or data is collected Sampling Unit: the unit of selection of a sample e.g. Households, people, etc… Study unit: the units on which information will be collected. E.g. individuals
- 14. •Probability sampling: •Simple random sampling (SRS), •Systematic random sampling, •Stratified sampling, •Cluster sampling, •Multistage sampling.
- 15. • Simple random sampling • Principle: Equal chance/probability of drawing each unit. • Procedure: List all units (persons) in a population, assign a number to each unit , and randomly select units. • Eg. Calculate the prevalence of anemia among 1000 pregnant women attendingANC (sample size 100). • List all pregnant women attendingANC. • Randomly select 100 numbers between (1 and 1000). • Lottery method • Table of random numbers • Using computer programs
- 16. • Systematic random sampling • Principle: Select sample at regular intervals based on sampling interval. • Procedure: • Number the units in the population from 1 to N, • Decide on the n (sample size) that you need, • Calculate the sampling fraction k (K = N/n), • Randomly select an integer between 1 to k, • Then take every kth unit through out the sampling frame.
- 17. • Eg.1: N= 12, n=4, calculate k? then select an integer b/n 1 and k and list the samples? • Eg. 2: N=200, n=50, calculate k and list the samples?
- 18. 18 • Stratified sampling • Applied when the source population is heterogeneous on one or more variables of interest. • The usual intension is to assure that the sample is representative based on the heterogeneous variable/s. • The population is first divided into classes (strata) base on the variable. • Then separate samples are taken from each stratum using simple or systematic random sampling technique. • The number taken from each stratum might be equal (non proportional stratified sampling) or the number is determined based on the proportion of each class in the source population (proportional stratified sampling). • Samples can be stratified across more than one variable.
- 19. Principle: • Select random samples from within homogeneous subgroups (strata) • Sampling frame divided into groups (age, sex, socioeconomic status) • Units in each group have the same probability of selection, but probability differs between groups. Procedure: • List all units (persons) in a population. • Divide the units into groups (called strata). • Assign a number to each unit within each stratum. • Select a random sample from each stratum (Simple or Systematic). • Combine the strata samples to form the full sample.
- 20. • A stratified sampling approach is most effective when: • Variability within strata are minimized and variability between strata are maximized. • The stratifying variable is strongly correlated with the desired dependent variable.
- 21. Cluster sampling Principle: Select all units within randomly selected geographic clusters. Procedure: • Divide population into geographic groups (clusters). • Assign a number to each cluster. • Randomly select clusters. • Sample all units within selected clusters OR select a random sample of units within selected clusters. • Cluster sampling; • Is a sampling method applied when the source population is composed of “natural” groups. • Assuming the groups are homogenous among each other, cluster sampling selects few groups (clusters) from the population as Primary Sampling Unit (PSU). • Then the required information is collected from all elements, Secondary Sampling Units (SSU), within each selected group. • Eg. Researchers usually use pre-existing units such as schools or cities as their clusters.
- 22. Multistage sampling • Is like cluster sampling, but involves selecting a sample within each chosen cluster, rather than including all units in the cluster. • Also called multistage cluster sampling. • Thus, multi-stage sampling involves selecting a sample in at least two stages. 22
- 23. Limitations of Sampling 23 Demands more rigid control in undertaking sample operation. Sampling error Sampling bias Minority and smallness in number of sub-groups often render study to be suspected. Sample results are good approximations at best.
- 24. 24 Errors When we take a sample, our results will not exactly equal the correct results for the whole population. That is, our results will be subject to errors. This error has two components:
- 25. 25 Sampling and Non-sampling errors a) Sampling error (i.e., random error) Random error, the opposite of reliability (i.e., Precision or repeatability), consists of random deviations from the true value, which can occur in any direction. can be minimized by increasing the sample size. Reliability (or precision): This refers to the repeatability of a measure, i.e., the degree of closeness between repeated measurements of the same value.
- 26. 26 b) Non Sampling error (i.e., Bias) Bias, the opposite of validity, consists of systematic deviations from the true value, always in the same direction. It is possible to eliminate or reduce by careful design of the sampling procedure.
- 27. 27 A study design is the process that guides researchers on how to collect, analyse and interpret observations. Research design is a master plan specifying the methods and procedures for collection and analyzing the needed information. It is a logical model that guides the investigator in the various stages of the research. Study Design
- 28. 28 Study designs could be exploratory, descriptive or analytical 1. Exploratory studies Are a small-scale study of relatively short duration, which is carried out when little is known about a situation or a problem. It may include description as well as comparison. Examples: A national AIDS Control Program wishes to establish counseling services for HIV positive and AIDS patients, but lacks information on specific needs patients have for support. To explore these needs, a number of in-depth interviews are held with various categories of patients (males, females, married and single) and with some counselors working on a program that is already underway.
- 29. 29 2. Descriptive studies studies that describe the patterns of disease occurrence and other health- related conditions by person, place and time. Characteristics Mainly concerned with distribution Useful for allocation of resources Important for hypothesis generation Less time consuming and less expensive Most common type of epidemiological design strategies in medical literature
- 30. 30 Uses of Descriptive Studies They can be done fairly quickly and easily Allow planners and administrators to allocate resources Provide the first important clues about possible determinants of a disease (useful for the formulation of hypotheses)
- 31. 31 Types of descriptive studies a) Case reports b) Case series c) Ecological studies d) Cross-sectional studies Ecological studies: data from entire populations are used to compare disease frequencies between different groups during the same period of time or in the same population at different points in time
- 32. 32 3.Analytic studies Studies used to test hypotheses concerning the relationship between a suspected risk factor and an outcome and to measure the magnitude of the association and its statistical significance.
- 33. Characteristics of Analytic Studies 33 Focus on the determinants (causes) of diseases. Used to test hypothesis Major distinguishing feature of analytic studies is the use of controls.
- 34. 34 Two broad categories 1. Observational (Analytic cross sectional studies, Case control Cohort studies, cohort studies, ) And 2. Interventional study design (experimental studies and quasi- experimental studies) Observational studies No human intervention involved in assigning study groups Simply observe the relationship between exposure and disease.
- 35. Things considered when choosing a study design 35 The objective/ the research question The time you have The money you have The expertise you have The requirements of an organization
- 36. 36 In planning any investigation, we must decide how many people need to be studied in order to answer the study objectives. If the study is too small, we may fail to detect important effects, or may estimate effects too imprecisely. If the study is too large, then we will waste resources. Sample Size Determination
- 37. Sample size determination depends on the: 37 Objective of the study Design of the study Descriptive/Analytic Degree of precision or accuracy – the allowed deviation from the true population parameter (can be within 1% or 5% etc) Plan for statistical analysis Degree of confidence level required, usually specified as 95% (level of confidence that the proportion in the whole population is indeed between (p-d) and (p+d))
- 38. 38 To estimate population mean, 𝑛 = 𝑍α/2 2 𝜎2 𝑑2 Determination of Sample Size for Estimating Means
- 39. 39 Estimating a mean • The same approach is used but with SE = / n • The required (minimum) sample size for a very large population is given by : n = Z2 2 / w2 Eg. A nurse wishes to estimate the mean serum cholesterol in a population of men. From previous similar studies a standard deviation of 40 mg/100ml was reported. If he is willing to tolerate a marginal error of up to 5 mg/100ml in his estimate, how many subjects should be included in his study ? ( =5%, two sided) a)If the population size is assumed to be very large, the required sample size would be: n = (1.96)2 (40)2 / (5)2 = 245.86 246 persons
- 40. 40 The minimum sample size (n) required for a very large population (N>10,000) is: Determination of Sample Size for Estimating proportion Where • n = required sample size, • p = proportion of the population having the characteristic, • q= 1-p • d = the degree of precision.
- 41. Example 41 Suppose that you are interested to know the proportion of infants who breastfed >18 months of age in a rural area. Suppose that in a similar area,the proportion (p) of breastfed infants was found to be 0.20. What sample size is required to estimate the true proportion within ±3% with 95% confidence. Let p=0.20, d=0.03, α=5%
- 42. 42 The above formulas are used with the assumption of a very large population (N>10,000) When the sample represents a significant (e.g. over 5%) proportion of the population, a finite population correction factor can be applied. This will reduce the sample size required. For fine population use the following formula. Finite population correction factor Where n = the adjusted sample size, n0 = the original required sample size and N = population size.
- 43. 43 Comparison of two proportions n (in each region) = f(,) (p1q1 + p2q2) / ((p1 - p2)² =type I error (level of significance) =type II error ( 1- =power of the study) power =the probability of getting a significant result f (,) =10.5, when the power =90% and the level of significance =5% f (,) =9.0, when the power =85% and the level of significance =5% f (,) =7.84, when the power =80% and the level of significance =5% Eg. The proportion of nurses leaving the health service is compared between two regions. In one region 30% of nurses is estimated to leave the service within 3 years of graduation. In other region it is probably 15%.
- 44. 44 Solution The required sample to show, with a 90% likelihood (power), that the percentage of nurses leaving the health service is different in these two regions would be: (assume a confidence level of 95%) neach = (1.28+1.96)2 ((.3.7) +(.15 .85)) / (.30 - .15)2 = 158 If 10% is added for non-response and other contingencies, neach = 158 + 16 = 174 Therefore, 174 nurses are required in each region.
- 45. Notation: 45 Take the values of p and standard deviations from other published research findings If there are more than one p value, take a p value that gives the maximum value when p is multiplied with q. Other option: do pilot study If the value of p is unknown, take 50% Consider non response (5-10%) in your sample size calculation
- 46. Variables 46 Variable is a concept which can take on different quantitative values. A variable is a quantity which can vary from one individual to another. Variable is a property that taken on different value. For example; height, weight, income, age etc. The main focus of the scientific study is to analyze the functional relationship of the variables.
- 47. Types of study variables 47 1. Dependent variable: a.k.a outcome variable If one variable depends or is a consequence of other, it is termed as dependent variable. 2. Independent variable: a.k.a predictor, factor, determinant, explanatory The variable that is antecedent to the dependent variable is termed as an independent variable.
- 48. 48 Operationalizing variables by choosing appropriate indicators is important Operationalizing variables means that you make them „measurable'. E.g. In a study on VCT acceptance, you want to determine the level of knowledge concerning HIV in order to find out to what extent the factor „poor knowledge‟ influences willingness to be tested for HIV. The variable ‘level of knowledge’ cannot be measured as such. You would need to develop a series of questions to assess a person‟s knowledge, for example on modes of transmission of HIV and its prevention methods. Operational Definition of Variables
- 49. 49 If 10 questions were asked, you might decide that the knowledge of those with: 0 to 3 correct answers is poor, 4 to 6 correct answers is reasonable, and 7 to 10 correct answers are good.
- 51. 51 Plan for Data Collection Why should you develop a plan for data collection? A plan for data collection should be developed so that: You will have a clear overview of what tasks have to be carried out, who should perform them, and the duration of these tasks; You can organise both human and material resources for data collection in the most efficient way; and You can minimise errors and delays which may result from lack of planning (for example, the population not being available or data forms being misplaced).
- 52. 52 Methods of Collecting Quantitative Data The most commonly used methods of collecting information (quantitative data) are: • The use of documentary sources • Interview administered questionnaire • Self-administered questionnaire Data collection Methods
- 53. 53 A.The use of documentary sources Clinical records and other personal records, death certificates, published mortality statistics, census publications, etc. Advantages: Documents can provide ready-made information relatively easily The best means of studying past events.
- 54. 54 Disadvantages Problems of reliability and validity (because the information is collected by a number of different persons who may have used different definitions or methods of obtaining data). There is a possibility that errors may occur when the information is extracted from the records. (This may be an important source of unreliability if handwritings are difficult to read. Since the records are maintained not for research purposes, but for clinical, administrative or other ends, the information required may not be recorded at all, or only partly recorded.
- 55. 55 B. Self-administered Questionnaire The respondent reads the questions and fills in the answers by himself. Advantages Simpler and cheaper: questionnaires can be administered to many persons simultaneously. They can be sent by post. Disadvantage Demands a certain level of education on the part of the respondent.
- 56. 56 C. Interview Questionnaire Interview may be highly structured interview or relatively unstructured. Advantages: Stimulate and maintain the respondent's interest Allay if anxiety is aroused (e.g., why am I being asked these question?) Repeat unclear questions “Follow-up” or “probing” questions to clarify a response Observations during the interview
- 57. 57 Disadvantages: Expensive and time taking Leading/guiding question Difficult to address sensitive issues Social desirability bias: Occurs because subjects are systematically more likely to provide a socially acceptable response. In general, apart from their expense, interviews are preferable to self-administered questionnaires provided that they are conducted by skilled interviewers.
- 58. 58 The choice of methods of data collection is based on: The accuracy of information they yield Practical considerations, such as, the need for personnel, time, equipment and other facilities, in relation to what is available.
- 59. Tools for data collection 59 The construction of a research instrument or tool for data collection is the most important aspect of a research project The famous saying about computers- “garbage in garbage out”- is also applicable for data collection. The research tool provides the input into a study and therefore the quality and validity of the output (the findings),are solely dependent on it.
- 60. Guidelines to Construct a data collection Tools: 60 Step I: Clearly define and individually list all the specific objectives or research questions for your study. Step II: For each objective or research questions, list all the associated questions That you want to answer through your study. Step III: Take each research question listed in step II and list the information Required to answer it. Step IV: Formulate question(s) to obtain this information.
- 61. 61 A questionnaire consists of a set of questions presented to a respondent for answers. The respondents read the questions, interpret what is expected and then write down the answers themselves. It is called an Interview Schedule when the researcher asks the questions (and if necessary, explain them) and record the respondent‟s reply on the interview schedule. Questionnaire design
- 62. 62 Questionnaires are a very convenient way of collecting useful comparable data from a large number of individuals. However, they can only produce valid and meaningful results if the questions are clear and precise and if they are asked consistently across all respondents Questionnaire should be developed and tested carefully before being used on a large scale. Therefore, careful consideration needs to be given to the design of the questionnaire.
- 63. 63 Types of questions 1. Closed ended questions: include all possible answers/prewritten response categories, and respondents are asked to choose among them. -e.g. multiple choice questions, scale questions 2. Open ended questions: allow respondents to answer in their own words. Questionnaire does not contain boxes to tick but instead leaves a blank section for the respondents to write in an answer. 3. Combination of both: -Begins with a series of closed –ended questions, with boxes to tick or scales to rank, and then finish with a section of open-ended questions or more detailed response.
- 64. 64 The type and content of a questionnaire depends much on your research question and research objectives (be clear about your dependent and independent variables) All questionnaires require a title (short description) It needs to be appealing and inviting It needs a confidential unique identifier How to construct questionnaires?
- 65. 65 In questionnaire design remember to: Use familiar and appropriate language Avoid abbreviations, double negatives, etc. Avoid two elements to be collected through one question Pre-code the responses to facilitate data processing Avoid embarrassing and painful questions Watch out for ambiguous wording Avoid language that suggests a response Start with simpler questions Ask the same question to all respondents Provide other, or don‟t know options where appropriate
- 66. 66 Provide the unit of measurement for continuous variables (years, months, k.g, etc) For open ended questions, provide sufficient space for the response Arrange questions in logical sequence Group questions by topic, and place a few sentences of transition between topics Provide complete training for interviewers Pre-test the questionnaire on 5% respondents in actual field situation Check all filled questionnaire at field level Include “thank you” after the last question
- 67. 67 A) Possible Sources of Bias during data collection: 1. Defective instruments 2. Observer bias 3. Effect of the interview on the informant 4. Information bias
- 68. 68 Data Quality Assurance Assuring data quality is important to get valid research findings It can be assured through: Providing training for data collectors Supervision Pre-testing and pilot study Assigning appropriate and skilled personnel
- 69. 69 Pre-test and Pilot study Before the collection of data can be started, it is necessary to test the methods and to make various practical preparations. Pre-tests or pilot studies allow us to identify potential problems in the proposed study. One of assuring data quality A pre-test usually refers to a small-scale trial of a particular research component. A pilot study is the process of carrying out a preliminary study going through the entire research procedure with a small sample.
- 70. 70 Plan for data processing and analysis Data processing and analysis should start in the field with፡- Checking for completeness of the data Performing quality control checks Sorting the data by instrument used and by group of informants. Data of small samples may even be processed and analyzed as soon as it is collected.
- 71. 71 The plan for data processing and analysis must be made after careful consideration of the objectives of the study as well as of the tools developed to meet the objectives. Preparation of a plan for data processing and analysis will provide you with better insight into the feasibility of the analysis to be performed as well as the resources that are required.
- 72. 72 What Should the Plan Include? When making a plan for data processing and analysis the following issues should be considered: 1. Sorting data, 2. Performing quality-control checks, 3. Data processing, and 4. Data analysis
- 74. 74 Ethical Considerations Why do we need ethical approval? Before you embark on research with human subjects, you are likely to require ethical approval. Ethical decisions are based on three main approaches: duty, rights and goal-based.
- 75. 75 A. Goal-based approach: assumes that we should try to produce the greatest possible balance of value over disvalue. • Discomfort to one individual may be justified by the consequences for the society as a whole. B. Duty-based approach: your duty as a researcher is founded on your own moral principles. As a researcher, you will have a duty to yourself and to the individual who is participating in the research. The researcher should not lye or deceive his subjects for getting good research outcome. If she/he did it, it is unethical.
- 76. 76 C. Rights-based approach: the rights of the individual are assumed to be all-important. • Thus a subject‟s right to refuse must be upheld whatever the consequences for the research.
- 77. 77 • Research studies should be judged ethically on three sets of criteria: 1. Ethical principles 2. Ethical rules 3. Scientific criteria.
- 78. 78 Dissemination and Utilization of Results – Feedback to the community – Feedback to local authorities – Identify relevant agencies that need to be informed Scientific publication Presentation in meetings/conferences Briefly describe how the study results can be best translated into application
- 79. 79 Thank you!