Data Collection biostats BJ revised 2021.ppt
- 1. 1 Overview of Data Collection Methods Dr B Jokomo
- 2. 2 Learning Objectives Describe various data collection techniques and state their uses and limitations Understand the benefits of using a combination of different data collection techniques State various sources of bias in data collection and ways of preventing bias Identify ethical issues involved in implementing research and ways of ensuring that study participants are not harmed
- 3. 3 Overview Data collection techniques systematically collect information about the subjects of study i.e. people, objects, phenomena and about the settings in which they occur. Very important: If this is done in a haphazard way it will be difficult to answer the research questions in a conclusive way
- 4. Quantitative and Qualitative methods •2 types of research methods •Qualitative •Quantitative 4
- 5. Types of research Quantitative research is concerned with questions about: how much? How many? How often? To what extent? Etc Quantitative – used to quantify the size, distribution and association of variables 5
- 6. Types of research Qualitative is concerned with finding the answers to questions which begin with: why? In what way? How? Why people behave the way they do Opinions and attitudes How people are affected by the events that go on around them Why certain occurrences have happened 6
- 7. Qualitative Qualitative - identify and explore variables that give insight into the nature and causes of certain problem e.g through use of Focus Group Discussions, Participant observation In-depth interviews 7
- 8. 8 Data Collection Methods Using Available Information Observing Interviewing Administering written Questionnaires
- 9. 9 Using Available Information This involves use of information collected by others, locating sources and retrieving the information • Service information system data, • Census data • Unpublished reports, • Publications Tool - checklist has to be designed paying special attention to the layout of the source documents from which data are to be extracted
- 10. 10 Advantages and Disadvantages of Existing Data Limitations Access to the data may be difficult Information may not always be complete and precise. Data from available source may be out of date Authenticity of information Limited variables available Variables inconsistently defined &recorded by different people at different times Advantages Inexpensive Quick although time often underestimated Allows historical comparison Data not influenced by project
- 11. 11 Observation Involves systematically selecting, watching and recording behavior and characteristics of humans, objects and phenomena Participant observation The observer or researcher takes part in the situation under investigation Non Participant Observation The observer or researcher watches the situation openly or concealed but does not participate NB Observation on objects e.g. measure weight, height, blood pressure
- 12. 12 Advantages and disadvantages of using the observation method Advantages Gives additional reliable information on behavior of people than interviews and questionnaires Checks on information collected especially on sensitive topics such as alcohol or drug use Can be a primary source of data Disadvantages Time consuming therefore useful in small scale studies Can be subjective- bias Participant observation – ethics?
- 13. 13 Interviews Involves oral questioning of respondents either individually or as a group. Can be Face to face Telephone Answers obtained can be tape recorded and transcribed • Key Informant interviews (understanding of problem – exploratory)
- 14. 14 Advantages and Disadvantages of Interviews Advantages They have a high degree of flexibility Excellent way of discovering subjective meanings and interpretations that people give to their expectations People’s responses are less influenced by their peers if conducted individually. They give access to events or activities that cannot be observed directly. No need for respondents literacy Disadvantages Social pressure on the respondent Time consuming & expensive Respondent may be feel inhibited on sensitive topics Interpersonal factors/ suspicion of respondent interfering with data collection Infringes on privacy
- 15. 15 Questionnaires Can be administered by the researcher or be self administered Two types of questions • Closed (Yes or NO) • Open ended (Allows for further elaboration)
- 16. 16 Advantages and Disadvantages of questionnaires Advantages Ensures privacy and anonymity Respondents are on their own therefore likely to give honest responses Easy to administer, reduces time and financial expenses Can reach a large group in short space of time Disadvantages No proportions of returns No control over data quality & form completion Lack clarification and further probing Responses are short and lack detail Introduction of researcher bias Not suitable for illiterates
- 17. Postal Questionnaires Can cover wide geographic region Limited to respondents with fixed contactable addresses Generally low response rate (<30%) Unsure who filled in the questionnaires 17
- 18. Telephone Surveys (Type of Individual interview) Can cover a wide area Exclusion of people with unlisted numbers or no phones (bias) Non verbal cues absent- affects communication Questionnaire has to be short- thereby limiting information Suspicion aroused when call received at home Call being ignored- unfamiliar number 18
- 19. 19 Differentiation between Data Collection Techniques and Tools Data collection techniques • Using available information • Observation • Interview • Questionnaires • FGD Data collection tools Checklist, data compilation forms (Case Record Forms) • Eyes and other senses, pen and paper, watches, scales microscope • Schedules,checklist, questionnaires,tape recoder • Questionnaires • tape recorder ,FGD Guide
- 20. Focus Group Discussion Qualitative method of data collection Group discussion of 6-12 people guided by a facilitator Group members talk freely and spontaneously about a certain topic More than a question and answer session or interaction Group members discuss the topic among themselves 20
- 21. Purposes of FGD Obtain in depth information on concepts, perceptions, and ideas of a group Develop relevant research hypotheses To focus research Exploring in greater depth the problem to be investigated and its possible causes Formulate appropriate questions Explore controversial topics 21
- 22. Strengths of FGDs Learning not just what the participants think, but why they think that way More natural and more flexible , since the researcher creates the discussion Implementation of FGDs is an iterative process- each FGD building on previous one- slighty elaborated or better focussed set of themes and discussions Powerful research tool providing valuable spontaneous information at a short time and low cost Can complement surveys or other qualitative and quantitative techniques 22
- 23. Limitations of FGDs Not to be used for quantitative purposes Testing of hypothesis or generalisation of findings which require more elaborate services Not advisable to use FGDs as a single tool.- combine with Key Informant Interviews and in depth interviews On very sensitive topics like sexual behaviour or LWHIV people may feel inhibited to express themselves openly 23
- 24. Triangulation Researchers often use combination of techniques , loosely structured interviews using open ended ?s - focus group discussions - observations These are termed qualitative research techniques. They produce qualitative information which is often recorded in narrative form. Use of different data collection techniques is called triangulation 24
- 25. Triangulation cont.. Qualitative Research Technique Involves the identification and exploration of a number of often related variables that give insight into the nature and causes of certain problems and into the consequences of those affected. Quantitative Research Techniques These are used to quantify the size, distribution, and association of certain variables in the study population. Both techniques are often used within a single study- (Mixed methods). 25
- 26. 26 Bias in Information collection Bias in information collection is a distortion that results in the information not being representative of the true situation Possible Sources of bias during data collection 1. Defective instruments Questionnaires with - Fixed or closed questions on topics about which too little is known
- 27. Bias cont.. - Open ended questions without guidelines on how to ask or answer - Vaguely phrased questions or - Questions asked in illogical order Weighing scales that are not standardised Prevention by carefully planning the data collection process and by pretesting the data collection tools. 27
- 28. 28 2. Observer Bias Can easily occur during observation or loosely structured interviews Data collector seeing or only hearing what they are interested in and missing information that is critical to the research
- 29. Prevention of Observer bias Prevention: • prepare guidelines for conducting loosely structured interviews • train data collectors • data collectors should work in pairs and discuss findings after the observation 29
- 30. 30 3. Effect of the Interview on the informant Possible factor in all interview situations Informant mistrusting intention of the interview, avoiding certain questions or giving misleading answers
- 31. Prevention - Adequately introduce the purpose of the study to informants - Take sufficient time to do the interview - Assure respondents that data collected will be confidential Awareness of potential biases can enable you to prevent them. If unsuccessful report honestly ways in which your data may be biased. 31
- 32. 32 Ethical Considerations Need to consider whether our research procedures likely to cause any physical or emotional harm Causes of harm –Violating informants rights to privacy by posing sensitive questions or by gaining access to records that may contain personal data –Observing the behaviour of informants without their being aware –Failing to observe or respect certain cultural values, traditions, or taboos
- 33. Prevention –Obtaining informed consent before the study or interview begins –Building rapport before discussing sensitive issues –Ensuring confidentiality of the data obtained 33