Introduction to Biostatistics

Werkneh Melkie (BSc PH, MPH in Biostatistics)

E-mail - [email protected]

Department of public health

College of Medicine and Health Sciences

Debre Markos university, Ethiopia

11/27/2022 Werkneh M. 1
Table of content
Chapter one: Introduction to Biostatistics
• Basic statistical concepts
• Classification of statistics
• Types of variables
• Application and limitation of statistics
• Data types and measurement scales
• Data collection methods
• Questionnaire design and interviewing techniques
Chapter two: Methods of data, organization, presentation and summarization
• Tables and diagrams
• Measures of central tendency
• Measures of variation

11/27/2022 Werkneh M. 2
Table of content…

Chapter three : Demographic statistics

Chapter four: Introduction to probability

• Definition of basic terms in probability

• Types of probability

• Rules of probability and types of events

• Probability distributions : binomial and normal distributions

11/27/2022 Werkneh M. 3
 Chapter one: Introduction to Biostatistics
Objectives:
At the end of this section the student will be able to:-
• Define Statistics and Biostatistics
• Discuss the role of statistics in health sciences
• Understand the importance of statistics in health sciences
• Describe methods of data collection
• Understand common types of variables and measurement scales
• Know how to design a questionnaires

11/27/2022 Werkneh M. 4
Introduction to Biostatistics
 what is statistics?

 Is it important to all professionals?

 Have you heared about vital statistics?

 Is a statistics a tool or science or art?

 How does your society deal with uncertainties?

 How does science deal with uncertainties?

 What is biostatistics?

11/27/2022 Werkneh M. 5
Definition

 Statistics is not just a series of formulas, rather a way of thinking

 Important to all professionals who want to help & make the world a better place

 We see ‘‘vital statistics’’ in the newspaper, announcements of life events such as births,
marriages and deaths

11/27/2022 Werkneh M. 6
Uncertainties
Scenario; court procedures: A crime has been discovered and a suspect has been
identified. After a police investigation collect evidence against suspect, presecutor presents
summarized evidence to a jury. The jurors are given the rules regarding convicting beyond a
reasonable doubt and about unanimous decision, and then debate. After the debate, jurors vote
and verdict is reached: guilty or not guilty. Why do we need to have this time-consuming,
cost-consuming process of trial by jury? One reason is that the truth is often unknown, at least
uncertain. Perhaps only the suspect knows but he or she does not talk. It is uncertain because
of variability (every case is different) and because of possibly incomplete information. Trial
by jury is the way our society deals with uncertainties; its goal is to minimize mistakes.
11/27/2022 Werkneh M. 7
Definitions

 ASA defined Statistics as “the science of learning from data, and of measuring,
controlling and communicating uncertainty.”

 It is a field of study concerned with the collection, organization, summarization

and analysis of data and drawing of inferences about a body of data when only a
pare of data is observed.

 It is a methodological discipline, that exists not for itself, rather to offer other
fields of study a coherent set of ideas and tools for dealing with data

11/27/2022 Werkneh M. 8
Definitions …
 It is a methodology that scientists and mathematicians have developed for
collecting, analyzing, interpreting data and drawing conclusions from information
and
 The detailed planning that precedes all these activities
 Statistics is a way of thinking about numbers (collection, analysis, and
presentation) with emphasis on relating their interpretation and meaning to the
manner in which they are collected
 Two main areas
 Mathematical and
 Applied statistics

11/27/2022 Werkneh M. 9
Classifications of statistics
 Mathematical statistics
• Development of new methods of statistical inference
• Requires detailed knowledge of abstract mathematics
 Applied statistics
• Application of methods of mathematical statistics to specific area
 Statistical methods can be used to find answers to the questions like:
 What kind and how much data need to be collected?
 How should we organize and summarize the data?
 How can we analyse the data and draw conclusions from it?
 How can we assess the strength of the conclusions and evaluate their
uncertainty?

11/27/2022 Werkneh M. 10
Cont’d…
 The term statistics is used to mean either statistical data or statistical
methods.

 Statistical data: numerical descriptions of things.

• These descriptions may take form of counts or measurements.

NB: Even though statistical data always denote figures (numerical descriptions) it
must be remembered that all 'numerical descriptions' are not statistical data.

In order say numerical descriptions called statistics they must possess the
following characteristics:
11/27/2022 Werkneh M. 11
Characteristics of statistical data
i. Must be in aggregates – means statistics are 'number of facts.‘ A single fact,
even though numerically stated, cannot be called statistics.
ii. Must be affected to a marked extent by a multiplicity of causes.
iii. They must be enumerated or estimated according to a reasonable standard of
accuracy.
If the basis happens to be incorrect the results are bound to be misleading.
iv. They must have been collected in a systematic manner
Facts collected in an unsystematic manner and without a complete awareness
of the object, will be confusing and cannot be made the basis of valid
conclusions.

11/27/2022 Werkneh M. 12
Characteristics of statistical data…
v. They must be placed in relation to each other
 Numerical facts may be placed in relation to each other either in point
of time, space or condition
 That is, they must be comparable
Statistical Methods
 It refers to a body of methods that are used for collecting, organising,
analysing and interpreting numerical data for understanding a
phenomenon or making wise decisions.
 It is a branch of scientific method and helps us to know in a better way the
object under study.

11/27/2022 Werkneh M. 13
Statistical cycle

11/27/2022 Werkneh M. 14
Statistical cycle …
• The Problem step is about turning vague feelings into precise informational
goals, and specific questions that should be able to be answered using data
• Forming useful questions that can realistically be answered using statistical data
• Plan step is about deciding what people/objects/entities to obtain data on, what
things we should “measure,” and how we are going to do all of this.
• Data step is about obtaining the data, storing it, and “whipping it into shape”
(data wrangling and cleaning).
• Analysis step obtaining results through detailed examination of anything
complex
• Conclusions step, are about making sense of it all and then abstracting and
communicating what has been learned.
• In fact there is back-and-forth between the major steps
11/27/2022 Werkneh M. 15
Biostatistics
• Statistics is a very broad subject, with applications in a vast number of different
fields — business, education, psychology, agriculture, and economics, to mention
only a few.
• When the data analyzed are derived from the biological sciences and medicine,
we use the term biostatistics to distinguish this particular application of statistical
tools and concepts.
• Biostatistics is a branch of applied statistics concerned with the application of
statistical methods to medical and biological problem.
• However, in some circumstances standard methods may be modified by
biostatisticians.

11/27/2022 Werkneh M. 16
Rationale of Biostatistics
 Statistical method is used to explain and predict some of the health outcomes
and the direction of epidemics and pandemics, and
 It definitely influences decision-makers in public health.
 Guide public health and other healthcare practitioners on how to go about
controlling these diseases. 
 Biostatistics is a critical and invaluable tool in developing public health policy
and initiatives
 Scientific community rely on accurate and timely data to deal with outbreaks
of infectious diseases such as Ebola and COVID-19.
 From designing a trial to protocol development, biostatistics helps the clinical
research realm in various ways. The below-mentioned areas are to name a few.

11/27/2022 Werkneh M. 17
Relevance of biostatistics
 Program evaluation
 To create population based interventions
 Identifying barriers
 To determine the accuracy of measurements
 To compare measurement techniques
 To assess diagnostic tests, to determine normal values
 To estimate prognosis and to monitor patients
 Provide a way of organizing information
 Assessment of health status

11/27/2022 Werkneh M. 18
Relevance of biostatistics
 Resource allocation
 To determine the nature of information
 Magnitude of association: Strong vs weak association
 Assessing risk factors: Cause & effect relationship
 Evaluation of a new vaccine or drug
 Is the effect due to chance or some bias?
 Drawing of inferences: Information from sample to population

11/27/2022 Werkneh M. 19
Limitation of statistics

• It deals with only those subjects of inquiry that are capable of being
quantitatively measured and numerically expressed

• It deals on aggregates of facts and no importance is attached to

individual items- Suited only if their group characteristics are desired
to be studied

• Statistical data are only approximately and not mathematically

correct

11/27/2022 Werkneh M. 20
Types of Statistics
Depending on how data are used, statistics divided into two
1. Descriptive statistics
2. Inferential statistics
1. Descriptive statistics
 All about a set of information that has been collected only
 Consists of the collection, organization, summarization, and presentation of data.
 Focus on just mere descriptive, do not involve generalizing beyond the data at
hand.
 There are 3 main types of descriptive statistics:
The distribution concerns the frequency of each value.
The central tendency concerns the averages of the values.
The variability or dispersion concerns how spread out the values are.
11/27/2022 Werkneh M. 21
Types of Statistics…

2. Inferential statistics.

 Generalizing from our data to another set of cases or from samples to populations

 Determining relationships among variables, and making predictions.

 Uses probability, i.e., the chance of an event occurring.

It has two main uses:

 Making estimates about populations
 Testing hypotheses to draw conclusions about populations
11/27/2022 Werkneh M. 22
Definition of common statistical terms
Data
Numbers or non-numbers (i.e. qualities or attributes of something) which contain
information
 The raw material of statistics
 Derived/resulted from either measurement or count
 Thus the purpose of statistics is to investigate and evaluate the nature and meaning of
these data
Information
 Information is data that have been processed, organized and has meaning to the user.
 It is a collection of meaningful facts and figures that can be used as a basis for
guidance and decision-making.

11/27/2022 Werkneh M. 23
Definition of common …
Parameter: It is numerical expression of population measurements
Example
 Population mean (µ),
 Population variance,
 Population standard deviation, etc.
Statistic: It is numerical expression of sample measurements.
Example:
 Sample mean,
 Sample variance (s2 ) and
 Standard deviation (s).

11/27/2022 Werkneh M. 24
Definition of common …
 Variable: a characteristic that takes different values in different persons, places, or
things /not the same when observed in different possessors/
E.g. blood pressure, heart rate, height, weights, ages, Sex, temperature, etc.

 Can be classified as
• Categorical (Qualitative ) and
• Numeric variables (Quantitative)

Quantitative Variables - is a quantifiable characteristic whose values are numbers

(except numbers which are standing up as codes for categories)
 Can be ordered or ranked and convey information regarding amount
11/27/2022 Werkneh M. 25
Definition of common…
 Numeric variables may be either continuous or discrete.
• E.g. heights, weights, and body temperatures etc.
 Discrete variables –
• can be assigned values such as 0, 1, 2, 3 and are said to be countable.
 Continuous variables
• Assume an infinite number of values in an interval between any two
specific values.
• Obtained by measuring
• Often include fractions and decimals

11/27/2022 Werkneh M. 26
Definition of common…
Categorical variables
• characteristic that can’t be quantifiable.
• Variables that can be placed into distinct categories
• Can be either nominal or ordinal.
• E.g. gender (male or female), religion, residence etc.
• Data: are the values (measurements/observations) that the variables can assume .
• Random variables: variables whose values are determined by chance
• Data set: a collection of data values
• Data value or a datum - each value in the data set

11/27/2022 Werkneh M. 27
Summary of variable classification

Variables may be classified into two main categories: categorical and numeric. Each
category is then classified in two subcategories: nominal or ordinal for categorical
variables, discrete or continuous for numeric variables.
11/27/2022 Werkneh M. 28
Measurement/observations
 Assignment of numbers to objects or events according to a set of rules
 Different set of rules may produces various measurement scales
 Basis of all scientific inquiry and essential component of research methodology.
 A critical junction between scientific theory and application,
 It is a process through which researchers describe, explain, and predict
phenomenon and constructs of our daily existence
Measurement is important in research design in two critical areas.
1. Allows researchers to quantify abstract constructs and variables.
2. Level of statistical sophistication used to analyze data derived from a study is
directly dependent on the scale of measurement that quantify variables of interest.

11/27/2022 Werkneh M. 29
Measurement scales

 Although variables could be broadly divided into categorical and numeric,

 It has been a common practice to see four basic types of data (scales of
measurement) under these broader nonmetric & metric measurements
 Nominal Scale
 Ordinal scale
 Interval scale
 Ratio scale
Nominal and ordinal scales are nonmetric measurement scales.

11/27/2022 Werkneh M. 30
Nominal Scale
 Least sophisticated and lowest type of measurement
 It consists of naming and used only to qualitative data
 Classifies data into mutually exclusive, exhausting categories
 No order or ranking can be imposed in a quantitative sequence
 Used only to qualitatively classify or categorize not to quantify
 No absolute zero point
 Impossible to use or apply standard mathematical operations
 Purely descriptive and cannot be manipulated mathematically
 e.g. Sex( Female, male), Exam result (Pass, Fail), Blood Group (A,B, O or
AB), Color of Eyes (blue, green brown, black), marital status, etc

11/27/2022 Werkneh M. 31
Ordinal scale
 Build on nominal measurement. However,
 The ability to measure both identity and magnitude of variables makes it a
higher level of measurement
 Ordering of variables with some number representing more than another
 Information about relative position but not the interval between the ranks or
categories
 Qualitative in nature and lacks the mathematical properties necessary for
sophisticated statistical analyses.
 Way of thinking is using the concept of greater than or less than
NB: it is impossible to express the real difference between measurements in numerical
terms. .

11/27/2022 Werkneh M. 32
Ordinal scale…

 There are both category and rank. However, precise differences between
the ranks do not exist
 spaces or intervals b/n the categories are not necessarily equal
 Ordinal measurements tell you the direction of difference between two
individuals
Example
 Pain level (Mild, Moderate, Severe), Tumors (Stage 0-IV)
 Response to treatment (poor, fair, good)
 Severity of disease (mild, moderate, severe)
 Income status (low, middle, high)

11/27/2022 Werkneh M. 33
Metric measurement scales
• Interval and ratio scales are the two types of metric measurement scales,
• They are quantitative in nature.
• they represent the most sophisticated level of measurement and
• lend themselves well to sophisticated and powerful statistical techniques

11/27/2022 Werkneh M. 34
Interval scale
 Build on ordinal measurement by providing precise/absolute/ differences between units
and
 Identify direction & magnitude of a d/ce
 Numbers scaled at equal/same distances
 There is no meaningful/true zero, instead zero point is arbitrary
 Lack of an absolute zero point makes division and multiplication impossible
 Addition and subtraction are possible
• e.g. on either the Fahrenheit or Celsius scale, zero does not represent a complete absence of
temperature and
• the difference between 10 and 20 degrees is the same as the difference between 40 and 50
degrees. There might be a qualitative difference but the quantitative difference is identical—10
degrees
• But 30 º C is not twice as hot as 15 º C
11/27/2022 Werkneh M. 35
Ratio scale
 The second type of metric measurement scale
 Identical to the interval scale, except that they have an absolute zero point
 Data is presented in frequency distribution in logical order.
 Unlike with interval scale data, all mathematical operations are possible
 Highest level of measurement and allow for the use of sophisticated
statistical techniques.
 Ratio matters – has a natural (true)zero value. E.g. Salaries i.e. a value of
zero indicates none of the variable.
 Examples include money, height, weight, and time
 Ten dollars is 10 times more than 1 dollar, and 20 dollars is twice as much as 10
dollars.

11/27/2022 Werkneh M. 36
Cont’d…
 There is not complete agreement among statisticians about the classification of
data into one of the four categories.

 Both interval and ratio data involve measurement

 Most data a nalysis techniques that apply to ratio data also apply to interval data

 Therefore, in most practical aspects, these types of data (interval and ratio) are
grouped under metric data

 In some other instances, these type of data are also known as numerical discrete
and numerical continuous
11/27/2022 Werkneh M. 37
Data collection methods/techniques

Before any statistical work, data must be collected.

data can be used to describe situations or events.
 Data collection techniques allow us to systematically collect data about our
objects of study (people, objects, and phenomena) and about the setting in
which they occur.
In the collection of data we have to be systematic.
If data are collected haphazardly, it will be difficult to answer our research
questions in a conclusive way.

11/27/2022 Werkneh M. 38
Data collection methods…
 Data can be collected in a variety of ways and
 Factors influencing choice of data collection methods are
 Type of data/variable
 Objective of the study
 Availability of data
 Money
 Time
 Quality needed
 Characteristics of study participants and
Types of data
 Primary source
 Secondary source
11/27/2022 Werkneh M. 39
Sources of Data

1- Routinely kept records

 Hospital medical records, Hospital accounting records

2- External sources
 Published reports, commercially available data banks, or the research literature

3- Surveys – DHS

4- Experiments
Data collection methods/techniques
Methods of data collection
• Formal testing (psychological, educational, academic, intelligence)
• Observation
• Face-to-face and self-administered interviews
• Postal or mail method and telephone interviews
• Using available information
• Focus group discussions (FGD)
• Other data collection techniques – Rapid appraisal techniques, life histories, case
studies, etc.
11/27/2022 Werkneh M. 41
Observation
 Systematically selecting, watching and recoding behaviors of people or
phenomena
 Give additional, more accurate information on behavior than interviews or
questionnaires
 Can be used in any kind of research (quantitative, qualitative)
 It can be simple visual observations or
 high level using sophisticated machines and equipment such as biochemical, X-
ray machines, microscope, clinical and microbiological examinations.
 Efficient way when researchers studies & quantifies some type of behavior.
 There should be standardized outline/guidelines for the observations prior to
actual data collection to ensure adequate validity and interrater reliability

11/27/2022 Werkneh M. 42
Observation
• The oldest and most basic tool
• Data’: body language, facial expression, behavior, other non-verbal expressions,
movements, etc.
• Observation of behaviors, actions, activities and interactions helps in understanding
more than what people say about situations, and complex situations more fully.
• e.g. a researcher may be interested in studying cooperative behavior of university students in a
classroom

11/27/2022 Werkneh M. 43
Advantage

 Provides deep understanding and relatively more accurate data on behavior and
activities
 Permit researcher/evaluator to enter into and understand situation/context.
 Gives more detailed and context related information
 Allows to observe whether people do what they say they do; often used for
exploring deviant behavior
 Useful to access tacit knowledge of subjects - the subconscious knowledge that
they would not be able to verbalize in an interview setting
 Provide direct information about behavior of individuals and groups.

11/27/2022 Werkneh M. 44
Disadvantages
 Time consuming and expensive
 Observer bias: selective perception of observer may distort/affect the investigation.
 Ethical issues concerning confidentiality or privacy may arise.
 May affect behavior of participants
 Need well qualified, highly trained observers; may need to be content experts
during the use of high level machines and good skills in research and local
language;
 Time lag between observation and note taking is likely thus requires good memory
and ability to take note afterwards

11/27/2022 Werkneh M. 45
 Interview
• Oral questioning of respondents to tell us their story in their own ways
assuming their perspectives are meaningful, knowable and can affect the
success of the research/project.
• Relatively simple approach, but it can produce a wealth of information
• Relatively inexpensive and efficient way to collect a wide variety of data that
does not require formal testing.
• The effectiveness of an interview depends on how it is structured
• Interviewer should be trained properly to avoid variation of collected data
• Responses can be recorded by:
 Writing them down
 Tape-recording
 Combination of them
11/27/2022 Werkneh M. 46
Interview …
Interviews can be conducted with varying degree of flexibility
High degree of flexibility: when
 Studying sensitive issues (e.g. commercial sex)
 The researcher has little understanding of the problem
 Is frequently applied in exploratory studies
Low degree of flexibility: when
 Researcher is relatively knowledgeable about expected answers or
 The number of respondents being interviewed is relatively large
 Questionnaires may be used with a fixed list of questions in a standard
sequence, which have mainly fixed answers
11/27/2022 Werkneh M. 47
Types of interview
In-depth Interviews

 Dialogue between a skilled interviewer and an interviewee to elicit rich, detailed

material that can be used in analysis

 Best conducted face to face,

 The interviewer does not follow a rigid form rather limited set of open-ended
questions with extensive probing

 Seek to encourage free and open responses in their own words

 Interview Guide need to be prepared.

11/27/2022 Werkneh M. 48
Conducting an In-depth Interview

Interview Guide
Includes a list of questions or issues that are to be explored
Helps the interviewer to remember the points to cover
Suggests ways of approaching and talking about topics make interviewing more
systematic and comprehensive.
Reminds the interviewer about probes and ways of asking questions
Gives a possible order of topics; and
Helps the interviewer to enable people to talk in their own ways, and fully as
possible.
Includes an introduction and way of ending the interview.

11/27/2022 Werkneh M. 49
Conducting an In-depth …

When to use In-depth interviews

 Highly sensitive issues
 To get detailed information about a person's thoughts and
behaviors
 When peer pressure is expected to distort facts

11/27/2022 Werkneh M. 50
Cont’d…
Important points to remember
 Avoid variations in the interview setting
 Consent of the respondent should be obtained
 Knowledge gap between the interviewer and interviewee must be considered to facilitate
understanding
 Use a tape record and or a note taker who can assist
 Confidentiality of information must be maintained
 Receive the information accurately
 Information can be distorted by interviewer
• Fatigue
• Bias (expectation of answers),
• Preoccupation with taking notes and
• By technical languages (Jargons) foreign to the interviewee
11/27/2022 Werkneh M. 51
Tips for Effective Interviewing

 Follow up on what the participant says

 Ask questions when you don’t understand
 Ask to hear more about a subject
 Listen more, talk less, and ask real questions
 Ask open-ended questions
 Follow up, don’t interrupt
 Ask participants to tell a story
 Keep participants focused, and ask for details
 Avoid reinforcing your participants’ responses
 Use an interview guide carefully
 Tolerate silence
11/27/2022 Werkneh M. 52
Advantages
 Usually yield richest data, details, new insights;
 Permit face-to-face contact with the respondents;
 Provide opportunity to explore topics in depth;
 Permits clarification of questions.
 Affordability to experience the affective as well as cognitive aspects of responses
 Allow interviewer to explain or help clarify questions, increasing the likelihood of useful
responses.
 Suitable for use with both literates & illiterates.
 Allow the interviewer to be flexible in administrating interview to particular individuals
or circumstances.
11/27/2022 Werkneh M. 53
Disadvantages

 Expensive and time consuming

 Data obtained may not be appropriate for extensive statistical analysis
because they simply describe a construct rather than quantifying it
 Need well qualified, highly trained interviewers;
 Interviewee may distort information through recall error, selective
perceptions,
Presence of interviewer can influence, desire to please interviewer
Flexibility can result in inconsistencies across interviews
 Volume of information too large, may be difficult to transcribe & reduce data

11/27/2022 Werkneh M. 54
Telephone interview
 They are less costly
 People may be more candid in their opinions since there is no face
to-face contact.
 A major drawback to the telephone survey is that some
 People may not have phones or may not answer when the calls are
made. chance of being surveyed??
 Unlisted numbers and cell phones, so they cannot be surveyed.
 Tone of the voice of the interviewer might influence the response of
the person who is being interviewed.
 We may not get a rich and detailed responses
11/27/2022 Werkneh M. 55
Mail interview

 Cover a wider geographic area than telephone in-depth interviews

 Less expensive to conduct.
Disadvantage
 Low number of responses and inappropriate answers to questions.
 People may have difficulty reading or understanding the questions.

11/27/2022 Werkneh M. 56
Use of documentary sources/available information
Available information includes
Clinical and other personal records,
death certificates,
published mortality statistics,
census publications, etc. Examples include
Official publications of Central Statistical Authority
Publication of Ministry of Health and Other Ministries
News Papers and Journals.
International Publications like Publications by WHO, World Bank, UNICEF
Records of hospitals or any Health Institutions.
Use of key informants is important technique to gain access to available
information
11/27/2022 Werkneh M. 57
Advantages

 Ease of access

 Low cost or free/ collection is inexpensive

 Time-saving

 Allow you to generate new insights from previous analysis

 Longitudinal analysis

 Anyone can collect the data

 A huge amount of secondary data with a wide variety of sources

11/27/2022 Werkneh M. 58
Disadvantages

 Might be not specific to your needs

 You have no control over data quality

 Data may not always be complete and precise enough, or too disorganized

 Biasness

 Not timely – collected before and can be out - dated

 You are not the owner of the information

11/27/2022 Werkneh M. 59
Self-administered questionnaire

Written questions are presented that are to be answered by the respondents

in written form
A written questionnaire can b administered in different ways, such as
 Sending questionnaires by mail
 Gathering all or part of the respondents in one place at one time, giving
oral or written instructions, and letting them fill out the questionnaires
 Hand-delivering questionnaires to respondents and collecting them later

11/27/2022 Werkneh M. 60
Focus group discussion (FGD)
 FGD is a qualitative data collection method of approximately 6-10 people guided
by a facilitator and discuss an issue/research topic among themselves in depth
 People with something in common discussing an issue /Homogenous group
composition/
 It is focused Ideal size 6-10, maximum =12, minimum=4
 Far advanced and used more than in-depth interview, because
 Group interaction stimulate richer responses and emergence of new ideas
 Observation: The researcher observes and gets first hand insights
 Cost and timing: (FGD) can be done more quickly and generally less expensively
than series of depth interviews

11/27/2022 Werkneh M. 61
FGD…

It requires
 Relaxed, informal/natural setting
 Time duration: 1:30 - 2:00 hours (it depends)
 One leader (moderator) One assistant (or co-moderator)
Good communication skills
Records all the verbal and non-verbal message, in addition to note
FGD guide: tool used to guide the discussion, probe/follow up questions
When to use FGD
 Idea generation
 Problem identification and definition
 Evaluation of message concepts
 Program design
11/27/2022 Werkneh M. 62
FGD… Advantages
Produce a lot of information in less time and cost /Richness of data/
Excellent for obtaining information from illiterate communities
generally more acceptable among community
Explore variety of opinions or views within a group
Flexibility in covering topics/May uncover unanticipated ideas that are important/
Group synergy (group dynamics)
Broader understanding by providing well-grounded data on social and cultural norms
with respect to health and disease, prevention of disease
Ability to study special respondents
youth
Professionals (doctors, lawyers)
11/27/2022 Werkneh M. 63
Disadvantages
 Needs more time for preparations
 Requires good facilitation skills
 Lack of generalizability (small sample size)
 High selection bias
 Might be misused
 FGD is not a replacement for quantitative research
 Subject to Interpretation - subjective
 Cost-per-respondent is high (compared to survey)
 Results dependent on skill of moderator in running the group and analysis
(researcher)
11/27/2022 Werkneh M. 64
Good facilitation skills

Build rapport and open the discussion with a general idea

 Give participants a chance to think about the questions/do not lookdown on
respondents/
 Covering the material in the guide
 Do not correct participants during discussions/a moderator is not a
teacher/judge/
 Turn the group away from irrelevant areas.
 Shows flexibility; e.g order of guide
 Links ideas together
 Encourages participation from everyone

11/27/2022 Werkneh M. 65
Good facilitation skills…
Treating the participant as the expert?/ does not agree or disagree with what is
said, no right or wrong answers
Active listening; Pay attention to what participants say and follow up with
relevant questions and probes
Don’t lose control
Build rapport with and among participants; it starts from welcoming
 Prevent dominance
Skills Required for Moderator
 Observation
 Interpersonal
 Communication
 Interpretive
11/27/2022 Werkneh M. 66
FGD… challenges
 More time for preparations
 The moderator may disrupt group interaction
 Group members can affect each other “polarization”(two division)
 FGs will not work well for some sensitive topics due to self-disclosure
 Individual behaviors
 Criticizing/undermine others’ ideas
 Blockers
 Recognition seekers
 Withdrawing - no view
 Dominating/authority, “I know everything “expect to others to confirm
 Fear of consequences
 Hip pocket decision, the influence of those in power
11/27/2022 Werkneh M. 67
 Questionnaire Design
General Considerations
 With only a few exceptions, long questionnaires get less response than short
questionnaires.

 Ask only questions that directly address study goals

 Involve experts in the questionnaire design process

 Formulate a plan for statistical analysis during the design stage of the project

68
 Questionnaire design…

Points to be taken into account when we design a questionnaire

 Keep questions short and simple (KISS)
 should be unambiguous & not double barreled
 Use simple and direct language.
 wording of a question should be simple & to the point
 To reduce misunderstanding, make items brief
The best kinds of questions are those allow a pre-printed answer to be ticked
Leading questions shouldn’t be asked
The questions should fall into a logical sequence
 It should translated into local languages to be used
The last step in questionnaire design is to test the questionnaire with a small number of
interviews before conducting your main interviews – pre-test/pilot
69
 Types of questions
Open-ended Questions
 Permit free responses that can be recorded in the respondent’s own words.
 It is used in Facts with which the researcher is not very familiar
 Opinions, attitudes, and suggestions of informants, or Sensitive issues
Closed Questions
 Offer a list of possible options or answers from which the respondents must
choose.
 choices should be exhaustive and mutually exclusive
 Keep the number of options as few as possible.

70
Reading assignments

 Advantage and dis advantages of primary and secondary data

Advantage and dis advantages of open and closed ended questions

 Steps in designing a questionnaire

11/27/2022 Werkneh M. 71
 Thank you!!

11/27/2022 Werkneh M. 72