RESEARCH

METHODOLOGY
By
Mrs. J. Hibeene
Nurse lecturer
SAMPLING
METHODS
 Introduction
• In research methodology, sampling is
concerned with the selection of a subset of
individuals from within a population to
estimate characteristics of the whole
population
 Cont.
• Researchers’ rarely survey the entire population
because the cost may be too high.
• The 3 main advantages of sampling are that the cost is
lower, data collection is faster and since the data set is
smaller, it is possible to ensure homogeneity & to
improve the accuracy & quality of the data.
 Definitions
• Sampling is a selection of a number of study
units from a defined study population
• Study Population :- The group from which we
wish to make an estimate
• Study population can be people, objects, animal
species etc
 Cont
• Some studies may involve a small number of
people and all of them can be included
• These parts are called units & the list
of all units is called the frame
• To draw a Sample from the total population, it is
necessary to focus on a large population for the
inclusion of some of its members in the
investigation.
 WHAT IS INVOLVED IN SAMPLING?.
• Four main aspects are considered:
Study Population - What is the group from which
we want to draw a sample?
Sample size - How many people do we need in our
sample?
Sampling method(s) - How will these people be
selected?
 CONT.
 Representativeness/Generalization - Is the
sample drawn representative to that
population to draw conclusions that are valid
for the whole population?
 Study Population
• Study Population has to be clearly defined

• For example, it has to be defined according to

age, gender, residence, economic status,
education status or ethnicity.
• Apart from human population, a study
population can be a village, household,
institutions records, etc.
 CONT

• Each study population consists of study units.

• How study population and units are defined

depends on the research problem.
 For Example:
Problem Study Population Study Unit
•Malnutrition -All Children 6-24 mths -one child
related to weaning in District X between 6-24
in District X months

•High Drop-out rates - All primary schools in -one in primary

schools in district Y primary school in District Y
District Y
•Poor Record-keeping -All TB records - one TB record in
for TB Patients registered in Hospital Z hospital Z
in Hospital Z
 Representativeness
• Representativeness of a sample has all the
important characteristics of the population from
which it is drawn.
• For example: To interview 100 mothers to obtain
a complete picture of the weaning practices in
district X, you would select these mothers from a
representative sample of villages.
 Cont..

• Interviewing mothers from <5clinics only may

not be representative of mothers under
study
 Sampling Methods
• One important issue to consider influencing
the choice of the appropriate sampling
method is the sampling frame
• Sampling Frame: Is the listing of all the units
that compose the study population (i.e.,
population characteristics)
 Types of Sampling Methods
• Two general types of sampling methods are
used. They include:
i. Probability Sampling methods

ii. Non-Probability sampling methods

 Probability Sampling Methods
These comprise:
 Simple random sampling

 Systematic sampling

 Stratified sampling

 Cluster sampling

 Multistage sampling
 What is a Probability Sampling?
• A probability sampling is a random selection
procedure to ensure that each unit of the
sample is chosen on the basis of chance.
• All units of the study population should have
an equal or at least a known chance of being
included in the sample .
 1. Simple Random Sampling

• Probability sampling requires that a listing of

all study units exists or can be compiled. This
listing is called the sampling frame
• Simplest method of probability sampling.
 Simple Random Sampling
• It involves:
• Making a numbered list of all the units in the
population from which you want to draw a sample.
• Deciding on the size of the sample

• Selecting the required number of sampling units- using

a lottery method or table of random numbers.
 Cont..
• A simple random sample of 50 students is to
be selected from a school of 250 students.
• Using a list of all 250 students, each student is
given a number (1 to 250), and these numbers
are written on small pieces of papers.
 Cont..
• All the 250 papers are put in a box, after which
the box is shaken vigorously to ensure
randomization.
• Then, 50 papers are taken out of the box and
the numbers are recorded.
• The students belonging to these numbers will
constitute the sample.
 2. Systematic Sampling
• In systematic sampling, individuals are chosen
at regular intervals (for example every 5th
person) from the sampling frame.
• We randomly select a number to tell us
where to start selecting individuals from the
list.
 Cont..
• For example, a systematic sample is to be
selected from 1200 students of a school. The
sample size selected is 100.
• The sampling fraction is:
100 (Sample size) = 1
1200 (Study population) 12
• Sampling interval is, 12
 Cont..
• Systematic sampling is usually less time
consuming and easier to perform than simple
random sampling.
 3. Stratified Sampling
• Is the division of sampling frame into groups or strata,
according to characteristics, such as residents from
urban & rural areas; or different age-groups, or high
cost areas & low cost areas, or distance.
• Random or systematic samples of predetermined size
will then have to be obtained from each group
(stratum).
 CONT.

• Stratified sampling is only applied when

proportion of the study population belonging
to each group is known.
 4. Cluster Sampling
• Cluster sampling is a selection of groups of
study units.
• Clusters are often geographic units (e.g.
Provinces, Districts, Villages) or organizational
units (e.g., clinics, hospitals, training
institutions).
 5. Multistage Sampling
• Multistage sampling procedure is carried out in
stages or phases & usually involves more than one
sampling methods.
• E.g., You may decide by selecting a province, then
district, villages, households, up to your
respondents. All these may require different
sampling methods
 Advantages of Cluster & Multistage
Sampling
• A sampling frame of clusters is sufficient

• It is within the clusters that you need to list &

sample the individual units
• Sample is easier to select than a simple random
sample of similar size
• Applicable for cross-sectional descriptive
studies.
 Advantage & Disadvantages of
Sampling in General
. Disadvantage
• By taking a sample rather than studying the
whole population we lose precision.
 Cont..
Advantages
• However, there are usually compensating
advantages:
1. Saving in labour & costs in general
2. Investigations which require the use of trained
personnel may be improved by the ability to
provide a trained observer to make every
observation
 Non Probability Sampling
• Non Probability sampling involves no random
sampling applied.
• Types of Non probability sampling:

1. Convenience sampling

2. Purposive sampling

3. Quota sampling
 Convenience Sampling:
• Convenience sampling is a method in which
for convenience sake, the study units that
happen to be available at the time of data
collection are selected in the sample.
• It does not produce representative findings
 Purposive Sampling:
• The principle of selection in purposive sampling
is the researcher’s judgment as to typicality or
interest
• It can be as a result of the prevailing conditions
affecting the population that the researcher
would be interested in as rationale for the
study.
 Quota Sampling
• Quota sampling is a method that ensures that
a certain number of sample units from
different categories with specific
characteristics appear in the sample so that all
these characteristics are represented.
 Cont..
• For example, if the researcher is interested in
economic status of respondents, he/or she
may include more units of measurement, and
therefore would need to increase the sample
by extending to 3 or 4 days over to obtain
desired sample
DATA COLLECTION
METHODS
 GENERAL OBJECTIVE
• By the end of the session, students
will gain knowledge in data
collection methods.
 SPECIFIC OBJECTIVES
1. Define Data collection
2. Outline guidelines for implementing a data
collection method
3. Discuss the following data collection
methods;
i. Self-reports
ii. Interview methods
iii. Observational methods
 DEFINITION
• Data collection is a systematic
process of gathering observations or
measurements to find a solution to a
research problem.
 DATA COLLECTION PROCESS
• When planning the data collection
process, the researcher is guided
by 5 important questions;
[Link]?
[Link]?
[Link]?
[Link]?
[Link]?
WHAT DATA WILL BE COLLECTED?
• Researcher must carefully consider
what information is needed to
answer the research question.
• The variables under study should be
clearly defined
• Determine the level of measurement
or measurement scale to be used
 LEVEL OF MEASUREMENT
NOMINAL SCALES
•Used when persons, events or phenomena are
separated into mutually exclusive categories e.g.
married or single, divorced or widowed, dead or
alive.
ORDINAL SCALE
•Used for variables that can be ordered or ranked
e.g. 1 plus, 2 plus or 3 plus pitting oedema; and
slight, moderate or intense pain.
 LEVEL OF MEASUREMENT CONT….
INTERVAL SCALES
•Variables are assigned real numbers which are
categorized and ordered with equal measurement
between each category. E.g. Temperature readings
on a thermometer.
•Natural zero point does not exist
RATIO SCALES
•Data is categorized and ranked. The difference
between ranks can be specified and a true or natural
zero point can be identified. E.g. Pain scale, level of
knowledge
 HOW WILL DATA BE COLLECTED?
• Researcher must use an instrument to
gather data
• The manner in which the data is
captured should be reliable
• Field workers should know how to use
the data collection instruments
• Raw data need to be stored in a safe
place
 WHO WILL COLLECT THE DATA?
• Teams of researchers usually collect data
• People outside the research team may
also be used to collect data
• It is important to ensure that data is collected in
the same manner, whenever more than one
person is involved to ensure inter rater
reliability
• Data collectors need training and the reliability
of the collected data should be checked
 WHERE WILL THE DATA BE
COLLECTED?
• The setting for data collection must
be carefully determined
• It could take place in a controlled
laboratory, clinic, a home,
community center and so on.
 WHEN WILL THE DATA BE
COLLECTED?
• The researcher must decide exactly
when the data will be collected and
how long the process will last.
• Can be determined by doing a pilot
study
 SELF-REPORT TECHNIQUES
• Self- Reports are methods of gathering
data where participants provide
information about themselves without
interference from the researcher.

• Self-report methods rely on

the honesty and self-awareness of the
participant hence research findings
may be flawed with ‘self-report bias’.
 UTILIZATION OF SELF-REPORT
TECHNIQUES
• Self-report techniques are used for collecting
data about what people believe, think or
know
• In addition to collecting factual information
about the participants, the purpose of the
questions is to discover their thoughts,
perceptions, attitudes, beliefs, motives, plans,
experiences, knowledge levels and memories
• The type of self-report instrument chosen
depends on the research objectives and
include questionnaires, scales and interviews
 QUESTIONNAIRES
• In a questionnaire process, the
respondent who is the unit of
analysis, writes down their answers
in response to questions in a printed
document
 ADVANTAGES OF QUESTIONNAIRES
• Quick way of obtaining data from a large group
of people
• Less expensive in terms of time and money
• One of the easiest research instruments for
testing validity and reliability
• Participants feel a greater sense of anonymity
and are more likely to provide honest answers
• Standard format for all participants
 DISADVANTAGES OF
• QUESTIONNAIRES
Mailing questionnaires may be expensive
• Response rate may be low
• Respondents may provide socially acceptable
answers
• Respondents may fail to provide answers to some
of the questions
• There is no opportunity to clarify questions which
may be misunderstood by respondents
• Participants must be literate
• Participants who respond may be not be representative
of the population
 INTERVIEWS
• When using interview method , responses are
obtained from participants in face-to-face
encounters, through telephonic conversations
or by electronic means
• Interviews are often used in exploratory,
descriptive research and in case studies
• Interviews are the most direct method for
obtaining facts from interviewees
• They are useful in ascertaining values,
preferences, interests, tasks, attitudes, beliefs
and experiences.
 STRUCTURED INTERVIEWS
• Formalized so that all interviewees hear the
same questions, in the same order and in the
same manner
• Most appropriate when straightforward, factual
information is desired
• The instrument used is an interview schedule
• The interview schedule uses closed ended or
fixed alternative questions
• It must be presented to each interviewee in
exactly the same manner
 UNSTRUCTURED INTERVIEWS
• The unstructured interview follows the research focus
• The wording and organization of questions and sometimes
even the topic is left to the interviewer’s discretion
• The interviews are conducted in a conversational manner,
but with a purpose
• They are particularly useful for exploratory or qualitative
research, where the researcher cannot structure the
questions before data collection takes place
• The interviewer may begin with a broad question, then
depending on the interviewee’s replies, researcher invites
them to add more information to clarify the initial
response
 SEMI STRUCTURED INTERVIEW
• The interviewer must ask a specified
number of questions, but can also
pose additional ones
• Both closed-ended and open-ended
questions are included
 FOCUS GROUP INTERVIEWS
• Includes groups of about 5 to 12 people
whose opinions and experiences are
requested simultaneously
• The method allows participants to share their
thoughts with one another, generate new
ideas and consider a range of views before
answering
 Cont’d
• Focus groups are particularly useful in
participatory and action research, where
members of the community are equal
participants in planning and implementation
• The disadvantage is that some people are
uncomfortable when asked to participate in
group discussions
 ADVANTAGES OF INTERVIEWS
• The participant need not be able to read or write
• Responses can be obtained from a wide range
of participants- almost all segments of the
population
• Responses and retention rate is high
• Non-verbal behavior and mannerisms can be
observed
• Questions may be clarified if they are
misunderstood
• In-depth responses can be obtained
DISADVANTAGES OF INTERVIEWS
• Training programmes are needed for interviewers
• Interviews can be time consuming and expensive
• Arranging interviews may be difficulty
• Participants may provide socially acceptable answers
• Participants may be anxious because answers are
being recorded
• Participants may be influenced by interviewer
characteristics
• Interviewers may misinterpret non-verbal behaviour
 OBSERVATIONAL METHODS
• Observational method is the use of
unstructured and structured observation
to measure a study variable
• Measurement by observation is most
commonly performed in qualitative
research and is useful for research
involving children
 Cont’d
• To use observational method, the
researcher needs to decide what is to be
observed and then determine how to
ensure that every variable is observed in
a similar manner in each instance
• Observational measurement tends to be
more subjective than other types of
measurement
 STRUCTURED OBSERVATION
• Structured observation utilizes
category systems for organizing and
sorting the behavior being observed
• First carefully determine what is to be
observed
• Next, decide how the observations are
to be made, recorded and coded.
 CATEGORY SYSTEMS
• Observational categories should be
mutually exclusive
• Overlapping categories will lead to
inconsistent data and should be
avoided
 CHECKLISTS
• Checklists are techniques used to indicate whether
the behavior occurred
• Tally marks are generally placed on the data
collection form each time the behavior is observed
• Behavior other than that on the checklist is ignored
• Multiple tally marks may be placed in various
categories while the researcher is observing a
particular event. However in other studies, the
observer is required to select a single category
• Checklists can also be used as self-report
instruments
 SCALES
• Scales are self-report data collection
instruments that ask respondents to record
their attitudes or feelings on a continuum
• A scale contains a set of numbers, letters or
symbols that have rules and which can be
used to locate individuals on a continuum
• Commonly used scales include; semantic,
Visual analogue, Likert and rating
 LIKERT SCALE
• Likert scale is an example of a summated
rating scale, which is frequently used to test
attitudes or feelings
• It is summative in that item scores are
added to obtain the final result
• It consists of a number of declarative
statements about a topic, with five or seven
responses for each statement, ranging from
‘strongly agree’ to ‘strongly disagree’
 EXAMPLE OF LIKERT SCALE

State Healthcare professionals should practice

ment therapeutic touch in patient care
1 situations

Strongly Disagre Uncert Agree Strongl

disagree e ain y agree

1 2 3 4 5
 LIKERT SCALE CONT….
• An approximate number of positively
worded items and negatively worded items
should be included in a Likert instrument
• To score it, responses of all items are added
to obtain a score . The values obtained are
treated as interval data
• Negatively worded items are often reverse
scored (coded), in which case ‘strongly
disagree’ is given a score of 5 on a 5 point
Likert scale, while ‘strongly agree’ is given a
score of 1.
 RATING SCALES
• Rating scales allows the observer to rate
the behavior or event on a scale.
• Provides more information for analysis
than does the use of dichotomous data,
which indicate only that the behavior
either occurred or did not occur.
 PHYSIOLOGICAL MEASURES
• Physiological measures use biometric
parameters and are commonly utilized in
clinical health science research
• The most familiar include; blood
pressure values, blood values, urine
values and electrocardiograms
• Advantages of physiological measures
include precision and accuracy
 RECORDS AND AVAILABLE DATA
• A researcher makes use of already available
data such as Hospital records, admission
charts, incident reports, care plans etc
• Records serve as an economic source of
information
• They permit an examination of trends over
time, and eliminate the need to seek
cooperation from participants
 SOURCES OF ERRORS IN USE OF
RECORDS
• Records may contain institutional biases
• Facts may be distorted
• Some facts may be omitted
• Record keeping may be erratic
• The collection of data may have been
stopped for political or financial reasons
• Some data may not be readily available
owing to their confidential nature
RESEARCH DESIGN
 GENERAL OBJECTIVE
• By the end of the session, students
will gain knowledge on research
design.
 SPECIFIC OBJECTIVES
• Define Research Design
• State the characteristics of a good
research design
• Describe the types of research
design
 DEFINITION
• “ A research design is a blueprint for
conducting the study that maximizes
control over factors that could interfere
with validity of the findings” (Burns et
al, 2005)
• Or, it is a strategy used by the
researcher to answer a research
question
 PURPOSE OF RESEARCH DESIGN
• The purpose of research design is to
obtain accurate responses to
objectives, questions or hypothesis
 CHARACTERISTICS OF A GOOD
STUDY DESIGN
• 1. Appropriate to the purpose of the
study
• 2. Feasible, given the realistic
constraints
• 3. Effective in reducing threats to
validity
 VALIDITY
• Study validity is the measure of the
truth or accuracy of a claim
• In ensuring validity, the study should
be designed well enough to provide a
valid test of proposition or an
approximate truth or falsity of the
proposition
• Validity provides a major basis for making
decisions about which findings are sufficiently
valid to add to the evidence base for patient care
 TYPES OF VALIDITY
• Statistical conclusion validity is
concerned with whether the
conclusions about relationships or
differences drawn from statistical
analysis are an accurate reflection of
the real world
• Internal validity is the extent to which
the effects detected in the study are a
true reflection of reality rather than
the result of extraneous variables
 TYPES OF VALIDITY (CONT..)
• Construct validity examines whether the
instrument measures the theoretical
constructs it purports to measure
• External Validity is concerned with the
extent to which study findings can be
generalized beyond the sample used in
the study
 FACTORS THAT DETERMINE
CHOICE OF RESEARCH DESIGN
– Type of problem
Research question or hypothesis
Study purpose
– Knowledge already available about the
problem.
– Resources available for the study
 BROAD CLASSIFICATIONS OF
RESEARCH DESIGN
Quantitative Research
– Research strategy that focuses on
quantification of data
Qualitative Research Design
– Collection and analysis of non-numeric
data to understand concepts, opinions
or experiences
 DIFFERENCES BETWEEN
QUANTITATIVE AND QUALITATIVE
RESEARCH DESIGN
Quantitative Qualitative
• One reality Several realities
• Numbers Words
• Hypothesis testing Hypothesis
generating
• Generalizing Limited generalizing
• Larger sample Small sample
 DIFFERENCES BETWEEN QUANTITATIVE
AND QUALITATIVE RESEARCH DESIGN
(CONT..)
Quantitative Qualitative
• Validity
Trustworthiness
• Objective Subjective
• Outcome oriented Process oriented
• Fragmented Holistic
 QUANTITATIVE RESEARCH DESIGNS

CATEGORIES OF QUANTITATIVE RESEARCH

[Link] or interventional study-
researcher actively introduces an intervention
or treatment
[Link]-experimental or non-interventional
study- researcher collects information without
introducing treatment or making changes
 EXPERIMENTAL DESIGNS
• TRUE EXPERIMENTAL DESIGNS
 These are interventional studies that meet the
following criteria;
Manipulation- the research controls the
independent variable to determine the
causal effect on the dependent variable
Control- strict rules imposed by the
researcher in the experiment process to
increase the probability that the study’s
findings are an accurate reflection of reality
 Cont’d

Randomization- researcher assigns

participants to experimental or control
groups randomly
QUASI EXPERIMENTAL DESIGNS

• In quasi experimental design, the rules

for a true experiment are relaxed
• For example researcher may omit a
control group; in this case the same
sample serves as both a control and
intervention group
• the researcher may omit
randomization if a control group is
used
 TIME SERIES DESIGNS
• Researcher collects data on the
dependent variable from the
experimental group at set intervals; both
before and after introduction of the
independent variable for comparison
• No control group is used for comparison;
participants act as their own control
 LIMITATIONS OF EXPERIMENTAL
DESIGNS
• Researcher cannot manipulate some variables
E.G. Health history, gender or age of study
participants
• Ethical considerations prohibit manipulation of
some variables for example withholding
treatment or exposing patient to dangerous
situations
• Experimental designs are costly
• High threat to internal and external validity
due to many sources of error and bias in the
experimental process
 TYPES OF NON-EXPERIMENTAL
DESIGNS
DESCRIPTIVE DESIGNS
•Exploration and description of phenomena in real
life situations
•Carried out when more information is required in
a particular field
•Variables are described in order to answer the
research question but there is no intention of
establishing a cause-effect relationship
•May be used to identify problems with current
practice or justify current practice
 LONGITUDINAL STUDIES
• Descriptive study in which the researcher
collects data at several points in time
from the same participants
• Longitudinal studies can be long term
and continue for many years making
them expensive and demanding in terms
of participation
 CROSS SECTIONAL STUDIES
• Descriptive studies in which data is
collected and examined at one point
in time
• Data is collected from different
participants and comparison is made
across groups
 CORRELATIONAL DESIGNS
• These are also known as ‘ex post facto’ or
‘after the fact’ design
• The purpose of this design is to determine
existing relationships between variables
• When a correlation exists, a change in one
variable corresponds to a change in others
• Researcher does not manipulate the
independent variable because the dependent
variable has already occurred
RETROSPECTIVE CORRELATIONAL
DESIGNS
• These are also referred to as case control
studies
• Researcher starts with an effect and works
backwards to determine and describe the
relationships which exists among the variables
• These studies are usually conducted to
determine factors which may have been a
cause of a disease
• Two groups are selected; cases (those with the
disease of interest) and controls (those without
the disease)
 PROSPECTIVE CORRELATIONAL
DESIGN
• This type of study is also known as cohort
study
• The researcher selects a population and
follows it over time to determine outcomes
of exposure to a naturally occurring event
• A sample of subjects is observed where
those exposed and not exposed are
compared in terms of the outcome
TYPES OF QUALITATIVE RESEARCH
DESIGNS
PHENOMENOLOGY STUDIES
•Study of phenomena by examining human
experience through descriptions provided by
the individual (lived experience)
•The purpose of phenomenology research is
to describe what people experience with
regard to phenomena and how they interpret
these experiences
•Phenomenologists view the person as
integral to the environment
 ETHNOGRAPHY
• Qualitative research approach which grew
out of anthropology and the study of culture
and customs of people
• The focus is the social and cultural world of
certain groups of people
• Ethnography is used to study entire cultural
groups, cultural sub-group in a broad sense
or it may only apply to the study of a subunit
 GROUNDED THEORY
• Grounded theory is an inductive
research approach in which theory
emerges from patterns observed in
data.
• The purpose of grounded theory is to
build theory and illuminate the area
of study
 REFERENCES
• Brink H., Van Der Walt C., and Rensburg G.
2018. fundamentals of Research Methodology
for Health Care Professionals 4th ed. Cape
town: Juta and Company
• Burns N. and Grove K. S. 2005. The Practice of
Nursing Research: conduct, critique and
utilization 5th ed. St Louis, Missouri: Elsevier
Saunders
PRE-TESTING
&
PILOT STUDY
 Definition

Pre-testing: A small-scale trial or testing of a

particular research component, e.g a research tool
or a questionnaire.

Pilot study: Process of carrying out a preliminary

study or testing the whole methodology of the study

Going through the entire research procedure with

a small sample.
Pre-testing and Pilot Study
Why do we carry out a pretest or pilot study?
-To identify potential problems in the proposed
study.
-To revise the methods and logistics of data
collection before starting the actual fieldwork.
 Cont’d

• A pretest is simpler and less time consuming

and costly than conducting an entire pilot
study.
• Most studies prefer a pre-testing than pilot
study.
Pre-testing the Methodology
What aspects of research methodology can be evaluated
during pre-testing?

1. Reactions of respondents to research procedures

can be observed to determine:
• Availability of the study population
• Acceptability of the methods used to establish contact
with the study population
• Acceptability of questions and willingness to respond
or answer to questions.
 Pre-testing
2. Data –collection tools can be pre-
tested to determine:
•Whether tools allow to collect
reliable information and suitable for
analysis
•Time spent for collecting
 Cont’d
• Whether there is need to revise the
format or presentation of questionnaires
in relation to:
- Logical sequence of questions
- Clear wording of questions
- Translations are accurate, &
- Space for answers is sufficient
 Pre-testing
•Need for pre-categorization and adjustment for the
coding system
•Need for additional instructions
3. Determine sampling procedures and activities of research
assistants

When do we carry out a pretest?

•It can be conducted during the workshop or before the main
study to allow alterations to be made on the tools.
QUANTITATIVE
DATA
ANALYSIS
 GENERAL OBJECTIVE
• By the end of the session, students
will acquire basic knowledge and
skills in quantitative data analysis
 SPECIFIC OBJECTIVES
1. Define data analysis
2. State factors to consider when choosing
appropriate quantitative data analysis
technique
3. Explain descriptive statistics
Describe frequency distributions
State measures of central tendency
State the measures of variability
Describe the normal distribution
 INTRODUCTION
• Once the researcher has collected data,
it is important to organize, analyze and
interpret the data in order to give it
meaning and answer the research
question
• The researcher should be familiar with
the appropriate data analysis technique
that suits the research question and
study design
 DEFINITION
• Data analysis is a systematic process
of sorting, organizing and
interpreting research findings using
scientific methods in order to draw
meaning and answer the research
question.
 FACTORS TO CONSIDER WHEN
CHOOSING AN APPROPRIATE
DATA ANALYSIS TECHNIQUE
• When analyzing quantitative data, the
following factors should be considered;
Research question
Aim/objectives of the study
Study design
Level of measurement
 DESCRIPTIVE STATISTICS
• Type of statistics used to explain and
summarize data
• It indicates how the data set looks
 FREQUENCY DISTRIBUTIONS
• A frequency is the number of times
an event, score, or result is
observed
• Each score can be listed individually
or grouped together with similar
scores
• For grouped data, classes should be
mutually exclusive
Frequency distribution of Pain Intensity for
Patients following traditional and new
analgesic treatments
Pain Intensity Experimental Control group
group (traditional) f
(analgesic) f
1 2 0
2 3 2
3 10 3
4 3 10
5 2 5

n = 20 n = 20
Grouped frequency distribution of
patients’ weight in a given ward
Weight Frequency (f)
96-100 3
91-95 4
86-90 5
81-85 9
76-80 13
71-75 12
66-70 4

n = 50
MEASURES OF CENTRAL TENDENCY
• Measures of central tendency are
statistics or numbers expressing the
most typical scores in a distribution
• The following are the measures of
central tendency;
Mean
Median
mode
MEAN
Mean obtained from pain
assessment
 MEDIAN
• Median is the midpoint score or value in a
group of data ranked from lowest to highest
• Median is the best value of central tendency
for ordinal data
• For example calculate the median for the following
pain scores; 4, 8, 7, 5, 6, 3, 2
• First reorder the data as follows; 2, 3, 4, 5, 6,
7, 8
• Therefore, the median is 5
 MODE

• The mode is the value or score which

occurs most frequently
• If the distribution only has one mode it is
called unimodal, two modes is bimodal
and more than two modes is multimodal
• For example; find the mode from the
following pain scores; 10, 7, 5, 6, 8, 7, 3,
 MODE (CONT…)
• The mode can be used with every
level of measurement and is the
most appropriate measure for
nominal data
• It is not the best measure of central
tendency
 MEASURES OF VARIABILITY
• Measures of variability describe
how widespread values or scores
are in a distribution
• The three statistics used to
describe variability are;
Range
Variance
Standard deviation
 RANGE
• The range is the difference between the
smallest and largest value in a
distribution
• What is the range from the following
pain score distribution; 10, 7, 5, 6, 8, 7,
3,
• Formula equals largest value minus
smallest value
• 10-3 = 7
 RANGE (CONT…)
• Range can be used with ordinal,
interval and ratio data
• The usefulness of the range is limited
because one extreme score can
drastically change the range
 VARIANCE
• Variance is the sum of squared
deviations about the mean, divided
by the total number of values
• It is an intermediate value used in
calculating the standard deviation
 STANDARD DEVIATION
• The standard deviation indicates
how values vary about the mean in a
distribution and is defined as the
square root of the variance
• The greater the standard deviation,
the more spread scores are in
relation to the mean
 THE NORMAL DISTRIBUTION
• The normal curve or normal distribution
represents a theoretical distribution of
population scores
• Normal distribution variables are frequently
found in biological, behavioral and clinical
sciences
• Variables such as blood pressure, height and
weight are normally distributed in the
population
 THE NORMAL DISTRIBUTION
(CONT….)
• A large number of data set is needed to
approximate the normal curve; at least
30
• A normal distribution is bell shaped and
symmetrical with maximum height at the
mean
• A few values occur on both extremes
 ASSUMPTIONS OF NORMAL
DISTRIBUTION
• The mean, median and mode are the
same
• Approximately 68.% of all observations
are located between the mean and one
standard deviation
• Approximately 95.% are located within
two standard deviations of the mean on
either side
The Normal Distribution curve
Sample data set for age normally
distributed
 CONCLUSION
• Data analysis enables the researcher to
establish meanings and draw conclusions
• To summarize quantitative date,
descriptive statistics are used
• Measures of central tendency are the
mean, median and mode
• Measures of variability include the
range, variance and standard deviation
END