Research methods

3/28/2015
Course Information
 Course Outline: QMS 225 course
content.doc
 Teaching Plan: Semester Teaching Plan
QMS 225.doc

3/28/2015
 Search for knowledge,
 A scientific and systematic
search for pertinent
information on a specific
topic (Kothari, 2009)
 Form of disciplined inquiry
that generates knowledge
(2008)
 A term used generously
for any kind of inquiry that
is intended to find out
interesting or new facts!
 Search for knowledge,
 A scientific and systematic
search for pertinent
information on a specific
topic (Kothari, 2009)
 Form of disciplined inquiry
that generates knowledge
(2008)
 A term used generously
for any kind of inquiry that
is intended to find out
interesting or new facts!
 In order to discover
answers to questions
through the application of
scientific procedures!
 To find out the truth which
is hidden and which has
not been discovered yet
 In order to discover
answers to questions
through the application of
scientific procedures!
 To find out the truth which
is hidden and which has
not been discovered yet

3/28/2015
 Provides the basis for nearly all government
policies in our economic system
 Helps solving various operational and
planning problems of business and industry
 Helps in studying social relationships and in
seeking answers to various social problems
 Is a fountain of knowledge for the sake of
knowledge and an important source of
providing guidelines for solving different
business, governmental and social problems.
 Provides the basis for nearly all government
policies in our economic system
 Helps solving various operational and
planning problems of business and industry
 Helps in studying social relationships and in
seeking answers to various social problems
 Is a fountain of knowledge for the sake of
knowledge and an important source of
providing guidelines for solving different
business, governmental and social problems.

3/28/2015
 Theory essentially means explanation of
certain social phenomena. Theorizing,
thus means formulating an explanation
 Is a systematic set of interrelated
statements that intends to explain some
aspect of social life
 Theories emanate from a paradigm
 Paradigm (perspectives, world view)

3/28/2015
Theory as paradigm (perspective)
 That which underpins research design
Theory as a ‘lens’
 That which may inform our understanding
of the phenomenon under investigation
Theory as new knowledge
 That which may emerge from our study
Theory as paradigm (perspective)
 That which underpins research design
Theory as a ‘lens’
 That which may inform our understanding
of the phenomenon under investigation
Theory as new knowledge
 That which may emerge from our study

3/28/2015
ObservationObservation
PatternPattern
Tentative
Hypothesis
Tentative
Hypothesis
TheoryTheory

3/28/2015
TheoryTheory
HypothesisHypothesis
ConfirmationConfirmation

3/28/2015
Research Process
Formulate a
Question
Select an Appropriate
Research Design
Collect
Data
Interpret
Findings
Disseminate
Review the
Literature

3/28/2015
 Research topic selection
 Formulating the research
problem
 Acquiring knowledge on
current theories and
researches- LR
 Identifying and labeling
the variables
 Defining concepts and
establishing operational
definitions
 Formulating hypothesis
(guides investigation)
 Research topic selection
 Formulating the research
problem
 Acquiring knowledge on
current theories and
researches- LR
 Identifying and labeling
the variables
 Defining concepts and
establishing operational
definitions
 Formulating hypothesis
(guides investigation)
 Selection of the
appropriate research
design
 Description of the sample
and sample size
 Sampling procedures and
techniques
 Selection of the
appropriate research
design
 Description of the sample
and sample size
 Sampling procedures and
techniques

3/28/2015
 Construction of the
data collection
instruments:
questionnaires ,
interview guides etc.
 Actual field work
 Construction of the
data collection
instruments:
questionnaires ,
interview guides etc.
 Actual field work
 Data processing:
sorting, coding and
entering into a data
analysis software
 Performing statistical
analysis
 Simple descriptions
and frequency tables,
inferential etc.
 Data processing:
sorting, coding and
entering into a data
analysis software
 Performing statistical
analysis
 Simple descriptions
and frequency tables,
inferential etc.

3/28/2015
Continues……
 Interpretation of the results
 What comes out from the processed data-
information!!
 Report writing
 Communicating the whole work done to
targeted audience
 Dissemination of the results
 Sharing of the research outcomes
 Interpretation of the results
 What comes out from the processed data-
information!!
 Report writing
 Communicating the whole work done to
targeted audience
 Dissemination of the results
 Sharing of the research outcomes

3/28/2015
 Is the arrangement of conditions for collection and analysis
of data in a manner that aims to combine relevance to the
research purpose
 Is merely a conceptual structure within which research is
conducted, it constitutes the blueprint for collection,
measurement and analysis of data (Kothari, 2009)
 The design decision happen in respect of the following
questions:
 What is the study about? Where will it be carried out? What
data do we need? What will be the sample design? What
techniques of data collection? How will the collected data
be analyzed? Inter- alia
 Is the arrangement of conditions for collection and analysis
of data in a manner that aims to combine relevance to the
research purpose
 Is merely a conceptual structure within which research is
conducted, it constitutes the blueprint for collection,
measurement and analysis of data (Kothari, 2009)
 The design decision happen in respect of the following
questions:
 What is the study about? Where will it be carried out? What
data do we need? What will be the sample design? What
techniques of data collection? How will the collected data
be analyzed? Inter- alia

3/28/2015
 Ensures reliability of the results
 Various research operations are
smoothened thereby making research as
efficient as possible yielding maximal
information with minimal expenditure of
efforts, time and money
 Caution!!
 To be done with care as any error may
upset the entire research project!!

3/28/2015
 Survey Design
 Case study design
 Experimental design

3/28/2015
 Are designed to collect information that
describe, explore and help the investigator
understand social life
 They attempt to quantify social phenomena
particularly issues, conditions and problems
that are prevalent in the society
 The focus is on the link among a smaller
number of attributes across a sample of
cases depending on the resources available
and size of the population
 Are designed to collect information that
describe, explore and help the investigator
understand social life
 They attempt to quantify social phenomena
particularly issues, conditions and problems
that are prevalent in the society
 The focus is on the link among a smaller
number of attributes across a sample of
cases depending on the resources available
and size of the population

3/28/2015
 They are used to establish cause and effect
relationships by manipulating variables and
conditions
 Normally, an investigator wishes to establish
the effect of some process or intervention,
often referred to as “treatment” on some
subjects or experimental units.
 Mostly preferred in physical and biological
sciences as it allows a greater degree of
control and manipulation of variables
 They are used to establish cause and effect
relationships by manipulating variables and
conditions
 Normally, an investigator wishes to establish
the effect of some process or intervention,
often referred to as “treatment” on some
subjects or experimental units.
 Mostly preferred in physical and biological
sciences as it allows a greater degree of
control and manipulation of variables

3/28/2015
True Experimental design:
 A method of choice when attempting to
determine a cause and effect relationship
 Control and experimental groups are
tested and it allows an investigator to
control for more confounding variables
that if not addressed may lead into
inaccurate findings!
True Experimental design:
 A method of choice when attempting to
determine a cause and effect relationship
 Control and experimental groups are
tested and it allows an investigator to
control for more confounding variables
that if not addressed may lead into
inaccurate findings!

3/28/2015
 Commonly used in the social science in assessing
the outcomes of social programmes
 The main difference with true experimental design
is that, random assignment of subjects or units to
experimental and control groups is not possible in
quasi experimental design.
• Also, inability of researcher to manipulate some variables
and control extraneous factors(Independent variables that
are not related to the purpose of the study, but may affect
the dependent variable )
 “The question in Q-E always has been what is the
method of choosing subjects to participate in the
study”
 Read also Mugenda (2008). “Social Science
Research”
 Commonly used in the social science in assessing
the outcomes of social programmes
 The main difference with true experimental design
is that, random assignment of subjects or units to
experimental and control groups is not possible in
quasi experimental design.
• Also, inability of researcher to manipulate some variables
and control extraneous factors(Independent variables that
are not related to the purpose of the study, but may affect
the dependent variable )
 “The question in Q-E always has been what is the
method of choosing subjects to participate in the
study”
 Read also Mugenda (2008). “Social Science
Research”

3/28/2015
• Accrual of detailed information from an individual
 In the social science and life sciences, a case study is
a research method involving an up-close, in-depth, and
detailed examination of a subject of study (the case), as
well as its related contextual conditions.
 Although no single definition of the case study exists,
case-study research has long had a prominent place in
many disciplines and professions, ranging from
psychology, anthropology, sociology, and political
science to education, clinical science, social work, and
administrative science.
 The "case" being studied may be an individual,
organization, event, or action, existing in a specific time
and place
• Accrual of detailed information from an individual
 In the social science and life sciences, a case study is
a research method involving an up-close, in-depth, and
detailed examination of a subject of study (the case), as
well as its related contextual conditions.
 Although no single definition of the case study exists,
case-study research has long had a prominent place in
many disciplines and professions, ranging from
psychology, anthropology, sociology, and political
science to education, clinical science, social work, and
administrative science.
 The "case" being studied may be an individual,
organization, event, or action, existing in a specific time
and place

3/28/2015
 Base on a limited number of cases and
conduct an in-depth study about the
phenomenon/object under investigation
 Phenomenon/ Object is investigated from
different directions and rely on multiple
sources of evidence

3/28/2015
DATA TYPESDATA TYPES

3/28/2015
 Provides data that are
“first-hand” to an
investigator
 They have not yet
undergone any statistical
process
 These are data from the
respondents directly
 Unprocessed/ raw data and
are fresh collected for the
first time
 Data have undergone at
least a statistical process
 Sources can be:
1. Public documents and or
official records- NBS etc
2. Non- government
productions (Private
documents, eg. Tax
hospital records)
3. Mass media: news papers,
TVs, radio programs

3/28/2015
 It is possible to derive own
set of secondary data using
primary data
 Primary data is tailored to
your specific needs
 Due to the nature of how data
is collected, the reliability is
assured than it is with
secondary data
 Getting direct source
information, no filtering
 Is more specific results than
secondary and is usually
based on statistical
methodologies that involve
sampling
 Expensive to get as it
involves surveys and
interviews

3/28/2015
 Less expensive to get than
primary data
 May be used to verify
researcher’s own findings
 Can be used to guide a
researcher formulate a well
thought research problem
 The information is filtered
through the world view of
whomever is reporting it
(questioning reliability)

3/28/2015
Problems with using data from
Secondary sources
 Validity & reliability: validity of information may
vary from source to source; eg information
obtained from a census is likely to be more valid
than that obtained from personal diaries
 Personal bias: the use of information from
newspapers or magazines may suffer the
problem of personal bias since the writers are
likely to exhibit less rigorousness and objectivity
than we would expect in research report
 Availability of data: do not assume that data will
be available and can be accessed!!!!
 Format: make sure the data you need are
available and to the required format you want!!

3/28/2015
 Data collection is a term used to describe
a process of preparing and collecting data
 Systematic gathering of data for a
particular purpose from various sources,
that has been systematically observed,
recorded, organized
 Data are the basic inputs to any decision
making process

3/28/2015
• To obtain information
• To keep on records
• To make decisions about important issues
• To pass information on to others

3/28/2015
Primary ResearchPrimary Research
Quantitative dataQuantitative data Qualitative dataQualitative data
Surveys
Personal
interviews
Mail
Telephone, fax,
email, web
Self
administered
Surveys
Personal
interviews
Mail
Telephone, fax,
email, web
Self
administered
ExperimentsExperiments
SimulationSimulation
Focus GroupsFocus Groups
Individual depth interviewsIndividual depth interviews
Case studiesCase studies

3/28/2015
Qualitative Information
 The qualitative variables
can not be numerically
specified
 Examples: religion, marital
status, ethnic, gender,
inter alia.
 Gives explanations on
what and how!!
Quantitative Information
 Numerically specified
Examples: Age, Height,
Weight, number of births,
number of road accidents,
etc.
 Based on numbers: for
example 25% of the 54
students taking QMS 225
passed. No details on how
and perhaps why?

3/28/2015
 Act of perceiving as
conducted between two
people! One who is asking
the questions- Interviewer
& the other to answer the
question- Interviewee
 Researcher meets an
interviewee face to face
and ask questions (face-
to- face)
 In most cases, interview
guides/ schedules are used
 Required data is obtained
quickly
 Researcher is assured that
an interviewee
understands the questions
correctly
 Information provided can
be checked for its validity

3/28/2015
 Time consuming and it is costly
 There is a chance of providing wrong
information if respondents becomes
suspicious
 There are events that an interviewee may
become overexcited on some questions
and thereby lengthen the interview

3/28/2015
 Interviewer has the pre
determined questions
 May also provide questions
with alternative answers
 There could be open-
ended and closed
questions depending on
the needs of the study
 A researcher can compare
answers from different
respondents to see their
validity
 Quantitative analysis can
be used
Disadvantage:
Inflexible because the
researcher does not add
more questions outside of
those originally prepared!!

3/28/2015
 One can use personal or
group interview
 There is an interview guide
but the researcher is
interested in getting
information which the
respondent feels it is
important
 Only minimal guidance is
provided through
stimulating discussion
 Flexibility: researcher does
not stick on his/her
questions
 Chances are given to
respondents to say more
on what they are even not
asked
Disadvantage:
Difficult to compare the
answers given because of
different questions are
asked to different people

3/28/2015
 Record answers by
using respondents
own words so that you
do not distort the
message!
 Do not forget to thank
the respondents!!

3/28/2015
 The circumstance of being in or around an
ongoing setting and recording facts as
they are observed
 The observer works or interacts with the
study setting
 In this method, an eye and an ear
becomes very useful instruments!!
 The method is also referred to as
participant observation or experimental
method

3/28/2015
 Researcher gets to know
more ideas of the group he
is studying
 Data are recorded as they
occur or observed
 It is possible to have the
inside of the group and
understanding it more
than when outside!
 Ethical consideration: you
observe people without
knowing they are
observed- dangerous!!!!
 Going native: You will lose
one site of perspective and
pretend to be in the site of
those observed!
 Information gathered may
not be complete
 Costly in terms of time and
money and risky

3/28/2015
Designing a Questionnaire
 A survey involves directly collecting information
from people (sometimes organizations) whom we
are interested in.
 The types of information we are asking will take
care of people’s level of knowledge, attitude,
personalities, beliefs or preferences
 Well designed questionnaires are highly
structured to allow the same type of information
to be gathered from a large number of people in
the same way and data to be analyzed
quantitatively and systematically

3/28/2015
Objectives!!
 Are two: i) to maximize the response rate
(number of subjects responding to our
questionnaire). ii) to obtain accurate and
relevant information for the survey
 Establish rapport, carefully administer the
questionnaire, establish purpose, keep
reminding those who are yet to respond
 Key- what questions we ask, how we ask,
the order we ask and what should be the
lay out of the questionnaire!

3/28/2015
Deciding what to ask
I: Ask Precise questions!
 Avoid ambiguous
questions: for eg- how
often did you borrow books
from the library?
 Time frame is missing:
how many books have you
borrowed from the library
within the past 3 months?
 Also using terms with more
than a single meaning
mislead!!!
II: Ensure those you ask
have necessary
knowledge!
E.g; an agricultural survey
should target farmers and
not entrepreneurs (unless
are engaged in farming
activities)
III: Sensitive issues
e.g How many times are you
beaten by your wife?
OR questions on income/
salary are likely to give
wrong answers!!!

3/28/2015
Length and Format
 No universal agreement about how long a
questionnaire should be;
 Avoid a time consuming questionnaire as may be
boring and hence affect the response rate
 Short and simple always attracts high response
rate: the design shall depend on the nature of
the study and type of respondents one deals
 Open- ended and closed questionnaires are
available
 Closed- respondents are forced to choose
between several given options!
 Open- ended: respondents formulate answers

3/28/2015
Questionnaire cont…
I: Closed:
Please, indicate your age by
placing a tick in the
appropriately category
Under 15
15-19 years
20-24 years
II: Open – ended
E.g: How would you describe
your current marital
status?
…………………………………..
 Always use simple and
everyday language
Do not ask questions that
are based on
presumptions!
> What contraceptives
do you use?
Avoid leading questions!
> Multiple sex partners is
bad, isn’t it?
Avoid double barred
questions!
> How often and how much
time do you spend in EPL?

3/28/2015
Arranging questions
 Go from general to particular
 Go from easy to difficulty
 Go from factual to abstract
 Start with closed format questions
 Begin with questions relevant to the
subject
 Do not start with demographic/ personal
information (like age, sex, marital, etc).

3/28/2015
Outline/ Coverage
 The concept
 Terminologies
 Principles of sampling
 Factors affecting the inferences drawn
from the sample
 Need for sampling
 Types of sampling
 Calculating sample size

3/28/2015
The concept of sampling
Sample
Finding out sample
statistics
Population
Estimate the population parameters

3/28/2015
Explaining the concept
 Eg. 1: Suppose you want to estimate the average
age of the students in BSc. AS and POM II
classes- there are two ways of doing this;
A: contact all students in the class, find out their
ages, add them up and obtain the average by
dividing with their number!
B: Select a few in the class, obtain their age, sum
them up and divide by their number to have an
estimate age that will represent the average class
age.
Eg. 2: Consider finding out average income of a
household in Morogoro that is spend on paying
schools fees!
Eg. 3: “Kipima joto”

3/28/2015
Sampling: definition
 A process of selecting a few (sample) from a
bigger group (sampling population) to become
the basis for estimating or predicting a fact,
situation or outcome regarding the bigger
group!
 Saves both time, financial and human resources
 Danger of compromising the level of accuracy in
findings bearing in mind that estimation!!!
 You do not find out the facts about the
population’s characteristics of interest to you
but only estimate them

3/28/2015
Terminologies
 Population/ study population: aggregate of all
cases that conform to some designated set of
specifications. It is the complete set of the units
of analysis.
 Eg. Countries of the world; students at Mzumbe
Main Campus; chairs in this campus, African
Heads of State, etc.
 It is that set whose behavior is being studied or
investigated. Therefore, the specific nature of the
population depends on the research problem.
 Sample:- the small group of students from
whom you obtain the information to make any
required estimate

3/28/2015
Terminologies cont….
 Sample size:- the total number of students (in
our case) from whom you obtain the required
information
 Sampling design/technique:- the way of selecting
the sample (students in our case)
 Sampling unit:- each student that become the
basis of selection of the sample
 Sampling frame:- a list identifying each student
 Sample statistics:- the findings on the basis of
information obtained from the sample; eg
average age!
 Population parameter:- the population
characteristics you want to investigate (like
average age of the class)

3/28/2015
Principles of sampling
1. In a majority of cases where sampling is done
there will be a difference between the sample
statistics and the true population mean
(consider 4 individuals with ages: 18,20,23,25)
2. The greater the sample size, the more accurate
will be the estimate of the true population mean
3. The greater the difference in the variable under
study in a population, for a given sample size,
the greater will be the difference between
sample statistics and the true population mean
(mostly applies to heterogeneous population)

3/28/2015
Factors affecting inferences
drawn from a sample
 The sample size:- findings based on large
samples have more certainty than those based
on smaller ones; the larger the sample size, the
more accurate will be the findings!
 The extent of variation in the sampling
population:- the greater the variation in the
sampling population with respect to the
characteristics under study for a given sample
size, the greater will be the uncertainty!
 In technical terms:- “the greater the std, the
higher will be the std error, for a given sample
size in your estimate”

3/28/2015
Points to Note in Sampling
 Achieve maximum precision in your estimates
within a given sample size
 Avoid bias in the sample selection
Bias in sample selection can be a result of:
1. Sampling is influenced by human choice
2. Sampling frame does not cover the sampling
population accurately and completely
3. A section of the population refuses to
cooperate!!

3/28/2015
Types of Sampling
Sampling Types
Probability Sampling Non- probability Sampling
Simple Random
Cluster
Stratified
Systematic
Quota
Snowball
Judgmental
Accidental
SystematicSystematicSystematic
Stratified
Systematic
Stratified
Systematic
Stratified
Systematic
Cluster
Stratified
Systematic
Simple Random
Cluster
Stratified
Systematic
Probability Sampling
Simple Random
Cluster
Stratified
Systematic
Simple Random
Cluster
Stratified
Systematic
Simple Random
Cluster
Stratified
Systematic
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Non- probability Sampling
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Quota
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Judgmental
Quota
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Snowball
Judgmental
Quota
Sampling Types
Simple Random
Cluster
Stratified
Systematic
Accidental
Snowball
Judgmental
Quota
Sampling Types
Simple Random
Cluster
Stratified
Systematic

3/28/2015
 A random sampling design ensure that each
element in the population has an equal and
independent chance of being selected
 Equal means the probability of selection of each
element in the population is the same,
 The choice of an element in the sample is not
influenced by other considerations such as
personal influence!
 The choice of one element is not dependent upon
the choice of another element in the sampling
 The selection or rejection of one element does
not affect the inclusion or exclusion of another!

3/28/2015
Advantages of random sample
 Mainly TWO:
 Since they represent the total sampling
population, the inferences drawn from
such samples can be generalized to the
total sampling population
 Statistical tests based upon the theory of
probability can be applied only to data
collected from random samples

3/28/2015
Table of random numbers
 Identify the total number of elements in the study
population, (e.g 50,100,420,800,1245). The total number
of population may run up to 4 or more digits!
 Number each element starting from 1
 If the table for random numbers is on more than 1 page,
choose the starting page by a random procedure. Select
column or row that will be your starting point with a
random procedure and proceed from there in a
predetermined direction
 Corresponding to the number of digits to which the total
population runs, select the same number, randomly of
columns or rows of digits from the table
 Decide on your sample size
 Select the required number of elements for your sample
from the table. Discard a number selected twice- sampling
without replacement!!

3/28/2015
Simple Random Sampling
(SRS)
 Most commonly method of selecting a sample
 Each element in the population is given an equal
and independent chance of being selected;
 Is selected using the following;
> Identify by a number all elements or sampling
units in the population
> Decide on the sample size
>Select sample using fishbowl, table of random
numbers or a computer program

3/28/2015
Stratified Random Sampling
 The accuracy of estimates depends on the extent of
variability or heterogeneity of the study population
 If heterogeneity is reduced by any means, for a given
sample size, greater accuracy is achieved in your estimate.
 Stratified sampling is based in this logic!
 The researcher attempts to stratify the population in such a
way that the population within a stratum is homogeneous
with respect to the characteristic on the basis of which it is
being stratified!
 Characteristics chosen as the basis of stratification has to
be clearly identifiable in the study population; eg,
stratifying on the basis of a gender is much easier than on
the basis of income or age! Or even attitude
 Once the sampling population has been separated into non-
overlapping groups the researcher selects the required
number of elements from each stratum using SRS

3/28/2015
Types of Stratified Sampling
 Proportionate and disproportionate
 Proportionate: number of elements from
each stratum in relation to its proportion
in the total population is selected
 Disproportionate: the size of a stratum is
not considered!

3/28/2015
Procedure for selecting a stratified
sample
 Identify all sampling units in the sampling
population
 Decide upon the different strata (k) into which
you want to stratify the population
 Place each element into the appropriate stratum
 Number every element in each stratum
separately
 Decide the total sample size (n)
 Decide whether to select following proportionate
of disproportionate stratified sampling and follow
the following:

3/28/2015
Proportionate & disproportionate
approach
 Determine the number
of element to be
selected from each
stratum = n/k
 Select the required
number of elements
from each stratum
with SRS technique
 Determine the
proportion of each
stratum in the study
population = elements
in each stratum/total
population size
 Determine the # of
elements to be
selected form each
stratum = np
 Select sample using
SRS

3/28/2015
Cluster Sampling
 If inability to identify each element in a
population
 If the total sampling population is large as in the
case of a country or city- it is hard to identify
each sampling unit
 Cluster- ability of the researcher to divide the
sampling population into groups, called clusters
and then select elements within the cluster using
SRS!
 Clusters can be formed on the basis of
geographical proximity or a common
characteristic

3/28/2015
Cluster sampling, cont…
 Suppose you want to investigate the attitude of post-
secondary students in TZ towards problems in higher
education!
 You will have to consider the following facts:
1. Higher education institutions are in more than 10 regions
in the country
2. There are different types of institutions (colleges,
universities, universities of technology etc)
3. Within each institution various courses are offered to both
undergraduate and postgraduates
4. Each academic course take 3-4 years
Huge Task is required- Cluster Sampling will appropriately suit
the purpose!!!

3/28/2015
Systematic Sampling
 The sampling frame is first divided into a number
of segments, called intervals
 From the first interval, using SRS technique , one
element is selected.
 Selection of subsequent elements from other
intervals depends upon the order of the elements
selected in the first interval
 If it in the first interval it is the 5th element, the
5th element in each subsequent will be chosen
 Summarized procedure follows;

3/28/2015
Selecting a sample- Procedure
 Prepare a list of all elements in the study
population (N)
 Decide on the sample size (n)
 Determine the width of the interval given
by Total population/ Sample size =k
 Using SRS select an element from the 1st
interval
 Select the same order element from each
subsequent interval

3/28/2015
Non random sampling
designs
Quota sampling
Accidental sampling
Purposive sampling
Snowball sampling

3/28/2015
Quota Sampling
 Non- probability sampling methods are used when the
number of elements in the population is either unknown or
can not be individually identified
 The main consideration in this method is the researcher’s
ease of access to the sample population.
 In addition to convenience the method is guided by some
visible characteristics, such as gender, race of the study
population of interest.
 The sample is selected from a location convenient to the
researcher and whenever a person with the visible and
relevant characteristics is seen
 The person will be asked to participate in the study
 The process continues until the researcher has been able to
contact the required number of respondents (quota)
 Example: obtaining a sample of 20 male students from
QMS 225 Class

3/28/2015
Advantages and disadvantages
Advantages
 Least expensive
 No prior information is
required (sampling
frame, etc)
 Guarantees inclusion
of the people you
want75
Disadvantages
 The sample is not
probability one
 The findings can not be
generalized to the total
population
 The identified units
included in the sample
might have unique
characteristics and thus
may not represent the
population sampled

3/28/2015
Accidental Sampling
 Conveniently accessing the sampling
population
 Unlike in quota sampling, this method
does not attempt to include people
possessing obvious/ visible characteristics
 Has more or less same merits and
demerits as quota sampling
 You are not guided by any obvious
characteristics and some people contacted
may not have the required information

3/28/2015
Purposive sampling
 Based upon the judgment of the
researcher as to who can provide the best
information to achieve the objectives of
the study
 Includes only people who are likely to
have the required information (well
informed about the subject matter) and
that will be willing to share
 Useful for constructing historical reality,
describe a phenomenon or develop
something which little is known

3/28/2015
Snowball sampling
 Sample is selected using networks
 A few individuals in a group or organization are selected
and the required information is collected from them
 Then are asked to identify other people in the group/
organization, and the people selected by them become a
part of the sample
 Information is collected from them and then they are asked
to recommend other people from the group/organization,
and so on
 The process continues until the required number or a
saturation point is reached, in terms of the information
being sought
 The choice of the entire sample rest upon the choice of
individuals at the first stage!!!!!

3/28/2015
Sample Size Computation
 How big a sample should I collect?
 What should be my sample size?
 How many cases do I need?
 Depends on what you want to do with the
findings and what type of relationships you want
to establish
 Individual purpose in undertaking research is the
main determinant of the level of accuracy
required in the results which in turn determine
the sample size!!!
 Three issues to consider:-

3/28/2015
Example
 You want to find out the average age of
students within an accuracy of 0.5 of a year,
that is you can tolerate an error of half a year
on either side of the true average age. Let us
also assume that you want to find the average
age within half a year of accuracy at 95%
confidence level, that is you want to be 95%
confident about your findings. The formula for
calculating confidence interval is:
 
n
txx 
05.0
ˆ 

3/28/2015
Formula cont…
Where:
sizesamplen
deviationdards
errordards
n
levelconfidenceattofvaluet
samplethefromcalculatedageaveragex
meanpopulationtheofvalueestimatedx






tan
tan
%95
ˆ
05.0



3/28/2015
Cont…
If we decide to tolerate an error of ½ years, that means
  5.005.0  x
n
tx 
In other words we would like   5.005.0 



n
t  ……I
96.105.0 t
Sample t- values
Level 0.02 0.10 0.05 0.02 0.01 0.001
t- value 1.282 1.645 1.960 2.326 2.576 3.291
Thus from (I), the sample size becomes; 




 

5.0
96.1 
n
Given any value of  (standard deviation), the sample size can be obtained.

3/28/2015
Obtaining Sigma (σ)
 Sigma can be obtained by:
1. Guessing (depends on how large your
sample size would be)
2. Consulting an expert
3. From previous and comparable studies
4. Carrying out a pilot study to calculate the
value
So if we assume σ=1 in our example, given
½ year of error, a sample of at least 16
students will be necessary for our study.

Research methods

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Research methods

Similar to Research methods (20)

Recently uploaded

Recently uploaded (20)

Research methods