Sei sulla pagina 1di 67

Chapter Four

Research Design

By;
Mehari H. (PhD Can, MBA, MA, BA)
E-mail: haileg2003@gmail.com
Contents

• Meaning of research design


• Characteristics of research design
• Some important concepts Related to Research Design
• Types of research design
• Sampling and sampling design
3.1.Meaning of Research Design
• Once the researcher has determined the specific question to be
answered and research question into a clear, measurable
hypothesis, it is time to consider a suitable research design.
• is a master plan/a blue print or conceptual structure/ that
specifies the methods and procedures for collecting, measuring and
analyzing the needed information.
• Provide information on your proposed design for tasks such as
sample selection and size, data collection method,
instrumentation, procedures, methods for analyzing research
material and ethical requirements.
• What methodological approach are you going to adopt?
• It specifies the details of the procedures necessary for obtaining the
information needed to structure and /or solve research problems.
3.2. Characteristics of Research Design
 Clearly stated the objective of the problem to be studied,
 Being flexible, suitable, efficient, economical, and so on
 Minimizes bias and maximizes the reliability of the data
collected and analyzed;
 Gives the smallest experimental error;
 Yields relevant information and provides an
opportunity for considering many aspects of a problem;
 Assumes the availability of skills, time and money of the
3.3. Some Important Concepts Related to
Research Design
Concepts and Variables
• A concept is a generalized idea about an objects,
occurrences, process, etc.,
• Concepts are highly subjective as their
understanding varies from person to person,
cannot be measured .
• A concept that can be measured by crude/refined
or subjective/ objective units of measurement is
called a variable. So, measurability is the main
difference between a concept and a variable.
The difference between Concept and Variable

Concepts Variables
 Effectiveness  Gender ( male/ female)
 Satisfaction  Attitude ( Good /Bad)
 Excellence  Age ( X years, Y months)
 Rich  Weight ( X Kg)
 etc  Income($ ----/Year)
 Religion(Orthodox, Muslim,
protestant, etc)
 Subjective impression  Measurable through the degree of
9/8/20 6
How to Convert Concepts in to Variables
 If you are using a concept in your study, you need to
consider its operationalization- i.e., how it will be
measured.
 In most cases, to operationalize a concept you first
need to go through the process of identifying
indicators.
 Concepts are converted in to variables using a set of
rules, benchmarks, yardstick called indicators.
 Indicators are a set of criteria reflective of the
concept-which can then be converted into variables.
 The process is called operationalization of concepts.
9/8/20 7
Example

Conversion of Concepts in to Variables


Concept Indicators Variables Decision level

Concepts Indicators Variables Decision Level

(Working definition)
 Rich a. Income a. Income per 1. If > $ 100,000
b. Asset year 2. If > $ 300,000
b. Total value of
assets
 High a. Average a. Percentage of 1. If > 70%
academic marks marks 2. If > 90%
achievemen obtained in b. Percentage of
9/8/20 8
Types of Variables

• Variables are important in bringing clarity and


specification to the conceptualization of a research
problem and to the development of a research
instruments.

• There are number of ways of classifying variables.

• Here we look from 2 perspectives:


a) The causal relationship
b) The unit of measurement
9/8/20 9
From the View of the Causal Relationship
A. Independent Vs. Dependent variables
Independent variables
• also known as the predictor variable
• this variable is the ‘cause’
• can be manipulated/treated or allowed to vary.
Dependent /Outcome/effect /variable
• also known as the criterion variable.
• the outcome of the changes brought about by
introduction of an independent variable
• this variable is the ‘effect’
• should only vary in response to the IV
B. Extraneous variables
• Independent variables that are not related to the purpose of
the study, but may affect the dependent variable
• These factors not measured in the study may
increase or decrease the magnitude or strength of
the relationship between independent and
dependent variables.
C. Intervening/linking or connecting/ Variables
• sometimes called confounding variables. It links the
independent and dependent variable.
• In certain situations the relationship between an IV
& DV can not be estimated with out the
intervention of another variable.
• The cause variable/IV/ will have the assumed
effect only in the presence of an intervening
variable.
E.g., Income medical care /nutrition/ Longevity
IV linking variable DV
i.e., Because of just having money doesn’t make
one live long. Other variables link b/n money &
long live. People with high income tend to have
better medical care than those with low income.
From the view point of the Unit of Measurement

a) Categorical /qualitative variables


• Categorizing or grouping of individuals, objects,
property and responses based on common
characteristics.

• Categorical variables can be of three types:


Constant;
Dichotomous; and
Polytomous
• When a variable can have only one value or
category, it is known as a constant.
• E.g., taxi, tree, water etc
• When a variable can have only two categories it is
known as a dichotomous variable.
• E.g. yes/no; good/bad; rich/poor.
• When a variable can be divided in to more than two
categories, it is called a polytomous variable.
• e.g., Religion (Christian, Muslim, Hindu); Attitudes
(strongly favorable, favorable, uncertain,
unfavorable, strongly unfavorable).
b) Continuous/quantitative variables

• have continuity in their measurement.


• For example, age, income, an attitude score.
• Age can be measured in years, months and days.
Similarly, income can be measured in Birr and cents.
Types of Measurement Scales
• Measurement is central to any scientific inquiry.
• A scale may be defined as "any series of items which is
progressively arranged according to value or magnitude into
which an item can be placed according to its quantification
• A scale is a continuous spectrum or series of categories.
• The purpose of scaling is to represent, usually quantitatively,
an item's, a person's, or an event's place in the spectrum.
• There are 4 types of measurement scales. These are:
Nominal/Classificatory Scale
Ordinal/Ranking Scale
Interval
Ratio
1. Nominal Scale or categorical
• Comprise categories that cannot be rank ordered
• The numbers or letters assigned to objects serve as labels for
identification or classification.
• For example,
• individual’s sex, religion, occupation, or political party
affiliation.
• In representing gender (male or female) the researcher might
assign numbers to each category (e.g., 2 for females and 1
for males).
• With these values, however, we can only tabulate the number
of males and females.
• The order is arbitrary and any calculation (e.g. average) is
meaningless
2. Ordinal Scale

• variables can be ordered or ranked in relation to the amount of the


attribute possessed.
• they are arranged in ascending or descending order.
• Arranges objects or alternatives according to their magnitude in an
ordered relationship.
• Respondents are asked to rank order their preferences, in ordinal
values
• Does not say anything about the distance or interval between the
values
Example:
 Socioeconomic status(low, middle, upper).
 "excellent," "good," "fair," or "poor."
 result of examination: first, second, third classes and fail;
• Since the numbers of this scale have only a rank meaning, the
appropriate measure of central tendency is the median.
3. Interval Scale

• Not only rank order values but also measure order


(distance) in units of equal intervals.
• The distance between the categories are identical across
the range of categories (e.g. 01, 02, 03, etc.)
• The location of the zero point is arbitrary- does not
signify absence.
• E.g. temperature

• Generally used tests for statistical significance are the


‘t’ test and ‘F’ test.
4. Ratio Scale
• Has all the properties nominal, ordinal and interval scale
plus it has a fixed starting point.
• Possess all these features (distance, order, and
origin/absolute zero).
• Examples:
• Income, age, height, weight, distance etc.
• Generally, all statistical techniques are usable with ratio
scales and all manipulations that one can carry out with
real numbers can also be carried out with ratio scale values.
Characteristics of different Levels of Scale Measurement

Types of Data characteristics Numerical Descriptive statistics Examples


scale operation

Nominal Classification but no Counting Frequency in each Gender (1=Male),


order, distance, or category 0= Female)

origin Percent in each


category
Mode

Ordinal Classification and Rank ordering Median Academic status


order but no Range 1=Freshman,

distance, or unique Percentile ranking 2= Sophomore,

origin 3= Senior
4= Graduating

Interval Classification, order Arithmetic Mean Temperature in


9/8/20 and distance, but no operations that Standard deviation degree 21
Rating scales used in Business Research

1. Dichotomous scale (e.g Do you own a car? Yes No)


2. Category scale- uses multiple items to elicit a single
response. Eg. Where do you live? Mekelle, Adigudom,
Maichew
3. Semantic differential scale- Several bipolar attributes
are identified at the extremes of the scale, and
respondents are asked to indicate their attitudes,
toward a particular individual, object, or event on each
of the attributes.
E.g.
• Responsive ・・・・・ Unresponsive
• Beautiful ・・・・・ Ugly
• Courageous ・・・・・ Timid
4. Numerical scale
• The numerical scale is similar to the semantic
differential scale, with the difference that
numbers on a five-point or seven-point scale
are provided, with bipolar adjectives at both
ends.
• E.g. How pleased are you with your new estate
agent?
• Extremely Pleased 7 6 5 4 3 2 1 Extremely Displeased
5. Itemized rating scale
• A five-point or seven-point scale with anchors, as needed, is
provided for each item and the respondent states the
appropriate number on the side of each item, or circles the
relevant number against each item, as per the examples that
follow.
6. Likert scale
• The Likert scale is designed to examine how strongly subjects
agree or disagree with statements on a five-point.

7. Fixed or constant sum rating scale


• The respondents are here asked to distribute a given number
of points across various items as per the example below. This
is more in the nature of an ordinal scale.
8. Graphic rating scale
• A graphical representation helps the respondents to indicate on this
scale their answers to a particular question by placing a mark at the
appropriate point on the line, as in the following example. This is an
ordinal scale, though the following example might make it look like an
interval scale.

9. Consensus scale
• Scales can also be developed by consensus, where a panel of judges
selects certain items, which in its view measure the relevant concept.
Testing goodness of measures
Validity
• is a test of how well an instrument that is developed
measures the particular concept it is intended to
measure.
• Validity is concerned with whether we measure the
right concept.
• It ensures consistent measurement across time and
across the various items in the instrument.
• Is an indication of the stability and consistency with
which the instrument measures the concept and
helps to assess the “goodness” of a measure.
Types of Validity
Reliability

• Reliability is the degree to which measures are free from


random error and therefore yield consistent results
• It is the degree to which measures are free from
error and therefore yield consistent results.
• When the outcome of the measuring process is
reproducible, the measuring instrument is reliable.
• Two dimensions underlie the concept of reliability:
• Repeatability, and
• internal consistency.
• Stability of measures
• The ability of a measure to remain the same over time-despite
uncontrollable testing conditions or the state of the respondents
themselves.
• is indicative of its stability and low vulnerability to changes in
the situation.
• This attests to its “goodness” because the concept is stably
measured, no matter when it is done.
• Two tests of stability are
• test-retest reliability and
• parallel-form reliability.
Test–retest Reliability
• The reliability coefficient obtained by repetition of the
same measure on a second occasion. i.e when a
questionnaire containing some items that are supposed to
measure.
• a concept is administered to a set of respondents, now
and again to the same respondents, Say several weeks to
six months later, then the correlation between the scores
obtained at the two different times from one and the
same set of respondents is called the test-retest
coefficient.
• The higher it is, the better the test-retest reliability and,
consequently, the stability of the measure across time.
Parallel-form Reliability
• When responses on two comparable sets of measures
tapping the same construct are highly correlated.
• Both forms have similar items and the same response
format, the only changes being the wording and the
order or sequence of the questions.
• What we try to establish here is the error variability
resulting from wording and ordering of the questions.
• If two such comparable forms are highly correlated (say .
8 and above), we may be fairly certain that the measures
are reasonably reliable, with minimal error variance
caused by wording, ordering, or other factors.
Internal consistency of measures
• The internal consistency of measures is indicative of the
homogeneity of the items in the measure that taps the
construct.
• In other words, the items should “hang together as a set,”
and be capable of independently measuring the same
concept so that the respondents attach the same overall
meaning to each of the items.
• This can be seen by examining whether the items and the
subsets of items in the measuring instrument are correlated
highly.
• Consistency can be examined through the inter-item
consistency reliability and split-half reliability tests.
Inter-item consistency reliability

• The inter-item consistency reliability is a test of the


consistency of respondents’ answers to all the items in a
measure.
• To the degree that items are independent measures of
the same concept, they will be correlated with one
another.
• The most popular test of inter-item consistency
reliability is Cronbach’s coefficient alpha which is used
for multipoint-scaled items, and the Kuder–Richardson
formulas (Kuder & Richardson, 1937), used for
dichotomous items. The higher the coefficients, the
better the measuring instrument.
Split-half Reliability
• Reflects the correlations between two halves of an
instrument.
• The estimates will vary depending on how the items in
the measure are split into two halves.
• Split-half reliabilities may be higher than Cronbach’s
alpha only in the circumstance of there being more than
one underlying response dimension tapped by the
measure and when certain other conditions are met as
well.
• Hence, in almost all cases, Cronbach’s alpha can be
considered a perfectly adequate index of the inter-item
consistency reliability.
Types of Research Designs
• Can be classified into one of three broad
categories based on the nature of research,
purpose of research, research questions, sample
selection, data collection methods, and data
analysis techniques:
1. Quantitative research designs
2. Qualitative research designs, and
3. Mixed-research designs.
QUANTITATIVE RESEARCH DESIGNS

1. Experimental Research Design


2. Longitudinal Research Design
3. Cross-Sectional Design
4. Correlational Research Design
5. Meta analysis
Experimental Research Design
• The researcher randomly selects the participants
and randomly assigns them to the experimental
conditions.
• The most conclusive and powerful of all the
research designs and the only research design
that tests research questions and hypotheses to
establish cause–effect relationships.
• For this reason, it is sometimes called the
‘‘Golden Design.’’
Longitudinal Research Designs
Involve repeated measurement or observation on the same
individuals at several points over a period.
Collects data at two or more points in time.
longitudinal research is better equipped than cross-sectional
research to address the question of change.
Focuses primarily on change, growth, and developmental
types of research questions across many different disciplines
such as
• medicine,
• public health,
• business, and
• social and
• behavioral sciences.
Correlational Research
• It describes and assesses the magnitude and degree of an
existing relationship between two or more continuous
quantitative variables with interval or ratio types of
measurements or discrete variables with ordinal or
nominal type of measurements.
• The most common statistic that measures the degree of
the relationship between a pair of
• Continuous quantitative variables, having interval and
ratio types of measurements, is the Pearson product–
moment correlation coefficient.
• Dichotomous (a nominal type of measurement having
only two categories), the Phi coefficient should be
used.
• Ordinal type of measurement, the Spearman rank
order correlation coefficient should be used.
• Statistical methods used for prediction purposes are
simple and multiple regression analyses.
Cross-Sectional Design
It addresses our need to document facts at a
single moment in time.
Are quite common in social research.
It is a reasonable strategy for pursuing many
descriptive and exploratory research projects.
Meta-Analysis Research
• is used to quantitatively and systematically summarize
and synthesize the research results and findings from a
collection of primary studies that address and test the
same research question.
• General stages that guide meta analysts in their
systematic quantitative review.
• These stages include;
1. formulating research problems,
2. collecting primary research studies,
3. evaluating primary studies,
4. analyzing and modeling the meta-analytic data, and
5. interpreting and presenting the meta-analytic results.
QUALITATIVE RESEARCH DESIGNS
MIXED-METHODS RESEARCH DESIGNS
• Involve research studies that employ both quantitative
and qualitative research methodologies.
• Mixed research methods combine the deductive and
inductive inquiries of the scientific research methods.
• the mixed-methods research designs can be classified
into three types:
1. Exploratory,
2. Explanatory, and
3. Triangulation.
Exploratory Research Designs
• Using this design, the researcher;
• first conceptualizes a qualitative research study.
• Second, the researcher collects and analyzes the
qualitative data.
• Third, the researcher uses the findings from the
qualitative data analysis to conceptualize a
quantitative research study.
• Finally, the researcher collects and analyzes the
quantitative data to validate the qualitative
findings.
Explanatory Research Designs
• Using this design, the researcher
• first conceptualizes a quantitative research
study.
• Second, the researcher collects and analyzes
the quantitative data.
• Third, the researcher conceptualizes a
qualitative research study.
• Finally, the researcher collects and analyzes
the collected qualitative data to clarify and
enhance the quantitative research findings.
Triangulation Designs
• Using this design, the researcher simultaneously
conceptualizes quantitative and qualitative
research studies.
• Then, the researcher simultaneously collects and
analyzes both quantitative and qualitative data.
• Finally, the researcher uses the results from the
quantitative and qualitative studies to validate
findings from both studies.
SAMPLING AND SAMPLE DESIGN
• Sampling is a technique that identifies a
representative number from a given universe of
population so as to predict a fact, circumstances or
results about the universe.
• Hence, sample is the smaller portion of the study
population or sub group of the population.
Sample Vs population

• Sample or units of analysis is the small group of people


from whom a researcher draws out information about
the total population.
• Population refers to the total population form which we
select the few to sample.
Sample size
• refers to the number of respondents involved in the
sample survey.
• A prudent choice of the sample size for a particular
survey involves many considerations, which include
among others: resources in manpower, cost per sample
unit and funds available.
• At any cost however, a sample should be adequate to
draw conclusions about the population.
• Sampling methods are classified into probability and
non-probability .
Sampling Methods and Techniques
• The process of probability sampling can be divided into
four stages:
(1) Identify a suitable sampling frame based on your
research question(s) and objectives.
(2) Decide on a suitable sample size.
(3) Select the most appropriate sampling technique and
select the sample.
(4) Check that the sample is representative of the target
population.
1.Probability sampling

• In this method, each member of the population


has a known, nonzero chance of being chosen as
subjects in the sample.
• Allows the chance of an element being selected to
be quantified (ideally equal).
A. Simple Random sampling

• Also known as unrestricted probability sampling design.


• Every element in the population has a known and equal
chance of being selected as a subject.
• Is the best known sampling strategy.
• A commonly used simple random sampling procedure is
to assign a number to each element in the sampling
frame and use an unbiased process, such as a random
number generator, to select elements from the sampling
frame.
B. Systematic sampling
• involves selecting elements from a population list
in a systematic rather than random fashion.
• The researcher calculates a sampling interval
rather than using a list of random numbers.
• The interval serves as quasi-random selection
method.
• Thus, in a systematic sample, every ‘n’th
member has a chance to be included in the
sample.
C. Stratified Sampling
• involves dividing the population into homogenous
groups (sub-groups or strata) each group containing
subjects with similar characteristics.
• Is used when the population is heterogeneous.
• Because the population is nonhomogeneous, the
sampling technique needs to ensure that the sample
set comprises of every representative from the
nonhomogeneous group or the distinct categories.
• Hence the population is split into different groups
called as “strata.” Each “strata” is more homogenous
and bears common characteristics. The data items
from these “strata” are sampled to form the sample
set.
• involves a process of stratification or segregation,
followed by random selection of subjects from each
stratum.
• The population is first divided into mutually exclusive
groups that are relevant, appropriate, and meaningful
in the context of the study.
• E,g.
• Stratifying customers on the basis of life stages, income
levels, stratifying companies according to size, industry,
profits, dividing a student population into female and
male, stratification of the population by geographical
area, market segment, consumers' age, consumers'
gender.
D. Cluster (Multistage) Sampling
• In cluster sampling we have to have a number of clusters
which are characterized by heterogeneity in between and
homogeneity within.
• is used for a variety of purposes particularly for large
sample surveys or a nation-wide survey.
• It is very convenient with respect to the time and money
allocated for a particular study.
• The sample is also reliable as it allows random allocation at
different stages. If we consider two stages to conduct the
survey, then it is called two-stage cluster sampling.
• If someone considers more than two stages to collect the
data, then it is called multistage sampling.
2. Non-probability sampling

• It is a type of sampling that does not use probability


theory and it is used when number of elements in a
population is not distinguished or known.
• The non-probability samples are drawn based on
judgment regarding the characteristics of the target
population and needs of survey.
• In this sampling method, some members of the eligible
target population have a chance of being chosen and
others do not.
• Examples of non-random sample include quota,
purposive/judgmental and snowball sampling.
A. Quota sampling
• the researcher divides the population into-sub groups
or categories such as men and women, arts and
commerce, younger and older, etc.
• Thus, the number of respondents is fixed in various
categories of the sample.
• The selection is influenced by researcher’s preference or
access to sampling.
• In quota sampling information about the whole
population is not needed.
B.Purposive Sampling
• Here, the researcher contacts only most relevant
respondent whom he/she thinks are with
sampling characteristics he/she is interested in.
• The judgment of the researcher is used in
selecting cases with a specific purpose in mind.
• It is usually common in gathering historical
facts, describing phenomena or when the subject
of discussion is not widely known.
C. Convenience Sampling
• refers to the collection of information from members of
the population who are conveniently available to
provide it.
• is most often used during the exploratory phase of a
research project and is perhaps the best way of getting
some basic information quickly and efficiently.
• Assume the researcher visits the mall and at that point
of time only four to five participants are available.
Inclusion of these samples will be biased and limited.
This sampling lacks generalization and is preferred for
pilot testing.
D.Snowball (Network or Chain) Sampling
• Snowball sampling is also known as network or
reputational sampling since it used uses
networking to obtain the required information
from those capable individuals.
• Here, the first people who are contacted will
create their own contact with others and so on
until the necessary data is collected.
• The quality of data that could be gathered from
such sample is therefore, liable to the influence of
the first group.
Thank You!!

Potrebbero piacerti anche