1. 11
Research DesignResearch Design
ConceptualConceptual structurestructure within whichwithin which
research is conducted.research is conducted.
Constitutes theConstitutes the blueprintblueprint for thefor the
Collection, Measurement and Analysis ofCollection, Measurement and Analysis of
data.data.
AnAn outlineoutline of what the researcher will doof what the researcher will do
from writing the hypothesis and itsfrom writing the hypothesis and its
operational implications to the finaloperational implications to the final
analysis of data.analysis of data.
2. 22
Design DecisionsDesign Decisions
What is the study about?What is the study about?
Why is the study being made?Why is the study being made?
Where will the study be carried out?Where will the study be carried out?
What type of data is required?What type of data is required?
Where can the required data be found?Where can the required data be found?
What periods of time will the studyWhat periods of time will the study
include?include?
What will be the sample design?What will be the sample design?
What techniques of data collection will beWhat techniques of data collection will be
used?used?
3. 33
How will the data be analyzed?How will the data be analyzed?
In what style will the report be prepared?In what style will the report be prepared?
PARTS OF RESEARCH DESIGNPARTS OF RESEARCH DESIGN
SamplingSampling DesignDesign: deals with the: deals with the
method of selecting items to be observed.method of selecting items to be observed.
ObservationalObservational DesignDesign: relates to the: relates to the
conditions under which the observationsconditions under which the observations
are to be made.are to be made.
4. 44
Statistical DesignStatistical Design : concerns with how: concerns with how
the information and data gathered are tothe information and data gathered are to
be analyzed.be analyzed.
Operational DesignOperational Design : deals with the: deals with the
techniques by which the procedurestechniques by which the procedures
specified in the sampling, statistical andspecified in the sampling, statistical and
observational designs can be carried out.observational designs can be carried out.
5. 55
Research Design shouldResearch Design should
contain:contain:
1.1. AA clearclear statementstatement of the researchof the research
problem.problem.
2.2. ProceduresProcedures andand techniquestechniques to be usedto be used
for gathering information.for gathering information.
3.3. TheThe populationpopulation to be studied.to be studied.
4.4. Methods to be used inMethods to be used in processingprocessing andand
analyzinganalyzing data.data.
6. 66
Need for Research DesignNeed for Research Design
Thoughtlessness in designing theThoughtlessness in designing the
research project may result in renderingresearch project may result in rendering
the research exercisethe research exercise futilefutile..
1.1. FacilitatesFacilitates smoothsmooth sailingsailing of the variousof the various
research operations.research operations.
2.2. Makes research as efficient as possibleMakes research as efficient as possible
yieldingyielding maximalmaximal informationinformation withwith
minimalminimal expenditureexpenditure of effort, time andof effort, time and
money.money.
7. 77
3.3. Has a great bearing on theHas a great bearing on the reliabilityreliability ofof
thethe resultsresults arrived atarrived at ..
4.4. Constitutes theConstitutes the foundationfoundation of the entireof the entire
edifice of the research work.edifice of the research work.
5.5. Helps the researcher to organize hisHelps the researcher to organize his
ideas in a form possible for him to lookideas in a form possible for him to look
forfor flawsflaws andand inadequaciesinadequacies..
6.6. Helps inHelps in critical evaluationcritical evaluation ..
7.7. Provides aProvides a comprehensivecomprehensive reviewreview of theof the
proposed study.proposed study.
8. 88
Features of a Good DesignFeatures of a Good Design
AppropriateAppropriate
EconomicalEconomical
Minimizes biasnessMinimizes biasness
Maximizes the reliability of data collectedMaximizes the reliability of data collected
Reliability of data analyzedReliability of data analyzed
OneOne singlesingle designdesign cannot serve thecannot serve the
purpose of all types of researchpurpose of all types of research
problems.problems.
9. 99
Basic Concepts Relating toBasic Concepts Relating to
Research DesignResearch Design
VariablesVariables::
A Concept which can take on differentA Concept which can take on different
Quantitative values.Quantitative values.
It is something whose magnitude canIt is something whose magnitude can
change.change.
Price, Cost, Profit, Saving, Investment,Price, Cost, Profit, Saving, Investment,
Export, Import, Income.Export, Import, Income.
10. 1010
Qualitative PhenomenaQualitative Phenomena or the Attributesor the Attributes
are also quantified on the basis of theare also quantified on the basis of the
presence or absence of the concerningpresence or absence of the concerning
attribute (s). (Honesty, BeautyâŚ.)attribute (s). (Honesty, BeautyâŚ.)
Continuous Variable:Continuous Variable: It can assume anyIt can assume any
numerical value within a specific range.numerical value within a specific range.
Phenomena which can take onPhenomena which can take on
quantitatively different values even inquantitatively different values even in
decimal points.decimal points.
Non-continuous Variable:Non-continuous Variable: VariablesVariables
expressed in integer values are non-expressed in integer values are non-
continuous variables (Discrete Variables).continuous variables (Discrete Variables).
11. 1111
Dependent Variable:Dependent Variable: If one variableIf one variable
depends upon or is a consequence of thedepends upon or is a consequence of the
other variable, it is termed as DV.other variable, it is termed as DV.
(Demand depends upon Price)(Demand depends upon Price)
Independent Variable:Independent Variable: The VariableThe Variable
whose value is used for the estimation ofwhose value is used for the estimation of
the value of the dependent variable isthe value of the dependent variable is
called as independent variable. (Pricecalled as independent variable. (Price
influences demand)influences demand)
The variable that is antecedent to theThe variable that is antecedent to the
dependent variable is known as IV.dependent variable is known as IV.
12. 1212
Extraneous Variable:Extraneous Variable: IndependentIndependent
variables that are not related to thevariables that are not related to the
purpose of the study, but may affect thepurpose of the study, but may affect the
DV are termed as EV. (Yielding as a resultDV are termed as EV. (Yielding as a result
of fertilizer used other than fertility of land).of fertilizer used other than fertility of land).
Experimental Error:Experimental Error: Any effect noticed onAny effect noticed on
DV as a result of EV is technically knownDV as a result of EV is technically known
as an Experimental Error.as an Experimental Error.
Study must be so designed that the effectStudy must be so designed that the effect
upon the DV is attributed entirely to the IVupon the DV is attributed entirely to the IV
and not to some EV or Variables.and not to some EV or Variables.
13. 1313
Control:Control: Good research design intends toGood research design intends to
minimize the influence of EV. When theminimize the influence of EV. When the
study is so designed to minimize thestudy is so designed to minimize the
effects of EV, the technical term Control iseffects of EV, the technical term Control is
used. (Use of Fertilizer on yielding).used. (Use of Fertilizer on yielding).
Confounded Relationship:Confounded Relationship: When the DV isWhen the DV is
not free from the influence of EVs, thenot free from the influence of EVs, the
relationship between the DV and the IVs isrelationship between the DV and the IVs is
said to be confounded by an EVs.said to be confounded by an EVs.
14. 1414
Research Hypothesis:Research Hypothesis: Is a predictiveIs a predictive
statement that relates an IV to a DV.statement that relates an IV to a DV.
When a prediction or a hypothesizedWhen a prediction or a hypothesized
relationship is to be tested by scientificrelationship is to be tested by scientific
methods, it is known as researchmethods, it is known as research
hypothesis.hypothesis.
Predictive statements, which are not to bePredictive statements, which are not to be
objectively verified or the relationships thatobjectively verified or the relationships that
are assumed but not to be tested,are assumed but not to be tested, are notare not
RH.RH.
15. 1515
Experimental hypothesis-testing research:Experimental hypothesis-testing research:
When the purpose of the research is toWhen the purpose of the research is to
test a RH, it is known astest a RH, it is known as hypothesis-hypothesis-
testing research.testing research.
Research in which theResearch in which the IV is manipulatedIV is manipulated isis
termed as EHTR. (Fertility of Soil).termed as EHTR. (Fertility of Soil).
Research in which anResearch in which an IV is notIV is not
manipulatedmanipulated is called as NEHTR.is called as NEHTR.
Experimental and Control Groups:Experimental and Control Groups: In anIn an
Experimental Hypothesis-testing research whenExperimental Hypothesis-testing research when
a group isa group is exposed to usual conditions,exposed to usual conditions, it isit is
known as Control Group. (Non-beneficiary of aknown as Control Group. (Non-beneficiary of a
loan)loan)
16. 1616
When the group is exposed to someWhen the group is exposed to some
special conditionspecial condition, it is known as, it is known as
Experimental GroupExperimental Group.. (Beneficiary of a(Beneficiary of a
Loan).Loan).
Treatments:Treatments: The different conditions underThe different conditions under
which Experimental and control groupswhich Experimental and control groups
are put usually referred to as Treatments.are put usually referred to as Treatments.
(Sanction of loan, Impact of varieties of(Sanction of loan, Impact of varieties of
fertilizers on the Yield).fertilizers on the Yield).
Experiment:Experiment: The process of examining theThe process of examining the
truth of a statistical hypothesis, relating totruth of a statistical hypothesis, relating to
some research problem, is known as ansome research problem, is known as an
experiment. (Examining the usefulness ofexperiment. (Examining the usefulness of
a newly developeda newly developed drug).drug).
17. 1717
Absolute Experiment:Absolute Experiment: Determining theDetermining the
impact of one variable on otherimpact of one variable on other
exclusively, is a case of Absolute Exp.exclusively, is a case of Absolute Exp.
(Impact of a fertilizer on the yield of a(Impact of a fertilizer on the yield of a
crop)crop)
Comparative Experiment:Comparative Experiment: Determining theDetermining the
impact of one variable as compared toimpact of one variable as compared to
other variables, is a case of Comparativeother variables, is a case of Comparative
experiment. (Determining the impact ofexperiment. (Determining the impact of
one fertilizer as compared to the impact ofone fertilizer as compared to the impact of
other fertilizers).other fertilizers).
18. 1818
Experimental Units:Experimental Units: The pre-determinedThe pre-determined
plots or the blocks, where differentplots or the blocks, where different
treatments are used, are known astreatments are used, are known as
Experimental Units. (Beneficiary, non-Experimental Units. (Beneficiary, non-
beneficiaryâŚ).beneficiaryâŚ).
Different Research DesignDifferent Research Design
Research Design in case ofResearch Design in case of ExploratoryExploratory
Research Studies.Research Studies.
Research Design in case ofResearch Design in case of DescriptiveDescriptive
and Diagnostic Research Studies.and Diagnostic Research Studies.
Research Design in case ofResearch Design in case of Hypothesis-Hypothesis-
testing Research Studies.testing Research Studies.
19. 1919
RD in case of Exploratory/RD in case of Exploratory/
Formulative Research StudiesFormulative Research Studies
Major emphasis in such studies is onMajor emphasis in such studies is on
Discovery of Ideas and InsightsDiscovery of Ideas and Insights..
Main purpose is that of formulating aMain purpose is that of formulating a
problem for moreproblem for more precise Investigationprecise Investigation oror
developing the working hypotheses fromdeveloping the working hypotheses from
an operational point of view.an operational point of view.
The RD must beThe RD must be FlexibleFlexible enough toenough to
provide opportunity for consideringprovide opportunity for considering
different aspects of a problem.different aspects of a problem.
20. 2020
Necessity is to make changes in theNecessity is to make changes in the
research procedures for gatheringresearch procedures for gathering
relevant data.relevant data.
Three Methods in RDThree Methods in RD
Survey of Concerning LiteratureSurvey of Concerning Literature
Experience SurveyExperience Survey
Analysis of âInsight-stimulatingâ examples.Analysis of âInsight-stimulatingâ examples.
21. 2121
The Survey of ConcerningThe Survey of Concerning
Literature:Literature:
Hypotheses stated by earlier workers mayHypotheses stated by earlier workers may
bebe reviewed.reviewed.
Usefulness of the hypotheses beUsefulness of the hypotheses be
evaluatedevaluated as a basis for further research.as a basis for further research.
Need to consider whether the alreadyNeed to consider whether the already
stated hypotheses suggeststated hypotheses suggest newnew
hypothesis.hypothesis.
Researcher shouldResearcher should review and build uponreview and build upon
the work already done by others.the work already done by others.
22. 2222
In cases where hypotheses have not yetIn cases where hypotheses have not yet
been formulated, the task is tobeen formulated, the task is to ReviewReview thethe
available material for deriving relevantavailable material for deriving relevant
hypotheses from it.hypotheses from it.
Bibliographical surveyBibliographical survey of studies shouldof studies should
be made by the researcher for preciselybe made by the researcher for precisely
formulating the Problem.formulating the Problem.
Works ofWorks of creative writerscreative writers provideprovide a fertilea fertile
ground for hypothesis formulation.ground for hypothesis formulation.
23. 2323
Experience SurveyExperience Survey
Survey of peopleSurvey of people who have had practicalwho have had practical
experience with the problem be studied.experience with the problem be studied.
Helps in obtainingHelps in obtaining insightinsight into theinto the
relationships between variables.relationships between variables.
Facilitates withFacilitates with new ideasnew ideas relating to therelating to the
research problem.research problem.
Need forNeed for careful selectioncareful selection of the peopleof the people
(respondents), found to be competent who(respondents), found to be competent who
can contribute new ideas.can contribute new ideas.
Need for preparation of anNeed for preparation of an interviewinterview
schedule.schedule.
24. 2424
Need forNeed for systematic questioningsystematic questioning ofof
informants.informants.
Interview must ensureInterview must ensure FlexibilityFlexibility ââ
respondents should be allowed to raiserespondents should be allowed to raise
issues & questions.issues & questions.
Need for sending the questionsNeed for sending the questions inin
advanceadvance to the respondents â advanceto the respondents â advance
thinking and for contributing effectively.thinking and for contributing effectively.
Facilitates to define the problem moreFacilitates to define the problem more
conciselyconcisely and helps in theand helps in the formulation offormulation of
the research hypothesis.the research hypothesis.
25. 2525
Analysis of âInsight-stimulatingâAnalysis of âInsight-stimulatingâ
Examples:Examples:
Suitable in areas whereSuitable in areas where little experiencelittle experience toto
serve as a guide.serve as a guide.
Consists ofConsists of intensive studyintensive study of selectedof selected
instances of the phenomenon.instances of the phenomenon.
Existing recordsExisting records may be examined.may be examined.
Unstructured interviewingUnstructured interviewing may be done.may be done.
Drawing together diverse information intoDrawing together diverse information into
aa unified interpretationunified interpretation is the main featuresis the main features
for evoking insights.for evoking insights.
26. 2626
Study of Individuals who are in transitionStudy of Individuals who are in transition
from one stage to another.from one stage to another.
Reactions of Individuals from differentReactions of Individuals from different
social strata.social strata.
RD must continue to be Flexible.RD must continue to be Flexible.
Many different facets of a ProblemMany different facets of a Problem
may be considered as and whenmay be considered as and when
they arise.they arise.
27. 2727
Overall Design:Overall Design:
Flexible Design. Design must provideFlexible Design. Design must provide
opportunity for considering differentopportunity for considering different
aspects of the problem.aspects of the problem.
Sampling Design:Sampling Design:
Non-probability sampling design-Non-probability sampling design-
Purposive or Judgment sampling.Purposive or Judgment sampling.
Statistical Design:Statistical Design:
No pre-planned design for analysis.No pre-planned design for analysis.
Observational Design:Observational Design:
Unstructured instruments for collection ofUnstructured instruments for collection of
data.data.
28. 2828
RD in case of Descriptive and DiagnosticRD in case of Descriptive and Diagnostic
Research StudiesResearch Studies
Concerned withConcerned with narration of factsnarration of facts andand
characteristics of an individual, group or acharacteristics of an individual, group or a
situation.situation.
Diagnostic Research Studies determineDiagnostic Research Studies determine
thethe frequencyfrequency with which somethingwith which something
occurs (whether certain variables areoccurs (whether certain variables are
associated).associated).
Aim is to obtainAim is to obtain complete and accuratecomplete and accurate
informationâprocedure must be carefullyinformationâprocedure must be carefully
planned.planned.
29. 2929
RD must enough provision for protectionRD must enough provision for protection
against biasagainst bias and mustand must maximize reliabilitymaximize reliability..
RD must beRD must be rigidrigid and not flexible withand not flexible with
focus on:focus on:
1.1. What the study is about?What the study is about?
2.2. Why the study is being made?Why the study is being made?
3.3. What techniques of gathering data will beWhat techniques of gathering data will be
adopted?adopted?
4. How much material will be needed?4. How much material will be needed?
5. Where can the required data be found?5. Where can the required data be found?
30. 3030
5.5. What time period should the data beWhat time period should the data be
related?related?
6. How the data will be processed and6. How the data will be processed and
analyzed?analyzed?
7. What style will be followed in preparing the7. What style will be followed in preparing the
report?report?
First, theFirst, the objectivesobjectives should be clearlyshould be clearly
specified to ensure that the data collectedspecified to ensure that the data collected
are relevant.are relevant.
Second,Second, techniquestechniques for collecting datafor collecting data
must be devised.must be devised.
31. 3131
While collecting information, adequateWhile collecting information, adequate
safeguards againstsafeguards against biasbias andand unreliabilityunreliability
must be ensured.must be ensured.
Questions must beQuestions must be well examinedwell examined and beand be
mademade unambiguousunambiguous..
Interviewers must be instructed not toInterviewers must be instructed not to
express theirexpress their own opinionown opinion..
Observers must be trained toObservers must be trained to uniformlyuniformly
recordrecord a given item of behavior.a given item of behavior.
Pre-testingPre-testing the data collection instrumentsthe data collection instruments
are necessary.are necessary.
32. 3232
StructuredStructured Instruments are necessary forInstruments are necessary for
such studies.such studies.
To make statements about the population,To make statements about the population,
samplesample has to be designed.has to be designed.
Random SamplingRandom Sampling procedure should beprocedure should be
followed.followed.
Data should be examined forData should be examined for
completeness, consistency and reliability.completeness, consistency and reliability.
Processing and analyzing procedure shouldProcessing and analyzing procedure should
bebe planned beforeplanned before research work started.research work started.
33. 3333
Statistical computations like averages,Statistical computations like averages,
percentages and coefficients must bepercentages and coefficients must be
worked out in the most proper way.worked out in the most proper way.
Layout of the report needs to beLayout of the report needs to be wellwell
plannedplanned..
Minimization ofMinimization of biasbias and maximization ofand maximization of
reliabilityreliability of evidence collected should beof evidence collected should be
ensured.ensured.
Overall Design:Overall Design:
RigidRigid Design. Protection against biasnessDesign. Protection against biasness
and maximization of reliability.and maximization of reliability.
34. 3434
Sampling Design:Sampling Design:
Probability Sampling Design.Probability Sampling Design.
Statistical Design:Statistical Design:
Pre-planned Design for Analysis.Pre-planned Design for Analysis.
Observational Design:Observational Design:
Structured or well thought out Instruments forStructured or well thought out Instruments for
collection of Data.collection of Data.
Operational Design:Operational Design:
Advanced Decisions about OperationalAdvanced Decisions about Operational
Procedures.Procedures.
35. 3535
Research Design in case ofResearch Design in case of
Hypothesis- testing Research StudiesHypothesis- testing Research Studies
Researcher tests theResearcher tests the hypotheseshypotheses of causalof causal
relationships between variables.relationships between variables.
Helps in drawingHelps in drawing inferencesinferences about causality.about causality.
RD often mean the design ofRD often mean the design of Experiments-Experiments-
known as Experimental Studies.known as Experimental Studies.
The study of Experimental Designs has itsThe study of Experimental Designs has its
origin inorigin in Agricultural Research ( Prof. Fisher).Agricultural Research ( Prof. Fisher).
36. 3636
ProcedureProcedure : Dividing Agricultural fields into: Dividing Agricultural fields into
different blocks and conductingdifferent blocks and conducting
Experiments on each blocks, informationExperiments on each blocks, information
gathered and inferences drawn found to begathered and inferences drawn found to be
more reliable.more reliable.
Three Principles of RD:Three Principles of RD:
The Principle of ReplicationThe Principle of Replication
The Principle of RandomizationThe Principle of Randomization
The Principle of Local ControlThe Principle of Local Control
37. 3737
Principle of ReplicationPrinciple of Replication
Experiment should beExperiment should be repeatedrepeated more thanmore than
once.once.
Each treatment is applied inEach treatment is applied in many experimentalmany experimental
units.units.
Facilitates in bringing theFacilitates in bringing the statistical accuracystatistical accuracy ofof
the experiments.the experiments.
Example:Example: Examining yield rate of two varietiesExamining yield rate of two varieties
of Paddy: Dividing the field intoof Paddy: Dividing the field into several partsseveral parts,,
grow one variety ingrow one variety in halfhalf of these parts and theof these parts and the
other variety in theother variety in the remainingremaining parts.parts.
38. 3838
Drawing conclusion by comparing the data ofDrawing conclusion by comparing the data of
yield rate of the two varieties.yield rate of the two varieties.
The result so obtained will be moreThe result so obtained will be more reliablereliable byby
applying the principle of Replication.applying the principle of Replication.
Increases theIncreases the precisionprecision of a study.of a study.
Principle of RandomizationPrinciple of Randomization
ProvidesProvides protectionprotection against the effects ofagainst the effects of
Extraneous factors by Randomization.Extraneous factors by Randomization.
39. 3939
Indicates that designing the Experiment shouldIndicates that designing the Experiment should
be done in such a way that thebe done in such a way that the variationsvariations
caused bycaused by ExtraneousExtraneous factors can all befactors can all be
combined under the heading ofcombined under the heading of âChanceâ.âChanceâ.
Example: Growing of rice in first half of theExample: Growing of rice in first half of the
parts of a field and the other variety in otherparts of a field and the other variety in other
half, it may so happen that the fertility of landhalf, it may so happen that the fertility of land
may be different in both the cases resulting inmay be different in both the cases resulting in
unrealisticunrealistic findings.findings.
Assigning of variety of rice to be grown inAssigning of variety of rice to be grown in
different parts of the field on the basis ofdifferent parts of the field on the basis of
Random Sampling TechniqueRandom Sampling Technique..
40. 4040
Randomization principleRandomization principle protectsprotects the effects ofthe effects of
the Extraneous factors (soil fertility differences)the Extraneous factors (soil fertility differences)
and helps inand helps in better estimationbetter estimation of Experimentalof Experimental
error.error.
Principle of Local ControlPrinciple of Local Control
The Extraneous factor is made to varyThe Extraneous factor is made to vary
deliberately over a wide range as necessary sodeliberately over a wide range as necessary so
that the variability can be measured andthat the variability can be measured and
eliminated from the experimental error.eliminated from the experimental error.
Example: Treatments (Varieties of Rice)Example: Treatments (Varieties of Rice)
Extraneous factor (Soil fertility)Extraneous factor (Soil fertility)
41. 4141
As per the Principle :As per the Principle : DivideDivide the field intothe field into
severalseveral homogeneous partshomogeneous parts, known as, known as blocksblocks..
ThenThen dividedivide each such block into parts equal toeach such block into parts equal to
the number ofthe number of treatmentstreatments..
Assigned theAssigned the treatments randomlytreatments randomly to theseto these
parts of a block.parts of a block.
Blocks are the levels at which we hold anBlocks are the levels at which we hold an
Extraneous factor fixed, to measure itsExtraneous factor fixed, to measure its
contribution to the total variability of the data.contribution to the total variability of the data.
42. 4242
Types of Experimental DesignTypes of Experimental Design
Experimental Design refers to theExperimental Design refers to the frameworkframework oror
structurestructure of an Experiment.of an Experiment.
Types:Types:
Before-and-after without control DesignBefore-and-after without control Design
After-only with control DesignAfter-only with control Design
Before-and-after with control DesignBefore-and-after with control Design
Before-and-after without control DesignBefore-and-after without control Design
AA single testsingle test group or area is selected.group or area is selected.
The Dependent variable is measuredThe Dependent variable is measured beforebefore
thethe introductionintroduction of the treatment.of the treatment.
43. 4343
Treatment is then introduced.Treatment is then introduced.
The dependent variable is measured after theThe dependent variable is measured after the
treatment has been introduced.treatment has been introduced.
The effect of treatment: Level of PhenomenonThe effect of treatment: Level of Phenomenon
after the treatment minus the Level ofafter the treatment minus the Level of
Phenomenon before the treatment.Phenomenon before the treatment.
Test area: Level of Phenomenon Treatment Level of PhenomenonTest area: Level of Phenomenon Treatment Level of Phenomenon
Before treatment (X) Introduced After treatment (Y)Before treatment (X) Introduced After treatment (Y)
Treatment Effect = ( Y ) â ( X )Treatment Effect = ( Y ) â ( X )
44. 4444
After â only with Control DesignAfter â only with Control Design
Two groupsTwo groups or areas (test area and controlor areas (test area and control
area) are selected.area) are selected.
TreatmentTreatment is introduced into theis introduced into the test areatest area
only.only.
TheThe DVDV is measured inis measured in both the areasboth the areas at theat the
same time.same time.
Treatment impact is assessed byTreatment impact is assessed by subtractingsubtracting
thethe value of DVvalue of DV in thein the control areacontrol area from itsfrom its
value in thevalue in the test areatest area..
45. 4545
Test Area: Treatment Introduced Level of Phenomenon after
Treatment (Y)
Control Area: Level of Phenomenon without
Treatment (Z)
Treatment Effect = (Y) â (Z)
46. 4646
Assumptions of the Design:Assumptions of the Design:
Two areas mustTwo areas must be identicalbe identical with respect towith respect to
their behavior towards the phenomenontheir behavior towards the phenomenon
considered.considered.
FailureFailure to assume so, there is possibility ofto assume so, there is possibility of
extraneous variationextraneous variation in entering into thein entering into the
treatment effect.treatment effect.
This Design is Superior to before â and âThis Design is Superior to before â and â
after without control design.after without control design.
47. 4747
Before â and â after with Control DesignBefore â and â after with Control Design
Two areasTwo areas are selected for the study.are selected for the study.
DVDV is measured in both the areas for anis measured in both the areas for an
identical time-periodidentical time-period before the treatment.before the treatment.
TreatmentTreatment is introduced into theis introduced into the testtest area only.area only.
DVDV is measured in both for an identical time-is measured in both for an identical time-
period after theperiod after the introductionintroduction of the treatment.of the treatment.
The treatment effect is determined byThe treatment effect is determined by
subtractingsubtracting the change in thethe change in the DV in the controlDV in the control
area from the change in thearea from the change in the DV in test areaDV in test area..
48. 4848
Test Area: Level of Phenomenon Treatment Level of Phenomenon
before treatment (X) introduced after treatment (Y)
Control Area: Level of Phenomenon Level of Phenomenon
without treatment without treatment
(A) (Z)
Treatment Effect = (Y - X) â (Z - A)
Time Period I Time Period II
49. 4949
Completely Randomized Design (C.R. Design)Completely Randomized Design (C.R. Design)
Involves two principles: Principle ofInvolves two principles: Principle of ReplicationReplication
and Principle ofand Principle of RandomizationRandomization..
Subjects areSubjects are randomlyrandomly assigned toassigned to
experimental treatments.experimental treatments.
ExampleExample: If out of 10 subjects, the study will: If out of 10 subjects, the study will
test 5 undertest 5 under treatment Atreatment A and 5 underand 5 under treatmenttreatment
BB, the randomization process gives every, the randomization process gives every
possible group of 5 subjects selected from a setpossible group of 5 subjects selected from a set
of 10 anof 10 an equal opportunityequal opportunity of being assigned toof being assigned to
treatment A and treatment B.treatment A and treatment B.
50. 5050
In C.R. Design, all the variations due toIn C.R. Design, all the variations due to
uncontrolleduncontrolled extraneous factorsextraneous factors are includedare included
under the heading ofunder the heading of chancechance variation.variation.
Two forms of the C.R. Design areTwo forms of the C.R. Design are::
Two-group Simple Randomized DesignTwo-group Simple Randomized Design
Random Replications DesignRandom Replications Design
Two-group Simple Randomized DesignTwo-group Simple Randomized Design
Population / Universe is defined.Population / Universe is defined.
From the population a sample is selectedFrom the population a sample is selected
Randomly.Randomly.
51. 5151
Items after being selectedItems after being selected randomlyrandomly from thefrom the
population, bepopulation, be randomlyrandomly assigned to theassigned to the
ExperimentalExperimental andand ControlControl groups.groups.
This yieldsThis yields two groupstwo groups as representatives ofas representatives of
the population on the basis of Principle ofthe population on the basis of Principle of
RandomizationRandomization..
Two groups (experimental and control) ofTwo groups (experimental and control) of
such design are givensuch design are given different treatmentsdifferent treatments..
RandomizesRandomizes thethe differencesdifferences among theamong the
sample items.sample items.
52. 5252
Individual differencesIndividual differences among thoseamong those
conducting the treatments are not eliminated.conducting the treatments are not eliminated.
FailureFailure of controlling theof controlling the extraneous variableextraneous variable
bring misleading results.bring misleading results.
ExampleExample: Comparing two groups of Ph.D.: Comparing two groups of Ph.D.
scholars who have beenscholars who have been randomlyrandomly selectedselected
and assigned.and assigned. Two differentTwo different treatments-treatments-
usual training and specialized training areusual training and specialized training are
given to two groups.given to two groups. Ho: Greater gains forHo: Greater gains for
the group receiving specialized trainingthe group receiving specialized training..
Study each groupStudy each group beforebefore andand afterafter thethe
training.training.
53. 5353
Compares theCompares the amount of gainamount of gain for the twofor the two
groups to accept or reject the Ho.groups to accept or reject the Ho.
Fails to controlFails to control the differential effects of thethe differential effects of the
extraneousextraneous independent variables.independent variables.
(Individual differences among those(Individual differences among those
conducting the training program).conducting the training program).
55. 5555
Random Replications DesignRandom Replications Design
ServesServes twotwo purposes: Providespurposes: Provides controlscontrols forfor
the differential effects of thethe differential effects of the extraneousextraneous
independent variables by providing aindependent variables by providing a
number ofnumber of repetitionsrepetitions for each treatment.for each treatment.
RandomizesRandomizes any individual differencesany individual differences
among thoseamong those conducting treatmentsconducting treatments..
In replication design there areIn replication design there are twotwo
populations.populations.
SampleSample is takenis taken randomlyrandomly from thefrom the
population available for study.population available for study.
56. 5656
TheThe samplesample isis randomlyrandomly assigned toassigned to
ExperimentalExperimental andand ControlControl groups.groups.
Sample is takenSample is taken randomlyrandomly from thefrom the
population available to conductpopulation available to conduct ExperimentsExperiments
and beand be randomly assignedrandomly assigned to the groups.to the groups.
Variables relating to both populationVariables relating to both population
characteristics are assumed to becharacteristics are assumed to be randomlyrandomly
distributed amongdistributed among twotwo groups.groups.
Example: Because of eight groups eightExample: Because of eight groups eight
individuals should be selected and randomlyindividuals should be selected and randomly
assigned to eight groups.assigned to eight groups.
57. 5757
Population
(Available for study) Population
(Available to conduct treatments)
Random Selection
Random Selection
Sample
( To be studied)
Sample
( To conduct treatments )
Random Assignment
Group 1
Group 2
Group 3
Group 4
Group 5
Group 6
Group 7
Group 8
E
E
E
E
C
C
C
C
Random Assignment
Treatment A
Treatment B
Independent Variable or Causal Variable
58. 5858
Developing a Research PlanDeveloping a Research Plan
After identifying and defining the problem,After identifying and defining the problem,
researcher must arrange the ideas andresearcher must arrange the ideas and
write them in the form of anwrite them in the form of an ExperimentalExperimental
Plan known as Research Plan.Plan known as Research Plan.
Research Plan must contain the following:Research Plan must contain the following:
ResearchResearch objectiveobjective should be clearlyshould be clearly
stated.stated.
Problem to be studied mustProblem to be studied must explicitlyexplicitly bebe
stated to know thestated to know the informationinformation to beto be
obtained for solving the problem.obtained for solving the problem.
59. 5959
EachEach major conceptmajor concept should be defined inshould be defined in
operational termsoperational terms which researcher wants towhich researcher wants to
measure.measure.
Should contain theShould contain the methodmethod to be used into be used in
solving the problem. An overall description ofsolving the problem. An overall description of
the approach to be adopted is usually given.the approach to be adopted is usually given.
Must state the details of theMust state the details of the techniquestechniques to beto be
adopted. (Data collection, Nature ofadopted. (Data collection, Nature of
instruments to be used etc.).instruments to be used etc.).
A clear mention of theA clear mention of the populationpopulation to beto be
studied should be made.studied should be made.
60. 6060
If the study happens to be sample based,If the study happens to be sample based,
the plan should state thethe plan should state the sampling plansampling plan i.e.,i.e.,
how the sample is to be identified.how the sample is to be identified.
Plan must contain thePlan must contain the methodsmethods to be used into be used in
processingprocessing the data.the data. StatisticalStatistical and otherand other
methods to be used must be indicated in themethods to be used must be indicated in the
plan.plan.
Result ofResult of pilot testpilot test, if any, should be, if any, should be
reported.reported.
TimeTime andand costcost budgets for the researchbudgets for the research
should be prepared and laid down in theshould be prepared and laid down in the
plan.plan.