Business Research Methodology for Researhers.ppt

To equip students with knowledge of business research

1. Research Methods

2. What is Research
Therearemanywaysofdefining“research”
•Toresearchistocarryoutadiligentinquiryor
acriticalexaminationofagivenphenomena.
•Researchalsoinvolvesacriticalanalysisof
existingconclusionsortheorieswithregardto
newlydiscoveredfacts.
•Researchisasystematic,controlled,empirical,
andcriticalinvestigationofhypothetical
prepositionsaboutthepresumedrelations
amongnaturalphenomena.

3. What is Research
•Researchistheprocessofarrivingat
dependablesolutionsthroughasystematic
collection,analysisandinterpretationofdata.
•Researchisthecarefulandsystematicinquiry
intoorexaminationofafieldofknowledgein
ordertoestablishfactsandprinciples.
•Alldefinitionsemphasizethatresearchisa
process,notanevent.Itmustthereforebe
carefullyplanned,implemented,disseminated,
andconsumed.

4. Purpose of Research
•Themainpurposeofresearchistodiscover
newknowledge.
–Thisinvolvesthediscoveryofnewfacts,their
correctinterpretationandpracticalapplication.
•Secondly,istodescribeaphenomenon.
–Accurateidentificationofanyeventinvolves
thoroughdescription.

5. Purpose of Research
•Thirdly,istoenableprediction.
–Thisistheabilitytoestimatephenomenon.
–Wesometimesuseasetofvariablestopredicta
givenvariable.
•Thefourthpurposeofresearchistoenable
control.
–Inscientificresearchcontrolisconcernedwith
theabilitytoregulatethephenomenonunder
study.
–Manyscientificexperimentsaredesignedto
achievethisobjective.

6. Purpose of Research
•Thefifthpurposeofresearchistoenable
explanationofphenomenon.
–Explanationinvolvesaccurateobservationand
measurementofagivenphenomenon.
•Thesixthpurposeofresearchistoenable
theorydevelopment.
–Theorydevelopmentinvolvesformulatingconcepts,
lawsandgeneralizationaboutagivenphenomenon

7. Characteristics of Research
•Researchissystematic
•Researchiscontrolled
•Researchisempirical.Itdealswithdata,
whichistestedscientifically.
•Researchisself-correcting.Theresultsof
researchareopentopublicscrutiny

8. Natureof Research
•Itstrivestobeobjectiveandlogical.
•Itisbasedonobservableexperience
orempiricalevidence.
•Itischaracterizedbypatientand
unhurriedactivity.
•Itdemandsaccurateobservations,
reservationsanddescriptions.

9. Natureof Research
•It is directed towards the solution of
the problem.
•It is carefully recorded and reported.
•It requires expertise.
•Itinvolvesgatheringnewdatafrom
primaryorfirsthandsourcesorusing
existingdatafornewpurpose.

10. TYPES OF RESEARCH
Descriptivevs.Analytical
•Descriptiveresearchincludessurveysandfact-finding
enquiriesofdifferentkinds.
•Insocialscienceandbusinessresearchwequiteoftenuse
thetermExpostfactoresearchfordescriptiveresearch
studies.
•Themaincharacteristicofthismethodisthatthe
researcherhasnocontroloverthevariables;hecanonly
reportwhathashappenedorwhatishappening.
•Themethodsofresearchutilizedindescriptiveresearch
aresurveymethodsofallkinds,includingcomparativeand
correlationmethods.

11. •Inanalyticalresearch,ontheother
hand,theresearcherhastouse
factsorinformationalready
available,andanalyzetheseto
makeacriticalevaluationofthe
material.

12. Appliedvs.Fundamental
•Applied (or action)research
aimsatfindingasolutionforan
immediateproblemfacingasocietyor
anindustrial/businessorganization.
•Fundamental(tobasicorpure)
research ismainlyconcerned
withgeneralizationsandwiththe
formulationofatheory.

13. Quantitativevs.Qualitative
•Quantitativeresearchisbasedon
themeasurementofquantityor
amount.
•Qualitativeresearch,ontheother
hand,isconcernedwithqualitative
phenomenon,i.e.,phenomenarelating
toorinvolvingqualityorkind.

14. Conceptualvs.Empirical
•ConceptualResearch
isthatrelatedtosomeabstractidea(s)
ortheory.
•EmpiricalResearch
Itisdata-basedresearch,comingup
withconclusionswhicharecapableof
beingverifiedbyobservationor
experiment.Wecanalsocallitas
experimentaltypeofresearch.

15. SomeOtherTypesofResearch
•One-timeresearchorlongitudinalresearch.Intheformer
casetheresearchisconfinedtoasingletime-period,whereas
inthelattercasetheresearchiscarriedonoverseveraltime-
periods.
•Field-settingresearchorlaboratoryresearchorsimulation
research,dependingupontheenvironmentinwhichitistobe
carriedout.
•Clinicalordiagnosticresearch.Suchresearchfollowcase-
studymethodsorin-depthapproachestoreachthebasic
causalrelations.
•Conclusion-orientedanddecision-oriented.Whiledoing
conclusion-orientedresearch,aresearcherisfreetopickupa
problem,redesigntheenquiryasheproceedsandisprepared
toconceptualizeashewishes.

16. •Theexploratoryresearchoritmaybe
formalized.
•Theobjectiveofexploratoryresearchis
thedevelopmentofhypothesesratherthan
theirtesting,whereasformalizedresearch
studiesarethosewithsubstantialstructure
andwithspecifichypothesestobetested.
•Historicalresearchisthatwhichutilizes
historicalsourceslikedocuments,remains,
etc.tostudyeventsorideasofthepast,
includingthephilosophyofpersonsand
groupsatanyremotepointoftime.

17. Types of
Research
Fundamental
Research
Empirical
Research
Analytical
Research
Qualitative
Research
Conceptual
Research
Applied
Research
Descriptive
Research
Quantitative
Research
Historical
Research

18. Theoretical Analysis
Thepurposeofthispartofresearch
istopresentthetheoreticalanalysis
oftheissueorproblemyouare
investigating.Thisisalsodescribed
aspresentingyourtheoreticalmodel.

19. Empirical Testing of the
Analysis
Thepurposeoftheempiricaltesting
partoftheresearchreportisto
providetheempiricalevidencefor
yourresearchargument.Thetheme
ofthissectionofthepapercanbe
summarized as:Given your
hypothesis,howdidyoutestitand
whatwereyourfindings?

20. Empirical Testing of the
Analysis
This section should include:
The data used;
The empirical model and type of
statistical analysis you employed;
The results you hypothesized;
The actual results; and
Your interpretation of the results.

21. ResearchApproaches
•Two basicapproachestoresearch,viz.,
quantitativeapproachandthequalitativeapproach.
•QuantitativeApproachsub-classifiedintoinferential
,experimentalandsimulationapproachestoresearch.
•QualitativeApproachtoresearchisconcernedwith
subjectiveassessmentofattitudes,opinionsand
behavior.

22. Research Methods versus
Methodology
•Researchmethodsortechniques,thus,refertothe
methodstheresearchersuseinperformingresearch
operations.
•Thus,whenwetalkofresearchmethodologywenot
onlytalkoftheresearchmethodsbutalsoconsider
thelogicbehindthemethodsweuseinthecontextof
ourresearchstudyandexplainwhyweareusinga
particularmethodortechniqueandwhywearenot
usingotherssothatresearchresultsarecapableof
beingevaluatedeitherbytheresearcherhimselforby
others.

23. Define Research
Problem
Review of the
Literature
Formulate
Hypothesis
Design Research
Collection of
Data
Analyze the
Data
Interpret
&
Report
Flow Chart of Research
Process

24. Significance of Research
•Itprovidesthebasisfornearlyall
governmentpoliciesinoureconomic
system.
•Ithelpsinsolvingvariousoperationaland
planningproblemsofbusinessandindustry.
•Itisanaidtodecisionmaking.
•Itestablishestherelationbetween
variables.

25. Significanceof Research
•Itisequallyimportantforsocial
scientistsinstudying social
relationshipsandinseekinganswersto
varioussocialproblems.
•Itprovidesabasisforinnovation.
•Itfacilitatestheprocessofthinking,
analysis,evaluationandinterpretation
ofvarioussituation.

26. Criteriafor Good Research
•Purposeshouldbeclearlydefined.
•Commonconceptsshouldbeusedthatcanbe
understoodbyall.
•Researchprocedureshouldbeexplainedin
detail.
•Researchdesignshouldbecarefullyplanned.
•Researchershoulddeclareallthepossible
errorsandtheirpossibleimpactonfinding.

27. Criteriafor Good Research
•Analysisofdatashouldbesufficiently
adequatetorevealsignificance.
•Themethodsofanalysisshouldbe
appropriate.
•Thevalidityandreliabilityofthedata
shouldbecheckedcarefully.
•Theresearcherhavegoodcommandover
researchmethodologiesandshouldbe
intelligentandexperience.

28. Limitationsof Research
•Itisbasedonsample&samplingresearch
lacksthecompleteaccuracy.
•Longtimeisrequiredintheresearch
procedures.
•Difficulttoevaluatetheeconomicbenefits
derivedfromtheresearch.
•Trainedpersonnelandalotoftimeare
requiredforresearch.
•Lackofadequateknowledgeofresearch.

29. LITERATURE REVIEW

30. What is LR?
Aliteraturereviewdiscussespublished
informationinaparticularsubjectarea,
andsometimesinformationinaparticular
subjectareawithinacertaintimeperiod.
Aliteraturereviewcanbejustasimple
summaryofthesources,butitusuallyhas
anorganizationalpatternandcombines
bothsummaryandsynthesis.

31. What is LR?
Asummaryisarecapoftheimportant
informationofthesource,butasynthesisisare-
organization,orareshuffling,ofthatinformation.
Itmightgiveanewinterpretationofoldmaterial
orcombinenewwitholdinterpretations.
Oritmighttracetheintellectualprogressionof
thefield,includingmajordebates.
Anddependingonthesituation,theliterature
reviewmayevaluatethesourcesandadvisethe
readeronthemostpertinentorrelevant.

32. What is LR?
Theformatofareviewofliteraturemayvaryfrom
disciplinetodisciplineandfromassignmentto
assignment.
Areviewmaybeaself-containedunit--anendin
itself--oraprefacetoandrationaleforengagingin
primaryresearch.Areviewisarequiredpartof
grantandresearchproposalsandoftenachapterin
thesesanddissertations.
Generally,thepurposeofareviewistoanalyze
criticallyasegmentofapublishedbodyofknowledge
throughsummary,classification,andcomparisonof
priorresearchstudies,reviewsofliterature,and
theoreticalarticles.

33. What is LR?
Aliteraturereviewistheeffectiveevaluation
ofselecteddocumentsonaresearchtopic.
Areviewmayformanessentialpartofthe
researchprocessormayconstitutea
researchprojectinitself.
Inthecontextofaresearchpaperorthesis
theliteraturereviewisacriticalsynthesisof
previousresearch.
Theevaluationoftheliteratureleadslogically
totheresearchquestion.

34. What is LR?A ‘good’ literature review…..
….. is a synthesis of available research
….. is a critical evaluation
….. has appropriate breadth and depth
….. has clarity and conciseness
….. uses rigorous and consistent
methods
A ‘poor’ literature review is…..
…..an annotated bibliography
….. confined to description
….. narrow and shallow
….. confusing and longwinded
….. constructed in an arbitrary way

35. Why write LR?
Literaturereviewsprovideyouwithahandy
guidetoaparticulartopic.Ifyouhavelimited
timetoconductresearch,literaturereviews
cangiveyouanovervieworactasastepping
stone.
Literaturereviewsalsoprovideasolid
backgroundforaresearchpaper's
investigation.Comprehensiveknowledgeofthe
literatureofthefieldisessentialtomost
researchpapers.

36. Why write LR?
Forprofessionals,theyareuseful
reportsthatkeepthemuptodate
withwhatiscurrentinthefield.
Forscholars,thedepthandbreadth
oftheliteraturereviewemphasizes
thecredibilityofthewriterinhisor
herfield

37. Why write LR?
Thepurposeofaliteraturereviewisforyou
totakeacriticallookattheliterature(facts
andviews)thatalreadyexistsintheareayou
areresearching.
Aliteraturereviewisnotashoppinglistof
everythingthatexists,butacriticalanalysis
thatshowsanevaluationoftheexisting
literatureandarelationshipbetweenthe
differentworks.
Itdemonstratestherelevanceofthe
research.

38. Why write LR?
Literaturecanincludebooks,journalarticles,
internet(electronicjournals),newspapers,
magazines,thesesanddissertations,
conferenceproceedings,reports,and
documentaries.
Literaturereviewsarewrittenoccasionallyin
thehumanities,butmostlyinthesciencesand
socialsciences;inexperimentandlabreports,
theyconstituteasectionofthepaper.
Sometimesaliteraturereviewiswrittenasa
paperinitself.

39. Why write LR?
Inthecontextofaresearchpaperona
thesis,theliteraturereviewprovidesa
backgroundtothestudybeingproposed.
Thebackgroundmayconsideroneormoreof
thefollowingaspectsdependingonthe
researchquestionbeingposed:
Theoreticalbackground–past,presentorfuture
Clinicalpractice–previousorcontemporary
Methodologyand/orresearchmethods
Previousfindings
Rationaleand/orrelevanceofthecurrentstudy

40. Why write LR?
In a broader contextHart (1998) lists the
following purposes of a review:
Distinguishingwhathasbeendonefromwhatneeds
tobedone;
Discoveringimportantvariablesrelevanttothe
topic;
Synthesisingandgaininganewperspective;
Identifyingrelationshipsbetweenideasand
practice;
Establishingthecontextofthetopicorproblem;

41. Why write LR?
Rationalisingthesignificanceoftheproblem;
Enhancingandacquiringthesubject
vocabulary;
Understandingthestructureofthesubject;
Relatingideasandtheorytoapplications;
Identifyingmethodologiesandtechniquesthat
havebeenused;
Placingtheresearchinahistoricalcontextto
show familiaritywithstate-of-the-art
developments.

42. Why write LR?
Its purpose is to:
Placeeachworkinthecontextofitscontributionto
theunderstandingofthesubjectunderreview
Describetherelationshipofeachworktotheothers
underconsideration
Identifynewwaystointerpret,andshedlightonany
gapsin,previousresearch
Resolveconflictsamongstseeminglycontradictory
previousstudies
Identifyareasofpriorscholarshiptoprevent
duplicationofeffort
Pointthewayforwardforfurtherresearch
Placeone'soriginalwork(inthecaseofthesesor
dissertations)inthecontextofexistingliterature

43. What should I do before writing
the literature review?
Clarify
Ifyourassignmentisnotveryspecific,seek
clarificationfromyoursupervisor/lecturer:
Roughlyhowmanysourcesshouldyouinclude?
Whattypesofsources(books,journalarticles,
websites)?
Shouldyousummarize,synthesize,orcritiqueyour
sourcesbydiscussingacommonthemeorissue?
Shouldyouevaluateyoursources?
Shouldyouprovidesubheadingsandother
backgroundinformation,suchasdefinitionsand/or
ahistory?

44. What should I do before writing
the literature review?
Find models
Lookforotherliteraturereviewsinyourareaofinterest
orinthedisciplineandreadthemtogetasenseofthe
typesofthemesyoumightwanttolookforinyourown
researchorwaystoorganizeyourfinalreview.
Youcansimplyputtheword"review"inyoursearch
enginealongwithyourothertopictermstofindarticles
ofthistypeontheInternetorinanelectronicdatabase.
Thebibliographyorreferencesectionofsourcesyou've
alreadyreadarealsoexcellententrypointsintoyourown
research.

45. What should I do before writing
the literature review?
Narrow your topic
Therearehundredsoreventhousands
ofarticlesandbooksonmostareasof
study.Thenarroweryourtopic,the
easieritwillbetolimitthenumberof
sourcesyouneedtoreadinordertoget
agoodsurveyofthematerial.
Yourinstructorwillprobablynotexpect
youtoreadeverythingthat'soutthere
onthetopic,butyou'llmakeyourjob
easierifyoufirstlimityourscope.

46. What should I do before writing
the literature review?
Consider whether your sources are
current
Somedisciplinesrequirethatyouuse
informationthatisascurrentas
possible.Inthesciences,forinstance,
treatmentsformedicalproblemsare
constantlychangingaccordingtothe
lateststudies.Informationeventwo
yearsoldcouldbeobsolete.

47. What should I do before writing
the literature review?
However,ifyouarewritingareviewinthe
humanities,history,orsocialsciences,asurveyof
thehistoryoftheliteraturemaybewhatis
needed,becausewhatisimportantishow
perspectiveshavechangedthroughtheyearsor
withinacertaintimeperiod.
Trysortingthroughsomeothercurrent
bibliographiesorliteraturereviewsinthefieldto
getasenseofwhatyourdisciplineexpects.
Youcanalsousethismethodtoconsiderwhatis
"hot"andwhatisnot.

48. What should I do before writing
the literature review?
Find a focus
A literature review, like a term paper, is
usually organized around ideas, not the
sources themselves as an annotated
bibliography would be organized. This
means that you will not just simply list
your sources and go into detail about
each one of them, one at a time.
No.

49. What should I do before writing
the literature review?
Asyoureadwidelybutselectivelyinyourtopic
area,considerinsteadwhatthemesorissues
connectyoursourcestogether.
Dotheypresentoneordifferentsolutions?
Isthereanaspectofthefieldthatismissing?
Howwelldotheypresentthematerialanddothey
portrayitaccordingtoanappropriatetheory?
Dotheyrevealatrendinthefield?
Aragingdebate?
Pickoneofthesethemestofocustheorganization
ofyourreview.

50. What should I do before writing
the literature review?
Construct a working thesis
statement
Thenusethefocusyou'vefoundto
constructathesisstatement.Yes!
Literaturereviewshave thesis
statementsaswell!
However,yourthesisstatementwillnot
necessarilyargueforapositionoran
opinion;ratheritwillarguefora
particularperspectiveonthematerial.

51. What should I do before writing
the literature review?
Some sample thesis statements for
literature reviews are as follows:
The current trend in treatment for
congestive heart failure combines
surgery and medicine.
More and more cultural studies scholars
are accepting popular media as a subject
worthy of academic consideration.

52. What should I do before writing
the literature review?
Consider organization
You've got a focus, and you've narrowed
it down to a thesis statement.
Now what is the most effective way of
presenting the information?
What are the most important topics,
subtopics, etc., that your review needs
to include?
And in what order should you present
them?

53. What should I do before writing
the literature review?
Developanorganizationforyourreviewat
bothaglobalandlocallevel:
First,coverthebasiccategories
Justlikemostacademicpapers,literaturereviews
alsomustcontainatleastthreebasicelements:an
introductionorbackgroundinformationsection;the
bodyofthereviewcontainingthediscussionof
sources;and,finally,aconclusionand/or
recommendationssectiontoendthepaper.

54. What should I do before writing
the literature review?
Introduction:Gives a quick idea of the topic
of the literature review, such as the central
theme or organizational pattern.
Body:Contains your discussion of sources and
is organized either chronologically,
thematically, or methodologically (see below
for more information on each).
Conclusions/Recommendations:Discuss what
you have drawn from reviewing literature so
far. Where might the discussion proceed?

55. What should I do before writing
the literature review?
The introductionshould provide the reader with the
scale and structure of your review. It serves as a kind of
map.
The bodyof the review depends on how you have
organised your key points. Literature reviews at
postgraduate level should be evaluative and not merely
descriptive. For example possible reasons for similarities
or differences between studies are considered rather
than a mere identification of them.
The conclusionof the review needs to sum up the main
findings of your research into the literature. The
findings can be related to the aims of the study you are
proposing to do. The reader is thus provided with a
coherent background to the current study.

56. What should I do before writing
the literature review?
Organizing the body
Onceyouhavethebasiccategoriesinplace,thenyou
mustconsiderhowyouwillpresentthesources
themselveswithinthebodyofyourpaper.Createan
organizationalmethodtofocusthissectioneven
further.
Tohelpyoucomeupwithanoverallorganizational
frameworkforyourreview,considerthesixtypicalways
oforganizingthesourcesintoareview:
Chronological
By publication
By trend
Thematic
Methodological
Questions for Further Research

57. What should I do before writing
the literature review?
Similartoprimaryresearch,developmentofthe
literaturereviewrequiresfourstages:
Problemformulation—whichtopicorfieldis
beingexaminedandwhatareitscomponent
issues?
Literaturesearch—findingmaterialsrelevantto
thesubjectbeingexplored
Dataevaluation—determiningwhichliterature
makesasignificantcontributiontothe
understandingofthetopic
Analysisandinterpretation—discussingthe
findingsandconclusionsofpertinentliterature

58. What should you write?
the accepted facts in the area
the popular opinion
the main variables
the relationship between concepts and
variables
shortcomings in the existing findings
limitations in the methods used in the existing
findings
the relevance of your research
suggestions for further research in the area.

59. What should you write?
Literaturereviewsshouldcomprisethefollowingelements:
Anoverviewofthesubject,issueortheoryunder
consideration,alongwiththeobjectivesofthe
literaturereview
Divisionofworksunderreviewintocategories(e.g.
thoseinsupportofaparticularposition,thoseagainst,
andthoseofferingalternativethesesentirely)
Explanationofhoweachworkissimilartoandhowit
variesfromtheothers
Conclusionsastowhichpiecesarebestconsideredin
theirargument,aremostconvincingoftheiropinions,
andmakethegreatestcontributiontotheunderstanding
anddevelopmentoftheirareaofresearch

60. What should you write?
In assessing each piece, consideration should be given to:
Provenance—Whataretheauthor'scredentials?Are
theauthor'sargumentssupportedbyevidence(e.g.
primaryhistoricalmaterial,casestudies,narratives,
statistics,recentscientificfindings)?
Objectivity—Istheauthor'sperspectiveeven-handed
orprejudicial?Iscontrarydataconsideredoriscertain
pertinentinformationignoredtoprovetheauthor's
point?
Persuasiveness—Whichoftheauthor'sthesesare
most/leastconvincing?
Value—Aretheauthor'sargumentsandconclusions
convincing?Doestheworkultimatelycontributeinany
significantwaytoanunderstandingofthesubject?

61. What should you write?
Layout
Makeyourliteraturereviewhaveanacademicand
professionalappearance.Herearesomepointstomake
thelookofyourreportappealingtothereader
Whitespace:leavespacebetweensections,especially
fromtheabstract.Thisgivesanunclutteredeffect.
Headings/sub-headings:thesehelptoseparateideas.
Textboxes:youcanusetheseforquotationsor
paraphrasingtoseparatethemfromtherestofyour
text.Itisalsopleasingtotheeye.

62. What should you write?
Graphics:Centreyourgraphics,suchas
diagramsortables,tohavespacearound
them.Trynottoburygraphicsinyourtext.
Pagination:Youcannumberpagesorsections
orboth,buttheimportantthingtodoistobe
consistent.Thecoverpagenormallyisnot
numbered.Thecontentpageandabstract
pageusuallyhaveaseparatenumbering
systemtothebodyofyourliteraturereview.

63. What should you write?
Language focus
Createabalancebetweendirectquotation
(citation)andparaphrasing.Avoidtoomuchdirect
quoting.Theverbtensechosendependsonyour
emphasis:
Whenyouarecitingaspecificauthor'sfindings,
usethepasttense:(found,demonstrated);
Whenyouarewritingaboutanacceptedfact,use
thepresenttense:(demonstrates,finds);and
Whenyouarecitingseveralauthorsormakinga
generalstatement,usethepresentperfecttense:
(haveshown,havefound,littleresearchhasbeen
done).

64. What should you write?
Final checklist
Have I fulfilled the purpose of the literature review?
Is it written at a level appropriate to its audience?
Are its facts correct?
Is all the information included relevant?
Are the layout and presentation easy on the eye?
Is the language clear, concise and academic?
Does the abstract summarise the entire review?
Does the introduction adequately introduce the topic?
Is the body organised logically?
Does the conclusion interpret, analyse and evaluate?
Are the recommendations reasonable?
Does the table of contents correspond with the actual contents? Are page
numbers correct?
Have I acknowledged all sources of information through correct
referencing?
Have I checked spelling, grammar and punctuation?
Have I carefully proof-read the final draft?

65. How to review?
The whole process of reviewing
includes:
a. Searching for literature
b. Sorting and prioritising the retrieved
literature
c. Analytical reading of papers
d. Evaluative reading of papers
e. Comparison across studies
f. Organising the content
g. Writing the review

66. How to review?
Comparison across studies
The aim is to extract key points by comparing and
contrasting ACROSS studies, instead of reading
one paper after another.
Key points for a review may concern areas of
similarities and/or differences in:
Research aim(s) or hypotheses
Research design and sampling
Instruments and procedures used
How data were analysed
Results or findings
Interpretations

67. How to review?
Find similarities and differences between
studies at different levels, e.g.:
-philosophy
-epistemology
-morality
-methodology
-methods
-types of data
-data analysis
-interpretation

68. How to review?
Setoutyourthinkingonpaperthrough
mapsandtrees.Feature map Classifies and categorises your thought in tabular form

Concept map
Links between concepts and processes, or shows relationship between
ideas and practice

Tree construction
Shows how topic branches out into subthemes and related questions or
represents stages in the development of a topic.

69. How to review?
Tips on writingSentences
Express one idea in a sentence. Ensure that all your sentences have
a subject, verb and object.

Paragraphs
Group sentences that express and develop one aspect of your topic.
Use a new paragraph for another aspect or another topic.

Consistent Grammar
Use sentences and paragraphs with appropriate use of commas,
colours and semi-colours. Incorrect use of punctuation can affect
the meaning.

Transition Words
Use words that link paragraphs and which show contrast and
development to your argument e.g. ‘hence’, ‘therefore’, ‘but’,
‘thus’, ‘as a result’, ‘in contrast’.

70. How to review?
•Pitfalls
-Vagueness due to too much or
inappropriate generalisations
-Limited range
-Insufficient information
-Irrelevant material
-Omission of contrasting view
-Omission of recent work

71. Example
Earlyworkshaveaddressedsomeoftheproblemsand
issuesdiscussedinvideoretrieval.Researchershave
developedideasandtoolsforsupportingvideoediting,
forexamplein[8].Theyhavedefinedaseamlessvideo
editinginthegradientdomain.Thespatio-temporal
gradientfieldsoftargetvideosaremodifiedormixedto
generateanewgradientfield,whichisusuallynot
integrateable.
Theyhavealsodescribedhowsemanticinformationabout
videocanbestructuredandusedforcontent-based
access.Fromageneralvideoarchivepointofview,the
problemwiththistoolisthelackofsupportformanaging
videodocumentstructures.Adigitalvideoarchive
servingdifferentcategoriesofusersshouldoffera
morestructuredwayofdescribingvideocontents

72. Example
HiddenMarkovModels(HMMs)arestatisticaltoolsthat
havebeenusedsuccessfullyinmodellingdifficulttaskssuch
asspeechrecognition[15]orbiologicalsequenceanalysis
[16].Inspiredbyasimilarspeechapplication,HiddenMarkov
model(HMM)hasalsobeenappliedtoactivityrecognition.
Thefirstapproachforthehumanmovementsbasedon
HMMswasdescribedin[13].Itdistinguishedbetweensix
differenttennisstrokes.
Thissystemdividedtheimageintomeshesandcountedthe
numberofpixelsrepresentingthepersonforeachmesh.The
numberswerecomposedtoafeaturevectorthatwas
convertedintoadiscretelabelbyavectorquantizer.The
labelswereclassifiedbasedondiscreteHMMs.In[8],an
HMMisusedasarepresentationofsimpleactionswhichare
recognizedbycomputingtheprobabilitythatthemodel
producesthevisualobservationsequence.In[14]layered
HMMswereproposedtomodelsinglepersonofficeactivities
atvarioustimegranularities

73. Example
Mostoftheexistingworkreliesonusingonlyasinglesource
ofinformation(example,eitheraudioorvisualtrackdata
alone).In[4],theaveragevideoshotactivityandtheduration
areusedasfeaturesforthecategorizationofmovies
accordingtotheactions.
Anactionscenewascharacterizedbytemporallylocalized
propertiesofvideoshotswhichhavelittleornorecurring
similarvisualcontents[5].
Althoughthesevisualcharactersareundoubtedlygood
indicatorsofrapidlyevolvingactioncontents,theyarenot
enoughtodeterminethedesiredaction.
Ontheotherhand,audio-basedactiondetectionwas
independentlyperformedonthesoundtrackin[6].
However,thisaudioalonemethodmayleadtomanypotential
falsedetectedcasesbecausemanysoundsoftenmix
differentnoisesandothersimilarbackgroundsound.

74. ExampleResearch Technique Features
Used
Domain Disadvantage /
Advantage
Future Direction
Lin et al. 2007

A priori algorithm

Association rule
mining

Pre-filtering
architecture

Audiovisual Weather

Sports

Commercial
Reduce the amount of
misclassification errors.

Able to identify a high
percentage of positive
instances in each concept

Due to the different
properties of the data
sets representing the
semantic concepts
such as weather,
commercial, and
sports, they proposed
to use different
strategies to merge the
rules.

Davis & Tyagi
2006
Probabilistic reliable-
inference framework

Hidden Markov
Model (HMM) output
likelihoods and action
priors

Maximum likelihood
(ML) and maximum
a posteriori (MAP)
Motion Walking, running,
standing, bending-
forward, crouching-
down, and sitting
The system only makes
classifications when it
believes the input is
‘good enough’ for
discrimination between
the possible actions

75. STUDY POPULATION AND SAMPLING
PROCEDURE
Population:Theentiregroupaboutwhichinformationis
desired.
Thetheoreticalaggregationofspecifiedelementsasdefined
foragivensurveydefinedbytimeandspace.
Definitionofpopulationgivesaclearstatementofthose
included.Thefollowingaresomeexamples:
Adultsandchildren10-59yearsofageresidinginArua
Municipality
Adults 25-59 years of age residing in Arua Municipality on
July 1, 2014
Employees working in Arua District
Sample:Aproportionorpartofthepopulation-usuallythe
proportionfromwhichinformationisgathered.Orthe
aggregationofthepopulationfromwhichthesampleis
actuallydrawn.

76. Importance of Sampling
•Sample
–Small representative subset of a larger population
•Random sample
–Every subject had equal chance of being selected
•Representative sample
–Characteristics of participants correspond to larger
population
–Cost
–Time
–Reduce fatigue
–Eliminate errors

77. Basic Concepts in Sampling
Population.Referstotheentiregroupof
people,eventsorthingsofinterestthatthe
researcherwishtoinvestigate.Forinstance,if
managementofMUBSwantstoinvestigatethe
behaviorofstudentsinMUBS,thenallMUBS
studentsbecomesthepopulation.
Parameter.numericalcharacteristicofapopulation
Element.Anelementisasinglememberofthe
population.Giventheaboveexampleeach
studentisanelement.

78. Basic Concepts in Sampling
Populationframe.Thisislistingofallelementsinthe
populationfromwhichthesampleisdrawn.Thepayrollof
anorganizationwouldserveasthepopulationframeifits
membersaretobestudied.
Sample.Isasubsetofthepopulation.Itcomprisessome
membersselectedfromit.Itshouldbenotedthatnotall
elementswouldformsample.Asampleisthusasubgroup
orsubsetofthepopulation.Bystudyingthesamplethe
researcherdrawconclusionsthatgeneralizethe
populationinterest.
Statistic.Numericalcharacteristicofasample

79. Basic Concepts in Sampling
Subject.Thisisasinglememberofthesample,justlikeanelement
whichisasinglenumberofthepopulation.If200membersfromthe
totalpopulationof1,000blue-collarworkersformedthesampleof
thestudy,theneachbluecollarworkerinthesampleisasubject.
Sampling.Samplingistheprocessofselectingobservations(a
sample)toprovideanadequatedescriptionandrobustinferencesof
thepopulation.
Thesampleisrepresentativeofthepopulation.
Thecharacteristicspopulationmean(µ)populationstandard
deviation(∂)andpopulationvariance(∂)2areknownaspopulation
parameters.
Thecentraltendencies,thedispersionsandotherstatisticsinthe
sampleofinteresttotheresearcharetreatedasapproximations
ofthecentraltendencies,dispersionsandotherparametersofthe
population.
Alltheconclusionsdrawnaboutthesampleunderstudyare
generalizedtothepopulation.Inotherwords,thesample
statisticssuchasthesamplemean(X),standarddeviation(S)and
thevariationinsample(S)2areusedasestimatesofthe
populationparameters.


80. Sample Size determination
Determinethelevelofprecisionrequiredi.e.
samplingerrorcanbetolerated
DeterminetheSizeofthepopulation—sample
sizematterswithsmallpopulations
Decideonthevariationwithinthepopulation
withrespecttothecharacteristicof
interest—whatyouareinvestigating
Determinesmallestsubgroupwithinthe
sampleforwhichestimatesareneeded
Ensurethatthesampleisbigenoughto
properlyestimatethesmallestsubgroup

81. Sample Size determination

82. Sampling Techniques
Random/Probabilisticsampling.Inprobability
samplingtheelementsinthepopulationhave
someknownchancesorprobabilityofbeing
selectedassamplesubjects.
Asamplesize‘n’isdrawnfromapopulation‘N’insucha
waythateverypossibleelementinthepopulationhas
thesamechanceofbeingselected.
Non-random/Non-probabilisticsampling.In
non-probabilitysampling,theelementsdonot
haveknownorpredeterminedchanceof
selection.
Eachofthesetwomajordesignshasdifferent
samplingstrategies

83. Sampling Techniques
Dependingonthelevelofgeneralizationrequired,demandof
timeandotherresourcesaswellasthepurposeofthe
study;differenttypesofprobabilityandnon-probability
samplingdesignsarechosen.
Random/Probabilisticsamplingstrategies
Simplerandomsampling
Systematicsampling
Stratifiedrandomsampling
Clustersampling
Non-random/Non-probabilistic sampling
strategies
Convenience sampling
Purpose sampling
Judgment sampling
Quota sampling

84. strategies
SimpleRandomsamplingAsamplesize
‘n’isdrawnfromapopulation‘N’insuch
awaythateverypossibleelementin
thepopulationhasthesamechanceof
beingselected.
Eachmemberofthepopulationhasan
equalandknownchanceofbeing
selected.
Takeanumberofsamplestocreatea
samplingdistribution.
Typically conducted “without
replacement.
Whenthereareverylargepopulations,
itisoften‘difficult’toidentifyevery
memberofthepopulation,sothepool
ofavailablesubjectsbecomesbiased.

85. Systematic sampling
Systematicsamplingisoftenused
insteadofrandomsampling.Itisalso
calledanNthnameselectiontechnique.
Aftertherequiredsamplesizehasbeen
calculated,everyNthrecordisselected
fromalistofpopulationmembers.
Aslongasthelistdoesnotcontainany
hiddenorder,thissamplingmethodisas
goodastherandomsamplingmethod.
Itsonlyadvantageovertherandom
samplingtechniqueissimplicity(and
possiblycosteffectiveness).

86. Stratified Sampling
Stratifiedsamplingiscommonlyusedprobabilitymethodthatis
superiortorandomsamplingbecauseitreducessamplingerror.
Astratumisasubsetofthepopulationthatshareatleastone
commoncharacteristic;suchasmalesandfemales.
Identifyrelevantstratumsandtheiractual
representationinthepopulation.
Thepopulationisdividedbycertaincharacteristicsinto
homogeneoussubgroups(strata)(e.g.,MUBS PhD
students,MastersStudents,Bachelorsstudents).
Elementswithineachstrataarehomogeneous,butare
heterogeneousacrossstrata.
Randomsamplingisthenusedtoselectasufficient
numberofsubjectsfromeachstratum.
Stratifiedsamplingisoftenusedwhenoneormoreofthe
stratumsinthepopulationhavealowincidencerelativeto
theotherstratums.
Asimplerandomorasystematicsampleistakenfrom
eachstratarelativetotheproportionofthatstratumto
eachoftheothers.

87. Cluster Sampling
ClusterSample:aprobabilitysampleinwhich
eachsamplingunitisacollectionofelements.It
combinesSRS,SS,stratificationandcluster
sampling.
Used when:
Researchers lack a good sampling frame for a
dispersed population.
The cost to reach an element to sample is very high.
Examples of clusters:
City blocks –political or geographical
Housing units –college students
Hospitals –illnesses
Automobile –set of four tires

88. Non-random/Non-probabilistic sampling
strategies
•Judgmentsamplingisacommonnon-probability
method.
•Thesampleisselectedbaseduponjudgment.
–anextensionofconveniencesampling
•Whenusingthismethod,theresearchermust
beconfidentthatthechosensampleistruly
representativeoftheentirepopulation.

89. Quota Sampling
Quotasamplingisthenon-probabilityequivalentof
stratifiedsampling.
Firstidentifythestratumsandtheirproportions
astheyarerepresentedinthepopulation
Thenconvenienceorjudgmentsamplingisusedto
selecttherequirednumberofsubjectsfromeach
stratum.
Snowballsamplingisaspecialnon-probabilitymethod
usedwhenthedesiredsamplecharacteristicisrare.
Itmaybeextremelydifficultorcostprohibitiveto
locaterespondentsinthesesituations.
Thistechniquereliesonreferralsfrominitial
subjectstogenerateadditionalsubjects.
Itlowerssearchcosts;however,itintroducesbias
becausethetechniqueitselfreducesthelikelihood
thatthesamplewillrepresentagoodcrosssection
fromthepopulation.

90. •Theresearchdesignprovidesthe
backbonestructureofthestudy,it
supportsthestudyandholdit's
together.
Def:
•Theresearchdesignreferstothe
researcheroverallplanforanswering
theresearchquestionortestingthe
researchhypotheses
Research Design

91. Research Design
A research designis:
•aframeworkfortheresearchplanof
action.
•amasterplanthatspecifiesthemethods
andproceduresforcollectingand
analyzingtheneededinformation
•astrategyforhowthedatawillbe
collected.

92. Purposes of Research Design
•Itprovidestheschemeforanswering
researchquestion.
•Itmaintainscontroltoavoidbiasthatmay
affecttheoutcomes.
•Itorganizethestudyinacertainway
defendingtheadvantagesofdoingwhile
beingawareandcautionaboutpotential
disadvantages

93. Characteristics of good design
•1-Appropriateness to the research question.
•2-Lack of bias.
•3-Precision.
•4-Power.
•Research design can be either quantitative or
qualitative. Both designs complement each
other because they generate different kinds of
knowledge that are useful in nursing practice.

94. Designs for nursing
research
The four types most commonly used in
nursing are:
* descriptive / exploratory survey
*correlational
*experimental
*Quasi Experimental Design

95. A-Descriptive: (exploratory
survey) studies
•Its purpose is to provide a picture of a
situation as it naturally happen,
•No manipulation variable is involved
•Some studies contain two variable, others
may include multiple variables.
•They use this design to search for accurate
information about the characteristic s of
particular subject, group, and event or about
the frequency of the phenomena.

96. •Advantages of descriptive:
•The great deal with information
obtained from large population and is
surprisingly accurate.
•Disadvantagesof descriptive:
•The information collected is very brief
and superficial.
•It is a time consumer process

97. B-Correlational design
•It examine relationships between or
among two or more variables in a
single group.
•It requires a large sample size.
Subjects are not divided into groups
•The correlational design doesn’t test
the cause and effect relationship

98. Advantages of Correlational
studies:
•Anincreaseflexibilitywheninvestigatingcomplex
relationshipamongvariables.
•Aneffectiveandefficientmethodofcollectingalarge
amountofdataaboutanissueofinterest.
•Exploring a relationship between variables.
Disadvantages 0F Correlational studies
•Theresearcherisnotemployingtherandomizationinthe
samplingprocedures.
•Theresearcherisunabletodeterminethecausal
relationshipbetweenthevariablesbecauseoflackof
manipulation,controlandrandomization.

99. C) Experimental Research
Design
•Test a causal hypothesis (Cause and effect
relationship) between the intervention or
treatment and the outcomes
Elements of true experiment
design:
1-control group
2-sample randomization
3-manipulation of the independent variable.

100. Types of Experimental
Design
There are 3 types of experimental design:-
1-Pre test –post test control group design:-
•theexperimentalandcontrolgroupsareboth
randomlyassignedfromthesamplethatwas
randomlyselected.
•Thetreatmentisundercontrolofthe
researcher,thedependentvariablesis
measuredtwicebeforeandaftermanipulation
oftheindependentvariables.
•theresearcherobservesthetwogroupsto
determineeffectofmanipulation(posttest).

101. Advantages of Pre test –post test
control group design
•Threatstointernalvalidityareminimized
(controlgroup)
•Presenceofrandomizationleadstoequal
chanceforthesubjecttobeselected.
•Allowtheresearchertoexaminechanges
occurs.
•Disadvantages :-
•Theresultscannotbegeneralizedonboth
groups.
•Mortalitycanaffectonegroupratherthan
theother

102. 2-Post Test Only Experimental
Group Design:
•Thepretestcannotbeestablished,andalsosubject
responsetoposttestcanbealteredbytheeffectof
thepretest,soposttestonlyisperformed.
Advantages:
•Random assignment decreases the threats to validity.
Disadvantages
•Withouttheeffectofpretesttheresearcherbecome
unabletodeterminetheeffectofthetreatment
received.

103. 3-Solomon four groups
of experimental design
•Fourgroupscanbeusedoneexperimentalandthree
arecontrol,sometimesaredividedequallytwocontrol
andtheotherareexperimental.
•Itisastrongerdesignthantheposttestonlybut
requiresocomplicatedstatisticalanalysis.
•Advantages
•Allowtheexaminertoexaminetheeffectofthepre
testontheposttest.
•Disadvantages
•Mortality can affect the results.

104. General Advantages of
Experimental Design
•Are most appropriate for testing the cause and
effect relationship between the variables.
•Most appropriate for testing hypothesis.
General Disadvantages of Experimental Design
•Forsomeethicalreasonsmanipulationcannot
beachievedespeciallyonthehumanstudies.
•Pre test can affect the response of post test.

105. D) Quasi Experimental Design:-
•Quasiexperimentisresearchdesigninwhichthe
researcherinitiatesanexperimentaltreatmentbutsome
characteristicoftrueexperimentislackingeithercontrol
orrandomization.
•italsousedtotestcauseandeffectrelationship,thebasic
problemwiththequasiexperimentistheweakened
confidenceinmakingcausalassertionsoitisusually
exposedtothesubjectcontamination.
•Example: examining differences in the amount of pain that
the patient feel during painful procedure and the effect of
nursing intervention on the experimental group. It might be
impossible to pretest the amount of pain

106. How and When to use Quasi-
experimental research
•Usedwhenyouhavecontroloverthe
“whenandthewhomofmeasurement”,
butlackcontroloverthe“whenandto
whomofexposure.”(completecontrolis
notpossible)
•Common in educational research.

107. COCLUSION
•Aresearchdesignisamasterplanthat
specifiesthemethodsandproceduresfor
collectingandanalyzingtheneeded
information.
•Itisastrategyforhowthedatawillbe
collected.
•Itprovidestheschemeforanswering
researchquestion(s).
•Itmaintainscontroltoavoidbiasthatmay
affecttheoutcomes.

108. DATA ANALYSIS AND
INTERPRETATION
•Data analysis
–Anattemptbytheresearchertosummarizecollecteddata.
•Data Interpretation
–Attempttofindmeaning
•Thisisessentialforascientificstudyandforensuringthatwe
haveallrelevantdatafordrawingrelevantconclusions/inferences.
•Dataprocessinginvolvesediting,coding,classificationand
tabulationofcollecteddatasothattheyareamenabletoanalysis.
•Preparingdataforanalysis
•Data
•Anecessityforalmostanyenterprisetocarryoutitsbusiness.
Consistsofrawfacts,andwhenorganizedmaybetransformedinto
information.
•Afterdatacollectionhasbeendone,dataneedtobeprocessedand
analyzedinaccordancewiththelaiddownobjectiveintheresearch
plan.

109. Data Analysis
•Analysis not left until the end
•Toavoidcollectingdatathatarenot
importanttheresearchermustask:
–How am I going to make sense of this data?
•As they collect data the researcher must
ask
–Why do the participants act as they do?
–What does this focus mean?
–What else do I want to know?
–What new ideas have emerged?
–Is this new information?

110. Data Analysis:
Qualitative Data
•Onewayistofollowthreeiterativesteps
1.Becomefamiliarwiththedatathrough
1.Reading
2.Memoing
2.Examthedataindepthtoprovidedetailed
descriptionsofthesetting,participants,and
activities.
3.Categorizingandcodingpiecesofdataand
groupingthemintothemes.

111. Data Analysis: Qualitative Data
Summarizing
•“Thefirsttimeyousitdownwithyourdataisthe
onlytimeyoucometothatparticularsetfresh”-
Kratowohl.
–ReadingandMemoing
•Readwritememosaboutfieldnotes.
–Describing
•Developcomprehensive descriptionsofsetting,
participants,etc.
–Classifying
•Breakingdataintoanalyticunits.
•Categories
•Themes

112. Data Analysis Strategies :
Qualitative Data
•Identifying themes
–Begin with big picture and list “themes” that
emerge.
•Events that keep repeating themselves
•Coding qualitative data
–Reduce data to a manageable form
–Often done by writing notes on note cards and
sorting into themes.
•Predetermined categories vs. emerging categories

113. Data Interpretation: Qualitative
Data
•Answer these four questions
–What is important in the data?
–Why is it important?
–What can be learned from it?
–So what?
•Remember
–Interpretation depends on the
perspective of the researcher.
•Why?

114. Interpretation
•One technique for data
interpretation(Wolcott)
–Extend the analysis by raising questions
–Connect findings to personal
experiences
–Seek the advice of “critical” friends.
–Contextualize findings in the research
•Converging evidence?
–Turn to theory

115. Mixed Methods?
•Acombinationofquantitativeand
qualitativetechniques.
–Under what circumstances might mixed
methods work?
–Under what circumstances might mixed
methods not work?
•Think epistemological perspectives.

116. Definition and Purpose
•Mixedmethodsresearch
–Theresearchproblemitselfdetermines
thechoiceofadesign
–Examples:
•Usingsurveystoidentifyspecificgroupsof
studentsandconductingfocusgroupswith
themtounderstandtheirviews
•Aseriesofinterviewsareconductedto
ascertainthecriticalissuesbothering
students,andasurveyofthestudentbodyis
conductedusingtheseissuesasvariables

117. Quantitative vs. Qualitative

118. DATA ANALYSIS: QUANTITATIVE
DATA

119. QUANTITATIVE DATA ANALYSIS
•Editing:Editingofdataisaprocessofexaminingthe
collectedrawdata(speciallyinsurveys)todetecterrors
andomissionsandtocorrectthesewhenpossible.
•Asamatteroffact,editinginvolvesacarefulscrutinyof
thecompletedquestionnairesand/orschedules.
•Editingisdonetoassurethatthedataareaccurate,
consistentwithotherfactsgathered,uniformlyentered,as
completedaspossibleandhavebeenwellarrangedto
facilitatecodingandtabulation.

120. QUANTITATIVE DATA ANALYSIS
•Theymustalsopossessthecharacteristicofexhaustiveness(i.e.,
theremustbeaclassforeverydataitem)andalsothatofmutual
exclusivelywhichmeansthataspecificanswercanbeplacedinone
andonlyonecellinagivencategoryset.
•Anotherruletobeobservedisthatofunidimensionalitybywhich
ismeantthateveryclassisdefinedintermsofonlyoneconcept.
•Codingisnecessaryforefficientanalysisandthroughitthe
severalrepliesmaybereducedtoasmallnumberofclasseswhich
containthecriticalinformationrequiredforanalysis.
•Codingdecisionsshouldusuallybetakenatthedesigningstageof
thequestionnaire.

121. QUANTITATIVE DATA ANALYSIS
•Classification:Mostresearchstudiesresultinalargevolume
ofrawdatawhichmustbereducedintohomogeneous
groupsifwearetogetmeaningfulrelationships.
•Thisfactnecessitatesclassificationofdatawhichhappens
tobetheprocessofarrangingdataingroupsorclasseson
thebasisofcommoncharacteristics.
•Tabulation:Whenamassofdatahasbeenassembled,it
becomesnecessaryfortheresearchertoarrangethesame
insomekindofconciseandlogicalorder.Thisprocedureis
referredtoastabulation.Thus,tabulationistheprocessof
summarizingrawdataanddisplayingthesameincompact
form(i.e.,intheformofstatisticaltables)forfurther
analysis.

122. QUANTITATIVE DATA ANALYSIS
•Inabroadersense,tabulationisanorderlyarrangementof
dataincolumnsandrows.Tabulationisessentialbecauseof
thefollowingreasons.
Itconservesspaceandreducesexplanatoryand
descriptivestatementtoaminimum.
Itfacilitatestheprocessofcomparison.
Itfacilitatesthesummationofitemsandthedetectionof
errorsandomissions.
Itprovidesabasisforvariousstatisticalcomputations.
Tabulationcanbedonebyhandorbymechanicalor
electronicdevices.

123. QUANTITATIVE DATA ANALYSIS
•Dataanalysismeanthecomputationofcertainindicesor
measuresalongwithsearchingforpatternsofrelationship
thatexistamongthedatagroups.
•Analysis,incaseofsurveyorexperimentaldata,involves
estimatingthevaluesofunknownparametersofthe
populationandtestinghypothesesfordrawinginferences.
•Analysismay,therefore,becategorizedasdescriptive
analysisandinferentialanalysis(Inferentialanalysisis
oftenknownasstatisticalanalysis).
•Descriptiveanalysisislargelythestudyofdistributionsof
onevariable.hisstudyprovidesuswithprofilesof
companies,workgroups,personsandothersubjectsonany
ofamultipleofcharacteristicssuchassize.
•Composition,efficiency,preferences,etc.Thisanalysiscan
beinformofusingchartsand/orgraphs.

124. QUANTITATIVE DATA ANALYSIS
•Analysismaybeinrespectofonevariable(describedas
unidimensionalanalysis),orinrespectoftwovariables
(describedasbivariateanalysis)orinrespectofmorethan
twovariables(describedasmultivariateanalysis).
•CorrelationAnalysisandCausalAnalysis.Correlation
analysisstudiesthejointvariationoftwoormorevariables
fordeterminingthedegreeofcorrelationbetweentwoor
morevariables.
•Causalanalysisisconcernedwiththestudyofhowoneor
morevariablesaffectchangesinanothervariable.Itisthus
astudyoffunctionalrelationshipsexistingbetweentwoor
morevariables.

125. QUANTITATIVE DATA ANALYSIS
•Thisanalysiscanbetermedasregressionanalysis.Causalanalysisis
consideredrelativelymoreimportantinexperimentalresearches,
whereasinmostsocialandbusinessresearchesourinterestliesin
understandingandcontrollingrelationshipsbetweenvariablesthen
withdeterminingcausesperseandassuchweconsidercorrelation
analysisasrelativelymoreimportant.
•Inmoderntimes,withtheaidofcomputersoftware,therehasbeen
arapiddevelopmentofmultivariateanalysiswhichmaybedefinedas
allstatisticalmethodswhichsimultaneouslyanalyzemorethantwo
variablesonasampleofobservations.Usuallythefollowinganalyses
areinvolvedwhenwemakeareferenceofmultivariateanalysis:
•(a)Multipleregressionanalysis:Thisanalysisisadoptedwhenthe
researcherhasonedependentvariablewhichispresumedtobea
functionoftwoormoreindependentvariables.Theobjectiveofthis
analysisistomakeapredictionaboutthedependentvariablebased
onitscovariancewithalltheconcernedindependentvariables.

126. QUANTITATIVE DATA ANALYSIS
•b)Multiplediscriminantanalysis:Thisanalysisisappropriate
whentheresearcherhasasingledependentvariablethat
cannotbemeasured,butcanbeclassifiedintotwoormore
groupsonthebasisofsomeattribute.Theobjectofthis
analysishappenstobetopredictanentity’spossibilityof
belongingtoaparticulargroupbasedonseveralpredictor
variables.
•(c)Multivariateanalysisofvariance(ormulti-ANOVA):
ThisanalysisisanextensionoftwowayANOVA,whereinthe
ratioofamonggroupvariancetowithingroupvarianceis
workedoutonasetofvariables.
•(d)Canonicalanalysis:Thisanalysiscanbeusedincaseof
bothmeasurableandnon-measurablevariablesforthe
purposeofsimultaneouslypredictingasetofdependent
variablesfromtheirjointcovariancewithasetof
independentvariables.

127. QUANTITATIVE DATA ANALYSIS
•Inferentialanalysisisconcernedwiththevarioustestsof
significancefortestinghypothesesinordertodetermine
withwhatvaliditydatacanbesaidtoindicatesomeconclusion
orconclusions.
•Itisalsoconcernedwiththeestimationofpopulationvalues.
Itismainlyonthebasisofinferentialanalysisthatthetask
ofinterpretation(i.e.,thetaskofdrawinginferencesand
conclusions)isperformed.
•Theimportantstatisticalmeasuresthatareusedto
summarizethesurvey/researchdataare:
(1)Measuresofcentraltendencyorstatisticalaverages;
(2)Measuresofdispersion;
(3)Measuresofasymmetry(skewness);
(4)Measuresofrelationship;and(5)othermeasures.

128. QUANTITATIVE DATA ANALYSIS
•AmongsttheMeasuresofCentralTendency,thethreemost
importantonesarethearithmeticaverageormean,medianandmode.
Geometricmeanandharmonicmeanarealsosometimesused.
•FromamongtheMeasuresofDispersion,variance,andthestandard
deviationarethemostoftenusedmeasures.Othermeasuressuchas
meandeviation,range,etc.arealsoused.Forcomparisonpurpose,we
usemostlythecoefficientofvariation.
•InrespectoftheMeasuresofSkewnessandKurtosis,wemostly
usethefirstmeasureofskewnessbasedonmeanandmodeoron
meanandmedian.Othermeasuresofskewness,basedonquartilesor
onthemethodsofmoments,arealsousedsometimes.Kurtosisisalso
usedtomeasurethepeakednessofthecurveofthefrequency
distribution.
•AmongsttheMeasuresofRelationship,KarlPearson’scoefficientof
correlationisthefrequentlyusedmeasureincaseofstatisticsof
variables,whereasYule’scoefficientofassociationisusedincaseof
statisticsofattributes.Multiplecorrelationcoefficient,partial
correlationcoefficient,regressionanalysis.

129. THANK YOU