You are on page 1of 35

Chapter One

What is Statistics?

Copyright 2009 Cengage Learning

1.1

What is Statistics?
Statisticsisawaytogetinformationfromdata.

Copyright 2009 Cengage Learning

1.2

What is Statistics?
Statisticsisawaytogetinformationfromdata
Statistics
Data

Information

Statisticsisatoolforcreatingnewunderstandingfromasetof
numbers.
Definitions:OxfordEnglishDictionary
Copyright 2009 Cengage Learning

1.3

Example 2.6 Stats Anxiety


Astudentenrolledinabusinessprogramisattendingthefirst
classoftherequiredstatisticscourse.Thestudentissomewhat
apprehensivebecausehebelievesthemyththatthecourseis
difficult.
Toalleviatehisanxietythestudentaskstheprofessorabout
lastyearsmarks.
Theprofessorobligesandprovidesalistofthefinalmarks,
whichiscomposedoftermworkplusthefinalexam.What
informationcanthestudentobtainfromthelist?
Copyright 2009 Cengage Learning

1.4

Example 2.6 Stats Anxiety

Copyright 2009 Cengage Learning

1.5

Example 2.6 Stats Anxiety


Typical mark
Mean (average mark)
Median (mark such that 50% above and
50% below)
Mean = 72.67
Median = 72
Is this enough information?

Copyright 2009 Cengage Learning

1.6

Example 2.6 Stats Anxiety


Are most of the marks clustered around the
mean or are they more spread out?
Range = Maximum minimum = 92-53 =
39
Variance
Standard deviation

Copyright 2009 Cengage Learning

1.7

Example 2.6 Stats Anxiety


Are there many marks below 60 or above 80?
What proportion are A, B, C, D grades?
A graphical technique histogram can provide
us with this and other information

Copyright 2009 Cengage Learning

1.8

Example 2.6 Stats Anxiety

Copyright 2009 Cengage Learning

1.9

Descriptive Statistics
Descriptivestatisticsdealswithmethodsoforganizing,
summarizing,andpresentingdatainaconvenientand
informativeway.
Oneformofdescriptivestatisticsusesgraphicaltechniques,
whichallowstatisticspractitionerstopresentdatainwaysthat
makeiteasyforthereadertoextractusefulinformation.
Chapter2introducesseveralgraphicalmethods.

Copyright 2009 Cengage Learning

1.10

Descriptive Statistics
Anotherformofdescriptivestatisticsusesnumerical
techniquestosummarizedata.
Themeanandmedianarepopularnumericaltechniquesto
describethelocationofthedata.
Therange,variance,andstandarddeviationmeasurethe
variabilityofthedata
Chapter4introducesseveralnumericalstatisticalmeasures
thatdescribedifferentfeaturesofthedata.
Copyright 2009 Cengage Learning

1.11

Case 12.1 Pepsis Exclusivity


Agreement
Alargeuniversitywithatotalenrollmentofabout50,000

studentshasofferedPepsiColaanexclusivityagreementthat
wouldgivePepsiexclusiverightstosellitsproductsatall
universityfacilitiesforthenextyearwithanoptionforfuture
years.
Inreturn,theuniversitywouldreceive35%oftheoncampus
revenuesandanadditionallumpsumof$200,000peryear.
Pepsihasbeengiven2weekstorespond.

Copyright 2009 Cengage Learning

1.12

Case 12.1 Pepsis Exclusivity


Agreement
Themarketforsoftdrinksismeasuredintermsof12ounce
cans.

Pepsicurrentlysellsanaverageof22,000cansperweek(over
the40weeksoftheyearthattheuniversityoperates).
Thecanssellforanaverageof75centseach.Thecosts
includinglaboramountto20centspercan.
Pepsiisunsureofitsmarketsharebutsuspectsitis
considerablylessthan50%.
Copyright 2009 Cengage Learning

1.13

Case 12.1 Pepsis Exclusivity


Agreement
Aquickanalysisrevealsthatifitscurrentmarketsharewere
25%,then,withanexclusivityagreement,

Pepsiwouldsell88,000(22,000is25%of88,000)cansper
weekor3,520,000cansperyear.
Theprofitorlosscanbecalculated.
Theonlyproblemisthatwedonotknowhowmanysoft
drinksaresoldweeklyattheuniversity.
Copyright 2009 Cengage Learning

1.14

Case 12.1 Pepsis Exclusivity


Agreement
Pepsiassignedarecentuniversitygraduatetosurveythe
university'sstudentstosupplythemissinginformation.

Accordingly,sheorganizesasurveythatasks500studentsto
keeptrackofthenumberofsoftdrinkstheypurchaseinthe
next7days.
Theresponsesarestoredinafileonthediskthataccompanies
thisbook.Case12.1

Copyright 2009 Cengage Learning

1.15

Inferential statistics
TheinformationwewouldliketoacquireinCase12.1isan
estimateofannualprofitsfromtheexclusivityagreement.The
dataarethenumbersofcansofsoftdrinksconsumedin7days
bythe500studentsinthesample.
Wewanttoknowthemeannumberofsoftdrinksconsumed
byall50,000studentsoncampus.
Toaccomplishthisgoalweneedanotherbranchofstatistics
inferentialstatistics.

Copyright 2009 Cengage Learning

1.16

Inferential statistics
Inferentialstatisticsisabodyofmethodsusedtodraw
conclusionsorinferencesaboutcharacteristicsofpopulations
basedonsampledata.Thepopulationinquestioninthiscase
isthesoftdrinkconsumptionoftheuniversity's50,000
students.Thecostofinterviewingeachstudentwouldbe
prohibitiveandextremelytimeconsuming.Statistical
techniquesmakesuchendeavorsunnecessary.Instead,wecan
sampleamuchsmallernumberofstudents(thesamplesizeis
500)andinferfromthedatathenumberofsoftdrinks
consumedbyall50,000students.Wecanthenestimateannual
profitsforPepsi.

Copyright 2009 Cengage Learning

1.17

Example 12.5
Whenanelectionforpoliticalofficetakesplace,thetelevision
networkscancelregularprogrammingandinsteadprovide
electioncoverage.
Whentheballotsarecountedtheresultsarereported.
However,forimportantofficessuchaspresidentorsenatorin
largestates,thenetworksactivelycompetetoseewhichwill
bethefirsttopredictawinner.

Copyright 2009 Cengage Learning

1.18

Example 12.5
Thisisdonethroughexitpolls,whereinarandomsampleof
voterswhoexitthepollingboothisaskedforwhomthey
voted.
Fromthedatathesampleproportionofvoterssupportingthe
candidatesiscomputed.
Astatisticaltechniqueisappliedtodeterminewhetherthereis
enoughevidencetoinferthattheleadingcandidatewillgarner
enoughvotestowin.

Copyright 2009 Cengage Learning

1.19

Example 12.5
TheexitpollresultsfromthestateofFloridaduringthe2000
yearelectionswererecorded(onlythevotesoftheRepublican
candidateGeorgeW.BushandtheDemocratAlbertGore).
Supposethattheresults(765peoplewhovotedforeitherBush
orGore)werestoredonafileonthedisk.(1=Goreand2=
Bush)

Xm1205
Thenetworkanalystswouldliketoknowwhethertheycan
concludethatGeorgeW.BushwillwinthestateofFlorida.

Copyright 2009 Cengage Learning

1.20

Example 12.5
Example12.5describesaverycommonapplicationof
statisticalinference.
Thepopulationthetelevisionnetworkswantedtomake
inferencesaboutistheapproximately5millionFloridianswho
votedforBushorGoreforpresident.
Thesampleconsistedofthe765peoplerandomlyselectedby
thepollingcompanywhovotedforeitherofthetwomain
candidates.

Copyright 2009 Cengage Learning

1.21

Example 12.5
Thecharacteristicofthepopulationthatwewouldliketo
knowistheproportionofthetotalelectoratethatvotedfor
Bush.
Specifically,wewouldliketoknowwhethermorethan50%
oftheelectoratevotedforBush(countingonlythosewho
votedforeithertheRepublicanorDemocraticcandidate).

Copyright 2009 Cengage Learning

1.22

Example 12.5
Becausewewillnotaskeveryoneofthe5millionactual
votersforwhomtheyvoted,wecannotpredicttheoutcome
with100%certainty.
Asamplethatisonlyasmallfractionofthesizeofthe
populationcanleadtocorrectinferencesonlyacertain
percentageofthetime.
Youwillfindthatstatisticspractitionerscancontrolthat
fractionandusuallysetitbetween90%and99%.

Copyright 2009 Cengage Learning

1.23

Key Statistical Concepts


Population
apopulationisthegroupofallitemsofinterestto
astatisticspractitioner.
frequentlyverylarge;sometimesinfinite.
E.g.All5millionFloridavoters,perExample12.5

Sample
Asampleisasetofdatadrawnfromthe
population.
Potentiallyverylarge,butlessthanthepopulation.
E.g.asampleof765votersexitpolledonelectionday.
Copyright 2009 Cengage Learning

1.24

Key Statistical Concepts


Parameter
Adescriptivemeasureofapopulation.
Statistic
Adescriptivemeasureofasample.

Copyright 2009 Cengage Learning

1.25

Key Statistical Concepts


Population

Sample

Subset

Parameter

Statistic

PopulationshaveParameters,
SampleshaveStatistics.
Copyright 2009 Cengage Learning

1.26

Descriptive Statistics
aremethodsoforganizing,summarizing,andpresenting
datainaconvenientandinformativeway.Thesemethods
include:
GraphicalTechniques(Chapter2),and
NumericalTechniques(Chapter4).

Theactualmethoduseddependsonwhatinformationwe
wouldliketoextract.Areweinterestedin
measure(s)ofcentrallocation?and/or
measure(s)ofvariability(dispersion)?

DescriptiveStatisticshelpstoanswerthesequestions
Copyright 2009 Cengage Learning

1.27

Inferential Statistics
DescriptiveStatisticsdescribethedatasetthatsbeing
analyzed,butdoesntallowustodrawanyconclusionsor
makeanyinterferencesaboutthedata.Henceweneed
anotherbranchofstatistics:inferentialstatistics.
Inferentialstatisticsisalsoasetofmethods,butitisusedto
drawconclusionsorinferencesaboutcharacteristicsof
populationsbasedondatafromasample.

Copyright 2009 Cengage Learning

1.28

Statistical Inference
Statisticalinferenceistheprocessofmakinganestimate,
prediction,ordecisionaboutapopulationbasedonasample.
Population
Sample
Inference

Statistic
Parameter

WhatcanweinferaboutaPopulationsParameters
basedonaSamplesStatistics?
Copyright 2009 Cengage Learning

1.29

Statistical Inference
Weusestatisticstomakeinferencesaboutparameters.
Therefore,wecanmakeanestimate,prediction,ordecision
aboutapopulationbasedonsampledata.
Thus,wecanapplywhatweknowaboutasampletothe
largerpopulationfromwhichitwasdrawn!

Copyright 2009 Cengage Learning

1.30

Statistical Inference
Rationale:
Largepopulationsmakeinvestigatingeachmemberimpractical
andexpensive.
Easierandcheapertotakeasampleandmakeestimatesaboutthe
populationfromthesample.

However:
Suchconclusionsandestimatesarenotalwaysgoingtobecorrect.
Forthisreason,webuildintothestatisticalinferencemeasuresof
reliability,namelyconfidencelevelandsignificancelevel.

Copyright 2009 Cengage Learning

1.31

Confidence & Significance Levels


Theconfidencelevelistheproportionoftimesthatan
estimatingprocedurewillbecorrect.
E.g.aconfidencelevelof95%meansthat,estimatesbasedonthis
formofstatisticalinferencewillbecorrect95%ofthetime.

Whenthepurposeofthestatisticalinferenceistodrawa
conclusionaboutapopulation,thesignificancelevel
measureshowfrequentlytheconclusionwillbewronginthe
longrun.
E.g.a5%significancelevelmeansthat,inthelongrun,thistype
ofconclusionwillbewrong5%ofthetime.

Copyright 2009 Cengage Learning

1.32

Confidence & Significance Levels


Ifweuse(Greekletteralpha)torepresentsignificance,
thenourconfidencelevelis1.
Thisrelationshipcanalsobestatedas:
ConfidenceLevel
+SignificanceLevel
=1

Copyright 2009 Cengage Learning

1.33

Confidence & Significance Levels


Considerastatementfrompollingdatayoumayhearabout
inthenews:
This poll is considered accurate within 3.4
percentage points, 19 times out of 20.

Inthiscase,ourconfidencelevelis95%(19/20=0.95),
whileoursignificancelevelis5%.

Copyright 2009 Cengage Learning

1.34

Statistical Applications in Business


Statisticalanalysisplaysanimportantroleinvirtuallyall
aspectsofbusinessandeconomics.
Throughoutthiscourse,wewillseeapplicationsofstatistics
inaccounting,economics,finance,humanresources
management,marketing,andoperationsmanagement.

Copyright 2009 Cengage Learning

1.35

You might also like