Professional Documents
Culture Documents
Period4
Math1040
Atkinson
SkittlesTermProject
Thepurposeofthisprojectistoputtogethertheconceptswehavebeenlearningsofar
thisyear.Thewaywewilldothis,ishavingeverystudentsubmitdatafroma2.17oz.bagof
OriginalSkittles.Theyrecordthenumberofeachcolortheyhadandweusethecompilationof
thatdataforourproject.UsingaPiechartandParetochart,wewilldepicttheamountsof
skittles.
Thegraphdepictsthecolorsofskittles.Theonlyproblemisthattheorangesectionrepresents
theyellowtotal,andthebluesectionrepresentstheorangetotal.
Youcanseethatthedifferencesintheamountofskittlespercolordonothaveahugesignificant
difference.Ifinditinterestingthattherearealotofpurpleincomparisontored.Ithoughtit
wouldbetheotherwayaround.Mydatalookedlikethis:
Red
Orange
Yellow
Green
Purple
Total
MyBag
10
12
12
13
56
Class
totals
166
183
208
187
203
947
Theoveralldatadoesntmatchmydata.Thoughitsclose,minehaspurplebeingthemostand
greenbeingtheleast.
Tocreateafrequencyhistogramandboxplot,weneedtoknowthe5numbersummaryforthe
data.Thereareatotalof56bags,andmybaghad56piecesinit.
Thehistogramforthedatalookslikethis:
Youcantellfromthishistogramthatmanyofthebagshadatleast59piecesinthem.
Tocreateaboxplot,weneedthe5numbersummary.
Mean
59.2
StandardDeviation
2.46
Min
Q1
Med
Q3
Max
54
57.5
59.5
61.3
62
Lookingatbothofthegraphs,youcanseethatmostofthebagsofcandyhadaround
5762skittles.Thatsnotveryspreadoutwhichisnicebecauseyoucantellthatifyougotabag
with63candies,youwereaboveaverage,andifyougotonewith55candies,youwerealittle
below.Mybaghadatotalof56candies,whichisbelowaverage.Imonthelowerendofthe
scale.
Reflection
Thedifferencebetweencategoricaldataandquantitativedataisthatcategoricaldatais
datathattakesitsdatabasedoncategoriesorcharacteristics.Quantitativeisdatathatisbasedon
numbers.Inthiscase,categoricaldataisusedinthePieChartandParetoChartwiththeamount
ofskittlesofacertaincolor.Quantitativeisusedwhenfindingtheaveragenumberofskittlesper
bag,andinfindingthe5numbersummary.Visually,categoricaldatamakessensetobeusedin
apiechartbecauseitshowsthecolorswhichalmostmatchperfectlywiththeskittleswhereas
quantitativedataworksonaboxplotsinceyouusethe5numbersummary,whichyoucantuse
oncategoricaldata.Ifyoutriedtodoaboxplotforquantitativedataitwouldntwork.For
categoricaldata,theinformationusedusuallyisawholenumber.Quantitativedatacanhave
decimalsbecauseyourefindingthestandarddeviationandaverages.
ConfidenceIntervalEstimates
Ingeneral,thepurposeofaconfidenceintervalistoestimatetherangeofvaluesthetrue
proportion,mean,orstandarddeviationcouldbein.
Toconstructa95%confidenceintervalforthetruepopulationofpurplecandies,youcan
usethe1propZIntonthecalculator.
Xnumberofpurplecandies:203
Numberofcandiestotal:947
ConfidenceLevel:.95
Yougetanintervalof(0.18822,0.2405).
Toconstructa99%confidenceintervalfortheestimateofthetruemeannumberof
candiesperbagyoucanusetheTIntervalfunction.
Meanofthesample:59.19
Standarddeviation:2.46
Numberofcandies:947
Confidenceintervalof.99
Yougetanintervalof(58.949,59.369)candiesperbag.
A98%confidenceintervalestimateforthestandarddeviationwouldusethevariables
n=16,s=2.46,=.02,L=5.229R=30.578.Youendupwiththis:
1.723<<4.167
Theseintervalsgivetheintervalsinwhichthetrueproportion,mean,andstandarddeviationof
thecandieslie.
HypothesisTests
Thepurposeofahypothesistestistotesttheclaimaboutapropertyofapopulation.Whatwe
getfromahypothesistestisthepvaluewhichtellsusthelevelofsignificancewithinthetest.
Totesttheclaimthat20%ofallSkittlescandiesaregreen,wewouldusea1propZTest.Using
thesevalues:
Nullhypothesis:0.2
XnumberofSkittles:187
NumberoftotalSkittles:947
Thatgiveustheteststatistic
z =
0.195and
pvalueof0.1975.Sincethepvalueisgreaterthan
=.01,thereissufficientevidencetosupporttheclaimthat20%ofallSkittlescandiesaregreen.
Iftheclaimisthatthemeannumberofcandiesina2.17ozbagofSkittlesis56,wecantestthis
claimusingtheTTest.Youinputthesevalues:
NullHypothesis:56
Samplemean:59.19
StandarddeviationofX:2.46
ThenumberofSkittles:947
Yougetateststatisticoft=39.91andapvalueof0.Becausethepvalueislessthanthe
=.05,thereisinsufficientevidencetosupporttheclaimthatthemeannumberofcandiesina
2.17ozbagofSkittlesis56.
Theresultsshowthat20%ofSkittlesaregreenandthemeannumberofskittlesina2.17ozbag
isnot56.
Reflection
Theconditionsfordoingintervalestimatesandhypothesistestsforpopulation
proportionsare:1)thatthesampleisasimplerandomsample,2)eitherpopulationis>30or
standarddeviationisknown,3)requirementsforbinomialdistributionaresatisfied,and4)
np
5and
nq
5.Theserequirementsarenotmetforthehypothesistestsbecausetherearemorethan
oneoutcome.
Therequirementsfortestingapopulationmeanare:1)Simplerandomsampleand2)the
populationisnormallydistributedand/or
n>
30.Bothoftheserequirementsweresatisfiedinmy
testingofapopulationmean.
Conditionsfordoinganintervalestimateforapopulationstandarddeviationare:1)
Simplerandomsampleand2)Populationmustbenormallydistributedevenifitisalarge
sample.ThevaluesIusedarenormallydistributed.
Possibleerrorscouldhavebeenmadebynotcountingcorrectlyorusingthewrongtests
tocalculatethevalues.Thesamplingmethodcouldbeimprovedbygivingeveryonea2.17oz
baginsteadofhavingpeoplebuythewrongsizeandhavingtomakethevaluesup.