View
1
Download
0
Category
Preview:
Citation preview
!"#$%&'($()*%$'+,-.'),%&%/*(*#-/'01'2#3('4"$(0(5'6,()7#8'9%&#:/%,
!"#$%#$&'()*+"#,*--.;%&&-/&'<,-.'=$>(,$'?@+*%
Il lustrated by Merchantfor the Brunswick Review
WhoIsEdwardTufte?ananalyticaldesigntheorist,educator,andlandscapesculptorbestknownforhisself-publishedbooksonanalyticaldesign
Anscombe’s Quartet
x y x y x y x y10 8.04 10 9.14 10 7.46 8 6.588 6.95 8 8.14 8 6.77 8 5.7613 7.58 13 8.74 13 12.7 8 7.719 8.81 9 8.77 9 7.11 8 8.8411 8.33 11 9.26 11 7.81 8 8.4714 9.96 14 8.1 14 8.84 8 7.046 7.24 6 6.13 6 6.08 8 5.254 4.26 4 3.1 4 5.39 19 12.512 10.8 12 9.13 12 8.15 8 5.567 4.82 7 7.26 7 6.42 8 7.915 5.68 5 4.74 5 5.73 8 6.89
I II III IV
Howwouldyoudescribethedifferencesinthesedatasets?
Anscombe’s QuartetMetric Valuemeanoftheyvalues 7.5equationoftheleast-squaredregressionline y =3+0.5x sumsofsquarederrors(aboutthemean) 110regressionsumsofsquarederrors(varianceaccountedforbyx) 27.5residualsumsofsquarederrors(abouttheregressionline) 13.75correlationcoefficient 0.82coefficientofdetermination 0.67
;$2,0<=9!2 >6*)898
6,()7#8&'!"#"$%$(*(
;$08"9)'9?*<+-9@'A9*)20$'B0))9-*8#0$
Otherreasons?
Visualizationisthehighestbandwidthchannelintothehumanbrain
Asdatavolumesgrow,visualizationbecomesanecessityratherthanaluxury.◦ “Apictureisworthathousandwords”
--Grolemund &Wickham,RforDataScience,O'Reilly2016
Visualizationisimportant…
GraphicalExcellence1. Showthedata
2. Inducetheviewertothinkaboutsubstanceratherthanmethodology
3. Avoiddistortion
4. Presentmanynumbersinasmallspace
5. Makelargedatasetscoherent
6. Encouragecomparisons
7. Revealthedataatseverallevelsfrombroadoverviewtofinestructure
8. Serveapurpose
9. Beintegratedwithstatisticalandverbaldescriptionsofdata
-- PrinciplesofGraphicalExcellence,EdwardTufte
6,()7#8("'%P8%""%/8%':#D%&'*7%'D#%>%,cY *7%':,%(*%&*'d'-+'#$%(&'Y #/'*7%'&7-,*%&*'*#.%Y>#*7'*7%'"%(&*'#/3Y #/'*7%'&.(""%&*'&)(8%I
C$9'07'8"9'0-4928'&)*+"#,*-'4#2+-*.2e@'f7#'?7@6$'+74+5("+5!$*89+74+:3+5("+&!"$5
f(,D%$'#/'&*-/%'#/'SSMH'4I9I'Q#$%('),-0(0"1'-"$%,R
g-*7#/:'&#.#"(,'#/'=@,-)%'@/*#"'hSJJO
^/%'-+'*7%'%(,"#%&*'*#.%&%,#%& 87(,*&
G#79H,.,-9'07'8"9'I*+*$929'J998-9
;I'F@:7'g%>.(/'6$.+$.0+;.9"*59+<=7.07./+>?@AB/+''+>CDE>CA
StriveForGraphicalIntegrity
Visualrepresentationsofdatamusttellthe
truth.
488-,$#/:'*-'?@+*%'*7%';#%'<(8*-,'-+'*7#&':,()7'#&'S`IUI''4'/@.%,#8("'87(/:%'-+'JMi'#&',%),%&%/*%$'01'(':,()7#8("'87(/:%'Q&#X%'-+'7-,#X-/*("'"#/%&R'-+'HUMiI
!"9'G#9'M*,80)
;#%'<(8*-,'L !#X%'-+'%++%8*'#/'*7%'D#&@("#X(*#-/!#X%'-+'%++%8*'#/'*7%'$(*(
GraphicalIntegrityTherepresentationofnumbers,asphysicallymeasuredonthesurfaceofthegraphitself,shouldbedirectlyproportionaltothenumericalquantitiesrepresented
?7%'/@.0%,'-+'#/+-,.(*#-/'8(,,1#/:'QD(,#(0"%R'$#.%/&#-/&'$%)#8*%$'&7-@"$'/-*'%P8%%$'*7%'/@.0%,'-+'$#.%/&#-/&'#/'*7%'$(*(I
()*+"#,*-'1$89&)#8.
6,()7#8&'.@&*'/-*'K@-*%'$(*('-@*'-+'
8-/*%P*I
()*+"#,*-'1$89&)#8.
6,()7#8&'.@&*'/-*'K@-*%'$(*('-@*'-+'
8-/*%P*I
()*+"#,*-'1$89&)#8.
6,()7#8&'.@&*'/-*'K@-*%'$(*('-@*'-+'
8-/*%P*I
()*+"#,*-'1$89&)#8.
6,()7#8&'.@&*'/-*'K@-*%'$(*('-@*'-+'
8-/*%P*I
()*+"#,*-'1$89&)#8.
N*?#<#E9'O*8*'1$%
3?*<+-9@'P#&"'0)'G05'O*8*'1$%')*8#0:
N*?#<#E9'O*8*'1$%
?,#.'*7%'+"@++'+,-.'1-@,':,()7#8&'(&'.@87'(&')-&�"%j
AvoidChartJunk
Theexcessiveandunnecessaryuseofgraphicaleffectsingraphsusedtodemonstratethe
graphicabilityofthedesignerratherthandisplaythedata.
Possiblytheworstgraphevermade
5Colors2Curvylines2Polygons2BrokenAxis3DimensionsTodisplay:
CHARTJUNK
5numbers
Aimforhighdatadensity
Theproportionofthetotalsizeofthegraph
thatisdedicateddisplayingdata.
Shrink
PrincipleManygraphscanbeshrunkwaydownwithoutlosinglegibilityorinformation
Seriesofthesamesmall graphrepeated
inonevisual
SmallMultiples
ClutterVisualizationtoocluttered?
Don'tremovedata,changethedesign.
Credibilitycomesfromdetailandinmanycasesonecanclarifya
designbyaddingdetail.
HyperaktandEkeneIjeomavisualizedmigrationsovertimeandspaceinTheRefugeeProjecthttp://www.therefugeeproject.org
Layering&
Separation
Usecolororotherdifferentiationtoseparateimportant
classesofinformation.
Mutedcolors,subtleshadingandthincontourlinesallowmultipletypesofdatatobelayeredtogetherinthis1958topographicmapofChattanooga,Tennessee.
Effectivelayeringofinformationisdifficultbecauseofinteractions
Can create non-informationpatternsandtexture
1+1=3(ormore)
N0#)9 9779,8
Selecttheappropriategraphicforthedata&analysis
Chart by Andrew Abela
Aesthetics&Technique
Carefullychosengraphicalformat
Designuseswords,numbers,anddrawings
Displaysanaccessiblecomplexityofdetail
Hasanarrativequality,astorytotellaboutthedata
Technicaldetailsofproductiondonewithcare
Avoidcontent-freedecoration,includingchartjunk
JuanVelasco.CornellornithologistEdwingScholesandbiologist/photographerTimLaman.SeniorGraphicsEditorFernandoBaptista,GraphicsSpecialistMaggieSmithandfreelanceresearcherFannaGebreyesus.NationalGeographic
Thefinalword…
Theprinciplesshouldnotbeappliedrigidlyorinapeevishspirit;theyare
notlogicallyormathematicallycertain;anditisbettertoviolate
anyprinciplethantoplacegracelessorinelegantmarksonpaper
--Tufte,VisualDisplayofQuantitativeInformation, page191
Recommended