larry hoyle institute for policy and social research university of kansas

43
Visualizing Two Social Networks Across Time with SAS®: Collaborators on a Research Grant vs. Those Posting on SAS- L Larry Hoyle Institute for Policy and Social Research University of Kansas 1 SGF2009 paper 229, Larry Hoyle

Upload: bonnie

Post on 24-Feb-2016

40 views

Category:

Documents


1 download

DESCRIPTION

Visualizing Two Social Networks Across Time with SAS®: Collaborators on a Research Grant vs. Those Posting on SAS-L. Larry Hoyle Institute for Policy and Social Research University of Kansas. Visualize These Data. Links. Nodes. A Social Network. Constellation Chart: Nodes. Nodes Have: - PowerPoint PPT Presentation

TRANSCRIPT

Implementing Stack and Queue Data Structures with SAS Hash Objects

Visualizing Two Social Networks Across Time with SAS:Collaborators on a Research Grant vs. Those Posting on SAS-LLarry HoyleInstitute for Policy and Social ResearchUniversity of Kansas

1SGF2009 paper 229, Larry HoyleVisualize These DataSGF2009 paper 229, Larry Hoyle2

NodesLinksA Social NetworkSGF2009 paper 229, Larry Hoyle3

Constellation Chart: NodesSGF2009 paper 229, Larry Hoyle4Nodes Have:Size (age)Color(gender)Tip (text)

Constellation Chart LinksSGF2009 paper 229, Larry Hoyle5Links Have:Width (Hours)Color(family)Tip (text)

Social Network GraphSGF2009 paper 229, Larry Hoyle6

Two SAS tools:Constellation Chart Applet (and Macro)Annotate FileConstellation Chart SliderSGF2009 paper 229, Larry Hoyle7Slider set to show only links with 19 or more hours spent together

Constellation Chart SliderSGF2009 paper 229, Larry Hoyle8Slider set to show only links with 14 or more hours spent together

Constellation Codetitle 'Mean Hours Spent Together';

%ds2const( ndata=Flints, ldata=FlintTimes, datatype=assoc,

minlnkwt=30, height=360, width=480, codebase=&jarpath, htmlfile=&outfile, colormap=y, fntsize=12,

nid=Person, nlabel=Person, nvalue=age, ncolor=gender, ncolfmt=Gcolor., ntip=ntip,

lfrom=PersonFrom, lto=PersonTo, lvalue=MeanHours, linktype=line, lcolor=linktype, lcolfmt=Lcolor., ltip=ltip, sclnkwt=N);SGF2009 paper 229, Larry Hoyle9FilesAppearanceNodesLinksTwo Different Sets of DataEach With Their Own ChallengesSAS-L (the SAS Listserv)Nodes are email addresses of posts (23,827)Links are posts to the same thread in the same year (267,209 messages to 82,279 threads ).

Kansas NSF EPSCoR GrantNodes are projects and nodes are peoplePeople have different roles (PI, researcher, support staff)Multiple types of links, together on:authorship, proposals, listed together in narrativeChanges across timeSGF2009 paper 229, Larry Hoyle10SAS-L Data Available on the WebSGF2009 paper 229, Larry Hoyle11

Linked-posting to the same threadData Cleaning Addresses ChangeSAS-L - Too Many Nodes for AppletApproach: Limit the number of nodesSGF2009 paper 229, Larry Hoyle12

SAS-L Those With Over 100 PostsSGF2009 paper 229, Larry Hoyle13

Most Links are With a Core GroupSGF2009 paper 229, Larry Hoyle14

Too Many Nodes for AppletApproach: Display All w/ SAS Annotate FileSGF2009 paper 229, Larry Hoyle15

SAS Annotate File Arrange NodesSGF2009 paper 229, Larry Hoyle16

How do you arrange the nodes in some meaningful way?All Nodes Around a Circle orMultidimensional Scaling of some or all nodes

proc mds data=SGF2009.TOPPOSTERSSIMILARITY out=SGF2009.TopPosters2D similar dimension = 2 level=ordinal;run;Problem: MDS on 23K nodes?SGF2009 paper 229, Larry Hoyle17

Scale the nodes with the most links (shown in red)Arrange the others randomly in a circle around them (shown in gray)Links to red nodes in blue, others in black

Zoom and Pan With AppletSGF2009 paper 229, Larry Hoyle18

With annotate Vector output (E.G.) RTF would allow zoom, but not tip on links3D with PROC G3D and AnnotateActiveX and Java Devices OnlySGF2009 paper 229, Larry Hoyle19

3D with PROC G3D and AnnotateGenerated in SAS 9.2SGF2009 paper 229, Larry Hoyle20

3D with PROC G3D and AnnotateGenerated From EG 4.1SGF2009 paper 229, Larry Hoyle21

3D with PROC G3D and AnnotateActiveX and Java Devices OnlySGF2009 paper 229, Larry Hoyle22

Kansas NSF EPSCoR Phase VVisualization NeedsShow relationships among 247 people And among 50 projectsShow change in collaboration across timeDifferentiate core peopleDifferentiate principal investigators (Pis)Differentiate institutionsAnimate across time

SGF2009 paper 229, Larry Hoyle23Projects Layer Arranged by People in Common Across all YearsSGF2009 paper 229, Larry Hoyle24

Core People Layer Arranged by Centroid of Projects to Which They BelongSGF2009 paper 229, Larry Hoyle25

People and LinksSGF2009 paper 229, Larry Hoyle26

PeopleColor indicates institutionWhite dot is Principal InvestigatorSize is count (e.g. publications)Large tan dot indicates core personLinksWidth represents count in commonPeople in Fixed Positions Allows Animation Across Time (2006)SGF2009 paper 229, Larry Hoyle27

People in Fixed Positions Allows Animation Across Time (2007)SGF2009 paper 229, Larry Hoyle28

People in Fixed Positions Allows Animation Across Time (2008)SGF2009 paper 229, Larry Hoyle29

Other Comparisons All Proposals and SubmissionsSGF2009 paper 229, Larry Hoyle30

Other Comparisons Successful ProposalsSGF2009 paper 229, Larry Hoyle31

Other Comparisons ProposalsSGF2009 paper 229, Larry Hoyle32

Other Comparisons Scientific ProductSGF2009 paper 229, Larry Hoyle33

Other Comparisons CombinedSGF2009 paper 229, Larry Hoyle34

Method ComparisonsAppletCoding is QuickSliderLink TipsMemory LimitsScreen Capture to PublishDynamic Pan and ZoomData Driven Color and SizeAnnotateAdditional Data StepsAnimated GIFHTML Link Tips (Difficult)Many Nodes PossibleHigh Quality ReproductionNo Tips (ODS Vector Output)Richer SymbologySGF2009 paper 229, Larry Hoyle35Animation Issues Fix Node PositionFix the position of nodes across all framesArrange in circleDimension reduction (MDS?)Example: KNEGIF.htmSGF2009 paper 229, Larry Hoyle36

Animation Issues - InterpolationDimension reduction that preserves orientation - then interpolate between observations SAS Example:could do something likeKansas Data Archive Bubble Plots

SGF2009 paper 229, Larry Hoyle37Chart from http://www.ipsr.ku.edu/ksdata/Inspired by Trendalyzer Software http://www.gapminder.org

Other ToolsSAS Graph NV WorkshopEnterprise Miner See paper 109-2009 Barry de Ville, Discover and Drive Brand Activity in Social NetworksSGF2009 paper 229, Larry Hoyle38Statistics - ClusteringClustering CoefficientGlobal Proportion of triads that have third linkSGF2009 paper 229, Larry Hoyle39BAC?When BA and BC are present,Is AC present?Statistics - BetweennessBetweenness CentralityIndividualSum of proportion of shortest paths that go through a given linkSGF2009 paper 229, Larry Hoyle40wxvyzContributing to Centrality for v wvz and wxz v is central 1 of 2 shortest w-z pathsStatistics - BetweennessBetweenness CentralityIndividualSum of proportion of shortest paths that go through a given linkSGF2009 paper 229, Larry Hoyle41wxvyzContributing to Centrality for v wvz and wxz v is central in 1 of 2 shortest w-z pathswvy - v is central in 1 of 1 shortest w-y pathsStatistics - BetweennessBetweenness CentralityIndividualSum of proportion of shortest paths that go through a given linkSGF2009 paper 229, Larry Hoyle42Contributing to Centrality for v wvz and wxz v is central in 1 of 2 shortest w-z pathswvy - v is central in 1 of 1 shortest w-y pathswx v is central in 0 of 1 shortest w-pathswxvyzQuestions?Larry [email protected] paper 229, Larry Hoyle43