DOC

A BRIEF GUIDE TO USING NETDRAW

By Elsie Harper,2014-07-05 11:52
7 views 0
A BRIEF GUIDE TO USING NETDRAW. NETDRAW IS A PROGRAM FOR DRAWING SOCIAL NETWORKS. OVERVIEW OF FEATURES. MULTIPLE RELATIONS. YOU CAN READ IN MULTIPLE RELATIONS ...

    A Brief Guide to Using NetDraw

NetDraw is a program for drawing social networks.

    Overview of Features

    Multiple Relations. You can read in multiple relations on the same nodes, and switch between them (or combine them) easily.

    Valued Relations. If you read in valued data, you can sequentially “step” through

    different levels of dichotomization, effectively selecting only strong ties, only weak ties, etc. In addition, you have the option of letting the thickness of lines correspond to strength of ties.

    Node Attributes. The program makes it convenient to read in multiple node attributes for use in setting colors and sizes of nodes (as well as rims, labels, etc.). In addition, the program makes it easy to turn on and off groups of nodes defined by a variable, such as males or members of a given organization. If the attributes are read in using the VNA data format (see below), they can be textual in addition to numeric. This means that instead of coding location as numeric codes 1, 2, 3, etc, you can simply write Boston, New York, Tokyo …

    Analysis. A limited set of analytical procedures are included, such as the identification of isolates, components, k-cores, cut-points and bi-components (blocks).

    2-Mode Data. NetDraw can read 2-mode data, such as the Davis, Gardner and Gardner data and automatically create a bipartite representation of it.

    Data Formats. The program reads Ucinet datasets (the ##h and ##d files), Ucinet DL text files, Pajek files (net, clu and vec), and the program’s own VNA text file format, which

    allows the user to combine node attributes with tie information.

    Saving Data. Using the VNA file format, the program can save a network along with its spatial configuration, node colors, shapes, etc. so that the next time you open the file, the network looks exactly like it looked before. The program can also save data as Pajek net and clu files, and Ucinet datasets (both networks and attributes).

    Saving Pictures. Network diagrams can be saved as bitmaps (.bmp), jpegs (.jpg), windows metafiles (.wmf) and enhanced metafiles (.emf). In addition, the program exports to Pajek and Mage.

    Printing. There is a Print button. This is very good for creating publication-quality diagrams because the results utilize the full resolution of the printer. (When you save an

    image like a bitmap to disk and insert into a document and then print that, the image resolution is no better than your screen’s.)

    Appearance Options. A full range of options is implemented, including the ability to change sizes and colors of nodes, node-rims, labels, lines and background. Different node shapes are not yet implemented. You can also rotate, flip, shift, resize and zoom configurations.

    Layout. Two basic kinds of layouts are implemented at present: a circle and an MDS/ spring embedding based on geodesic distance. The MDS includes options for exaggerating clustering, biasing toward equal-length edges, and turning on/off node-repulsion.

Getting Started

    Suppose you have a network currently stored as a Ucinet dataset. To draw it in network, just press the Open File button on the toolbar and select the file. The rest is automatic. Here is an example of drawing a file called campnet:

    BILL

    DON

    HARRY

    MICHAEL

    HOLLY

    GERYLEEPAT

    STEVEJENNIE

    PAM

    BRAZEYRUSSJOHNPAULINEBERTANN

    CAROL

    Now suppose you have some information about each person that you would like to use to in the display. You might enter the information in a text file called CampAttribs.txt (e.g., enter it in Excel and Save As text file) in the following format:

*node data

    id gender role betweenness

    HOLLY female participant 78.33333588

    BRAZEY female participant 0

    CAROL female participant 1.333333373

    PAM female participant 32.5

    PAT female participant 39.5

    JENNIE female participant 6.333333492

    PAULINE female participant 12.5

    ANN female participant 0.5

    MICHAEL male participant 58.83333206

    BILL male participant 0

    LEE male participant 5

    DON male participant 16.33333397

    JOHN male participant 0

    HARRY male participant 2.333333254

    GERY male instructor 54.66666794

    STEVE male instructor 16.83333397

    BERT male instructor 13.66666698

    RUSS male instructor 47.33333206

    Note that the values do not need to be numeric. They should be separated from each other by a comma, space or tab (and values that contain spaces should be enclosed in quotes as in “Bill Smith”).

    To read this file, go to File|Open|VNA|Attributes and select the file. This will read the file and open the Node Selector window, which looks like this:

    Using this window you can select an attribute (ID is selected by default), and then use that click on and off nodes with specific properties, such as females or instructors.

    Now suppose you want to change the colors of nodes to reflect a node attribute, such as wanting to paint men blue and women red. Go to Properties|Nodes|Colors|By Attribute. A dialogue box will open which lets you choose the attribute (gender) and then choose the color of each gender. The dialogue box looks like this:

    You might also want to change the shape of the nodes to reflect the role that person plays in the group (as indicated by the Role variable). To do this, go to

    Properties|Nodes|Shape|by attribute, which opens a dialogue box very similar to the color box, except instead of colors, there are shapes. Something similar can be done with the size of nodes.

    After setting each gender to the desired color, each role to desired shape, and making the size of the nodes proportional to their betweenness centrality, the network diagram looks like this:

    BILL

    DON

    HARRY

    MICHAELHOLLY

    PAT

    GERY

    PAMJENNIELEESTEVE

    CAROLPAULINERUSSJOHNBRAZEY

    BERTANN

    VNA Data Format

    The VNA data format allows the user to store not only network data but also attributes of the nodes, along with information about how to display them (color, size, etc.). A key feature of VNA attribute data is that textual data is permitted. In other words, instead of using numeric codes, the gender variable can have values like “male” and “female”.

Here is a short example of a vna file:

*node data

    ID name gender age

    j101 joe male 56

    w067 wendy female 23

    b303 bill male 48

    *tie data

    from to friends advice

    j101 w067 1 3

    w067 j101 0 1

    j101 b303 1 2

    w067 b303 0 6

VNA files are ordinary text files. They consist of sections called “star sections”. Not

    every file has to have every possible star section, and sections can be in any order. At the moment, there are 3 possible star sections (soon to be 6). They are:

*node data

    *node properties

    *tie data

    A description of each follows. At the end of this document is a complete VNA file.

Node Data Section

    The Node Data section contains variables that describe the actors in a network. Here is an example:

*node data

    id gender role betweenness

    HOLLY female participant 78.33333588

    BRAZEY female participant 0

    CAROL female participant 1.333333373

PAM female participant 32.5

    PAT female participant 39.5

    JENNIE female participant 6.333333492

    PAULINE female participant 12.5

    ANN female participant 0.5

    MICHAEL male participant 58.83333206

    BILL male participant 0

    LEE male participant 5

    DON male participant 16.33333397

    JOHN male participant 0

    HARRY male participant 2.333333254

    GERY male instructor 54.66666794

    STEVE male instructor 16.83333397

    BERT male instructor 13.66666698

    RUSS male instructor 47.33333206

The first line (“*node data”) identifies the section as containing node data.

The line following “*node data” is a list of variable names. The first variable is assumed

    to be a unique identifier. It can be numeric or text, as long as each node has a distinct

    value. If any value (for any variable) contains spaces or other extraneous punctuation, it

    should be enclosed in full quotes, as in:

“John Barrymore”

Following the line of variable names is the actual data corresponding to those variables.

    All following lines are assumed to be node data until a new star command is read or the

    end of the file is reached.

Node Properties Section

The node properties section is very similar to the node data section, except that the

    variables all refer to display characteristics of the nodes, such as size, color, and shape.

    Here is an example:

*Node properties

    ID x y color shape size

    "HOLLY" 1094 415 255 1 10

    "BRAZEY" 84 742 255 1 10

    "CAROL" 1224 996 255 1 10

    "PAM" 1249 722 255 1 10

    "PAT" 1291 551 255 1 10

    "JENNIE" 1518 686 255 1 10

    "PAULINE" 1051 928 255 1 10

    "ANN" 1330 876 255 1 10

    "MICHAEL" 791 365 255 1 10

    "BILL" 785 52 255 1 10

    "LEE" 80 619 255 1 10

    "DON" 994 195 255 1 10

    "JOHN" 776 894 255 1 10

    "HARRY" 945 214 255 1 10

    "GERY" 600 578 255 1 10

"STEVE" 338 636 255 1 10

    "BERT" 282 897 255 1 10

    "RUSS" 543 814 255 1 10

    As before, the first line (“*node properties”) identifies the section as containing node properties. The line following “*node propertiesis a list of variable names. Aside from

    the first variable, which must be ID, all the other variables can be in any order, and none of them have to be there at all. But if they are present, they must be named exactly as shown in the example. Variable “X” is the horizontal coordinate of a node. Variable “Y” is the vertical coordinate (the 0,0 point is the top left corner of the drawing area). Variable “Color” is the color of the node (in hexadecimal). Variable “Shape” is the shape of the node (circle, square, up-triangle, etc.). Variable “Size” is the size of the nodes in points. Finally, (not shown in the example), the variable “Shortlabel” gives the label for each node (if not given, the program uses the ID code).

    Following the line of variable names is the actual data corresponding to those variables.

Tie Data

The Tie Data section contains dyadic data the presence/absence or strength of tie

    among pairs of nodes on one or more relations. Here is an example:

*Tie data

    from to talk strength

    HOLLY PAM 1 1

    HOLLY PAT 1 3

    HOLLY DON 1 2

    BRAZEY LEE 1 1

    BRAZEY STEVE 1 2

    BRAZEY BERT 1 3

    CAROL PAM 1 1

    CAROL PAT 1 2

    CAROL PAULINE 1 3

    PAM JENNIE 1 3

    PAM PAULINE 1 1

    

    The second line contains the list of dyadic variables (relations), except that the first two variables are necessarily called “from” and “to” and identify the nodes that are tied. In this example, there are two relations (called “talk” and “strength”).

    Following the variable names are the actual ties. A data line such as “Holly Pam 1 1” indicates that Holly talks to Pam and their relationship has strength 1. Values of zero are assumed to indicate the absence of a tie on a given relation.

Putting it all together

Not all possible sections need to be in a given file just one will do. Here is an example

    of a file with all sections:

*Node data

    ID, gender, role, betweenness

     HOLLY female participant 78.33333588

     BRAZEY female participant 0

     CAROL female participant 1.333333373

     PAM female participant 32.5

     PAT female participant 39.5

     JENNIE female participant 6.333333492

     PAULINE female participant 12.5

     ANN female participant 0.5

     MICHAEL male participant 58.83333206

     BILL male participant 0

     LEE male participant 5

     DON male participant 16.33333397

     JOHN male participant 0

     HARRY male participant 2.333333254

     GERY male instructor 54.66666794

     STEVE male instructor 16.83333397

     BERT male instructor 13.66666698

     RUSS male instructor 47.33333206 *Node properties

    ID x y color shape size shortlabel HOLLY 1160 271 255 1 10 HOLLY BRAZEY 1214 577 255 1 10 BRAZEY CAROL 671 612 255 1 10 CAROL PAM 985 127 255 1 10 PAM

    PAT 802 402 255 1 10 PAT

    JENNIE 729 187 255 1 10 JENNIE PAULINE 69 590 255 1 10 PAULINE ANN 877 818 255 1 10 ANN

    MICHAEL 182 224 255 1 10 MICHAEL BILL 380 137 255 1 10 BILL

    LEE 617 44 255 1 10 LEE

    DON 281 656 255 1 10 DON

    JOHN 617 839 255 1 10 JOHN

    HARRY 382 410 255 1 10 HARRY GERY 1051 706 255 1 10 GERY STEVE 64 394 255 1 10 STEVE BERT 348 812 255 1 10 BERT

    RUSS 1176 426 255 1 10 RUSS *Tie data

    from to friends strength

    HOLLY PAM 1 1

    PAT HOLLY 1 2

    PAULINE PAT 1 2

    JOHN RUSS 1 3

    HARRY HOLLY 1 2

    HARRY MICHAEL 1 1

    BERT RUSS 1 3

    RUSS GERY 1 1

    RUSS STEVE 1 3

    RUSS BERT 1 2

    HOLLY BRAZEY 0 7 HOLLY CAROL 0 17 BRAZEY PAULINE 0 7 BRAZEY ANN 0 6

    BRAZEY MICHAEL 0 15 PAM MICHAEL 0 9 PAM BILL 0 16

    PAM LEE 0 13

    JENNIE BRAZEY 0 8 PAULINE JENNIE 0 5 PAULINE ANN 0 4 ANN PAT 0 7

    ANN MICHAEL 0 9 BILL LEE 0 10

    DON ANN 0 12

    DL Data Format

    The DL protocol is a flexible language for describing data and itself encompasses a number of different formats. Three of these formats nodelist, edgelist and fullmatrix are described here.

    A sample nodelist file called borg4cent.txt is provided with the program.

    Nodelist Format

    This is usually the most efficient format. Just create a text file using any word processor (make sure to remember to save as text). Enter the data in the following format:

    dl

    n = 50

    format = nodelist

    data:

    1 7 8 2

    3 19 21 49 6

    2 6

    

    The "DL" at the top is required and identifies the type of file. The "n=50" tells program to expect up to 50 distinct nodes. The "format = nodelist" tells the program to expect the node list format (as opposed to edge list and full matrix). The word "data:" (don't forget the colon) marks the end of information about the data and the beginning of the data itself.

    The first line of the data ("1 7 8 2") says that person 1 has ties to three people, who are 7, 8 and 2. The ordering of the people is arbitrary and makes no difference. The second line, "3 19 21 49 6" says that person 3 has ties to four people, who are 19, 21, 49 and 6.

    Important note: each value is separated by a space (or tab). Each value is a "sequential" ID number. By "sequential" I mean that the numbers run from 1 to n. You can't have arbitrary ID numbers like "1001" or non-numeric IDs like "BOS007" or "Steve" unless you add the words "Labels embedded" some time before the "data:" statement, as follows:

    dl

    n = 50

    labels embedded

    format = nodelist

    data:

    binlad geobus tonblai kenski

    bilste jeabar stebor judcla jandoe

    kenski jandoe

    

    These names or labels must be less than 20 characters long and should not contain spaces or punctuation (as in "osama bin laden") unless they are enclosed in full quotes. A sample data file called borg4cent.txt using nodelist format is provided with the program.

Report this document

For any questions or suggestions please email
cust-service@docsford.com