Software

We are currently compiling a list of social network analysis software programs. If you have suggestions for inclusion or would like to help review programs, please contact Maryann Durland, or any other member of our leadership team. We would like your suggestions for software, and any comments or ratings on software!

thank you!

In the mean time, check out this brief review of different tools and introduction to Gephi from Lada Admic, associate professor at the University of Michigan in the School of Information and the Center for the Study of Complex Systems and instructor for the on-line Coursera Social Network Analysis class. 

UCINET (http://www.analytictech.com/)
Advantages: UCINET allows for many network measures, and also provides options for conducting statistical tests. NetDraw, is a visual graphing option that is included. Relatively inexpensive. Disadvantages: Data input can be awkward, and there are many options for doing the same thing. If you are not sure what you want to do, it may be difficult to find the right routine. Visuals are of moderate quality. There is now a book out which takes you through some common routines. Borgatti, S. P., Everett, M. G., & Johnson, J. C. (2013). Analyzing social networks. SAGE Publications Limited. ISBN-13: 978-1446247419.  And a website to follow the examples in the book. 


NodeXL (http://nodexl.codeplex.com/)
NodeXL is a free (for the basic version), open-source template for Microsoft Excel that makes it easy to explore network graphs. NodeXL is particularly good for analysis of social media. Advantages: handles large data sets. Supports the most standard metrics and a range of layout options.  Includes a wide menu of filters and data partitioning options.  Provides aesthetically good visualizations. In particular, NodeXL offers a depiction of clusters and components that is not available in other packages, setting out each subgroup in its own space.  Imports and exports many data formats (professional version only) and allows relatively easy data editing. Easily imports social media data directly from the internet. Relatively easy to learn and well supported in book form. Disadvantages: Does not support statistical tests or models and provides a limited number of metrics. The  “professional” version is relatively expensive and must be renewed yearly.

Gephi (http://gephi.org/)

Gephi is open source, free of charge and is particularly good for sophisticated layouts and visualizations. Advantages: handles large data sets.  Supports the most standard metrics and a range of layout options.  Includes a wide menu of filters and data partitioning options. Many optional add-on packages are available supporting a wide range of analytical, visual and import/export options, including those for web applications. Imports and exports many data formats and allows relatively easy data editing.  Can generate aesthetically superior visuals, especially for large graphs. Reasonably well documented, particularly in book form. Free. Disadvantages: Does not support statistical tests or models and provides a limited number of metrics. Moderately difficult to learn.  Use of too many of the available visual options can lead to an “over-decorated” look, depending on your taste and audience. 

PAJEK (http://vlado.fmf.uni-lj.si/pub/networks/pajek/)
Package for large network analysis.

R (https://www.r-project.org) (https://statnet.org) (http://igraph.org)

Igraph: A package in R. Advantages: handles large data sets. Supports all standard metrics and many statistical tests and data/vector operations including some that may not be supported by UCINET.  Supports some visualization options not available elsewhere.  Imports and exports many data formats.  Well documented, particularly online. Free.  Disadvantages: more of a learning curve than others, including basic R knowledge and moderate coding. Only aesthetically moderate visualizations, although with many layout and decorative options.

Statnet: Another package in R, often used in conjunction with igraph. Advantages: supports a range of statistical tests and visualizations including Exponential Random Graph Models. Well documented, particularly online. Free. Disadvantages: like igraph, statnet involves more of a learning curve than others, including basic R knowledge and moderate coding. Only aesthetically moderate visualizations, although with many layout and decorative options.

Many analysts also use sna, which is another of the network analysis packages in R.


KRACKPLOT (http://www.contrib.andrew.cmu.edu/~krack/)

ANTHROPAC (http://www.analytictech.com/)
Helps collect and analyze structured qualitative and quantitative data including freelists, pilesorts, triads, paired comparisons, and ratings. ANTHROPAC's analytical tools include  techniques that are unique to Anthropology, such as consensus analysis, as well as standard multivariate tools such as multiple regression, factor analysis, cluster analysis, multidimensional  scaling and correspondence analysis. In addition, the program provides a wide variety of data  manipulation and transformation tools, plus a full-featured matrix algebra language. 

INFLOW (http://www.orgnet.com/inflow3.html)
Network visualization

Network Genie (https://secure.networkgenie.com/)
Online application for designing and managing social network projects, including the design of surveys and survey questions, the management of social network projects, the collection of social network survey data, and downloading and exporting data to social network analysis  programs

LinkaLyzer™, from MDLogix, Inc,
Accepts existing egocentric social network data in an SPSS, Microsoft® Excel, or  tab-delimited text file as input. Data can be in either Egocentric (one record per Primary subject) or  Egocentric-Dyadic (one record for each Secondary subject) format. LinkaLyzer lets you choose  the criteria used to determine whether different subjects in your data are really the same person. You can define combinations of numeric, string, and date variables to be used with a variety of comparison functions to find potentially matching subjects in your data. Once you have produced a set of potential matches, you can view and accept them one by one, or for large datasets you can have LinkaLyzer accept all of them automatically. When the matches are combined, the links of matched subjects are also combined, resulting in a new network.

FATCAT (http://www.sfu.ca/~richards/Pages/fatcat.htm)
Different kind of network analysis program. FATCAT works with categorical who-to-whom matrices, in which you select a variable that describes nodes to determine the categories for rows (who) and another one to determine the categories for columns (whom).

NEGOPY (http://www.sfu.ca/~richards/Pages/negopy.htm)
One of the oldest network analysis programs, NEGOPY finds cliques, liaisons, and isolates in networks having up to 1,000 members and 20,000 links. In use at over 100 universities and research centers around the world.

StOCNET (http://stat.gamma.rug.nl/stocnet/)
Open software system currently under development that will provide a new platform to make a number of statistical methods that are presently privately owned available to a wider audience. A new version that contains BLOCKS and SIENA can be downloaded.

GRADAP 2 (http://www.gamma.rug.nl/)
GRAph Definition and Analysis Package, can be used to define, manipulate, and analyze graphs and networks of various kinds.

PREPSTAR (http://kentucky.psych.uiuc.edu/pstar/index.html)

PSPAR (http://www.sfu.ca/~richards/Pages/pspar.html)
Sparse matrix version of PSTAR.

IKNOW (http://csu1.spcomm.uiuc.edu/projects/TECLAB/iknow/)
Tool that assists the study, creation, and growth of knowledge networks.

BLANCHE (http://csu1.spcomm.uiuc.edu/Projects/TECLAB/BLANCHE/)
Blanche is a computational modeling environment to specify, simulate, and analyze the evolution 
and co-evolution of networks.

VISONE (http://www.visone.de/)
Network visualization.

NETVIZ (http://www.netviz.org/
Network visualization