You can download either the 32-bit or 64-bit version of UCINET. The 32-bit version is the standard version and runs on both 32-bit and 64-bit Windows. The 64-bit version is limited in that it does not have all of the functions of the 32-bit version. It often crashes. Therefore, it is best used in tandem with the 32-bit version.

32-bit Installation Package. This installs the latest version of the 32-bit version of UCINET along with several helper programs (such as NetDraw), and puts a copy of all the standard datasets in a folder called Ucinet Data under your Documents folder. It runs on 64-bit and 32-bit Windows. The installation program is new as of version 6.531 and unfortunately can't automatically uninstall versions of UCINET prior to 6.531, so you should do that yourself prior to installing current version.

64-bit installation package. This is NOT recommended. If you have a 64-bit version of Windows, you can try this 64-bit version of UCINET. It lets you analyze much larger datasets, if you have the RAM memory for it. However, it is also flaky. Older parts of UCINET are not compatible with 64-bit execution, so these parts will crash. 

Trial vs "real" version. Actually, there is no separate trial version. If you download the program, it will run on your computer for 60 days without having to enter a registration code. Buying the program gets you that registration code. So whether you buy or just try, you download the same program.

Virus/malware issues. There was a time when the download files were stored in a DropBox location. For some reason, this caused Norton AntiVirus to flag the program as dangerous. So we don't use DropBox for this anymore. But you might still run into issues with Norton or some other anti-virus program when installing UCINET. One thing to be aware of is that Windows 8 will issue a warning that the publisher is unknown and probably dangerous. This just means we haven't sought (read: paid for) Microsoft certification.

Installation Notes


Installation works best if you right-click on the installation file (typically called something like setup32UCI6534.exe) and choose 'Run as Administrator'. Similarly, it is also helpful to use 'Run as Administrator' when running UCINET itself for the first time, as this allows you to register the program for all users of your machine. 


The best way to run UCINET on a Mac is to use a Windows emulator such as Parallels (or, of course, Bootcamp). However, it is (often) possible to run UCINET on a Mac using Wine. For more information, see this FAQ

Version Info

  • Version 6.666 | 31 Oct 2018
    Added two capabilities to the CLI. 
    • The degree command now accepts an optional parameter 'norm' to select normalized values. e.g.:
      ->dsp degree(campnet norm)
    • The symmetrize command now accepts an optional additional parameter 'both'. This determines how it handles missing values. If you include it, and either Xij or Xji are missing, then both Yij and Yji will be set to missing. If you omit it, then if just one of Xij or Xji are missing, Yij and Yji will be set to the non-missing value. e.g.,
      ->ymat = symmet(campnet min both)
    Posted Oct 31, 2018, 12:38 PM by Steve Borgatti
  • Version 6.665 | 4 Sept 2018
    • Fixed problem with installer that was installing wrong midas.dll file to user's system. This in turn caused Data|Filter|Filter rows using SQL to fail
    • Fixed bug in Data|Affiliations (2-mode to 1-mode). The option  sum(xi*yi)/min(x+,y+) was giving incorrect answers.
    • Updated help files
    Posted Sep 4, 2018, 12:34 PM by Steve Borgatti
  • Version 6.664 | 26 Aug 2018
    New capabilities
    • Added EI command in the CLI. It calculates the E-I "homophily" index at the individual level. It also includes a normalization based on the number of nodes in each category, as described in Everett, MG and Borgatti, SB. 2012. Categorical attribute based centrality: E–I and G–F centrality. Social Networks 34(4): 562–569. [pdf]
    • Added option to Data|Affiliations and Tools|Similarities & Distances to compute the sum of squared differences
    • Fixed Data|Random|Build matrix w/ fixed column marginals (indegrees). Previously, it was treating the column sums as stochastic. Now they are fixed.
    • Fixed help files for a number of routines

    Posted Sep 4, 2018, 11:05 AM by Steve Borgatti
Showing posts 1 - 3 of 245. View more »