Versions

This page shows changes to UCINET between versions. Just scroll down to see release notes in reverse chronological order.
Note: To find revision information for really old versions of UCINET look in these two places:
Related Links
  • To see a list of known bugs and user wish list items, check the fix list
  • To download the latest version, click download

News

  • Two-day workshop on social network analysis using UCINET WHOTaught by Steve Borgatti under the auspices of Statistical HorizonsWHENFriday, April 27, 2018 9:00 AM –Saturday, April 28, 2018 5:00 PM (Eastern Time)WHERETemple ...
    Posted Jan 14, 2018, 12:35 PM by Steve Borgatti
Showing posts 1 - 1 of 12. View more »

Version 6.666 | 31 Oct 2018

posted Oct 31, 2018, 12:34 PM by Steve Borgatti   [ updated Oct 31, 2018, 12:38 PM ]

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)

Version 6.665 | 4 Sept 2018

posted Sep 4, 2018, 11:47 AM by Steve Borgatti   [ updated Sep 4, 2018, 12:34 PM ]

Fixes
  • 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

Version 6.664 | 26 Aug 2018

posted Aug 26, 2018, 7:06 AM by Steve Borgatti   [ updated Sep 4, 2018, 11:05 AM ]

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
Fixes
  • 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

Version 6.662 | 30 July 2018

posted Jul 30, 2018, 6:38 PM by Steve Borgatti   [ updated Jul 30, 2018, 6:40 PM ]

  • Added new procedure to export edgelists in Data|Export|DL. Currently, the procedure is limited to edgelists, but will be expanded to include other dl formats as well
 download 

Version 6.661 | 22 July 2018

posted Jul 22, 2018, 6:25 AM by Steve Borgatti   [ updated Jul 22, 2018, 12:57 PM ]

  • added triad census function to the CLI. Syntax is triadcensus(<network>). Usage example:
    ->dsp triadcensus(campnet)
  • Also added node-level triad census function to CLI. This counts how often each node is involved in each kind of triad. Syntax: nodetriadcensus(campnet). Usage example:

    ->ntc = nodetriad(campnet)
    ->summed = transp(total(ntc col))
    ->dsp summed; triadcensus(campnet)

    Note that the summed the summed triad counts will not equal the overall triad census. In the sum, most triads will be triple-counted as each member of a given triad will occur once. Similarly, triads 012 and 102 will each be double-counted. 
Update 15:54
  • Added CLI command to list undirected triads with specified number of ties. Syntax: liststriads(<dataset> [0 1 2 3]). Usage:
    ->dsp listtriads(campnet 3)  // lists all closed triads (ones with 3 ties)
    ->dsp listtriads(campnet 2 3) //lists all open triads (2 legs) and all closed triads (3 legs)
    ->dsp listtriads(campnet 0) //lists only triads containing no ties

Version 6.660 | 8 July 2018

posted Jul 8, 2018, 2:28 PM by Steve Borgatti   [ updated Jul 10, 2018, 2:58 PM ]

  • Added procedure to CLI for exporting networks to Pajek .net files. Syntax is Pajek <network> [<coordinates>]. Usage example:
    ->pajek campnet   // creates a text file called campnet.net
  • Install package includes netdraw 2.164, which fixes a bug in reading vna files in which you could not have a variable name that started with the letters 'id'
Update 10 July 2018
  • replaced netdraw with version 2.165 which changes the way the default increment is set. Now, if the data have values greater than 1, the increment is set to 1. If the values are 1 or less, the increment is set to 0.1.

Version 6.659| 23 June, 2018

posted Jun 23, 2018, 10:40 AM by Steve Borgatti   [ updated Jun 23, 2018, 7:51 PM ]

  • Updated help file
  • added options to Transform|Match Marginals

Version 6.658 | 3 June 2018 (updated 22 June)

posted Jun 3, 2018, 7:03 AM by Steve Borgatti   [ updated Jun 22, 2018, 3:43 PM ]

New Features
  • In the CLI, changed Degree command so that when given directed data, it produces both indegree, outdegree and degree (on the underlying graph, i.e., symmetrized by maximum).
  • Added temporary help button in CLI to online help
  • (22 June) updated the help files

Version 6.657 | 30 May 2018

posted May 30, 2018, 8:49 AM by Steve Borgatti

Bugs
  • Fixed bug in CLI's alters-in-common procedure (aic). It had actually been calculating alters not in common.
New features
  • In Network|Cohesion|Alters in Common, the user can now specify different generalizations of simple counts for the case where ties are valued
  • The CLI's version of AIC now has this as well. Syntax is aic(<network> [out|in|BOTH|recip]  [min|pos|PROD]). Usage examples:
    ->x = aic(samplk1 min)
    ->dsp aic(samplk2 out pos)
 download 

Version 6.656 | 27 May 2018

posted May 27, 2018, 3:42 PM by Steve Borgatti   [ updated May 27, 2018, 3:44 PM ]

  • Added procedure under Transform|Make Structural Effects for Dyadic Regression. This creates matrices that can be used to test a number of structural effects using QAP regression, such as 
    • Reciprocity
    • Transitivity -- tendency for a-->b and b-->c to result in a-->c
    • Cyclicity -- tendency for a-->b and b-->c to result in c --> a
    • Alters in common (out) -- how often a and b send ties to same others
    • Alters in common (in) -- how often a and b receive ties from same others
    • Out coverage -- proportion of a's outgoing ties that are also b's
    • In coverage -- proportion of a's incoming ties that are also b's
    • Reciprocal distance -- reciprocal of geodesic distance from a to b
    • In pref attach -- tendency to send ties to those w/ many incoming ties
 download 

1-10 of 245