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

  • Week-long workshop on SNA The LINKS Center at the University of Kentucky is offering its annual 1-week summer workshop on social network analysis June 6-10, 2016 on the University of Kentucky campus ...
    Posted Mar 15, 2016, 12:54 PM by Steve Borgatti
Showing posts 1 - 1 of 9. View more »

Version 6.616 | 30 May 2016

posted May 30, 2016, 3:14 PM by Steve Borgatti

  • Fixed bug in the CLI's Forfiles command. Previously, commands of the form ->forfiles samplk? draw were failing, while commands like ->forfiles samplk* $-cent = centrality($) were succeeding.

Version 6.615 | 29 May 2016

posted May 30, 2016, 7:07 AM by Steve Borgatti   [ updated May 30, 2016, 3:09 PM ]

  • Added "number of holes" statistic to CLI's Holes routine. It simply counts the number of pair of alters that are not connected. This captures both network size and density
  • Added limited Frequencies command to the CLI. Usage: 
->dsp freq(sampson)

Matrix: Frequencies

          1   2   3   4   5   6   7   8   9  10 
        SAM SAM SAM SAM SAM SAM SAM SAM SAM SAM 
        PLK PLK PLK PDL PES PDE PIN PNI PPR PNP 
          1   2   3   K       S       N       R 
        --- --- --- --- --- --- --- --- --- --- 
    1 0 251 249 250 259 252 248 253 256 267 265 
    2 3  18  18  18  15  17  16  17  16  13  13 
    3 1  18  21  19  15  18  20  17  16  14  14 
    4 2  19  18  19  17  19  22  19  18  12  14 

Matrix: Proportions

            1     2     3     4     5     6     7     8     9    10 
        SAMPL SAMPL SAMPL SAMPD SAMPE SAMPD SAMPI SAMPN SAMPP SAMPN 
           K1    K2    K3    LK     S    ES     N    IN     R    PR 
        ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 
    1 0 0.820 0.814 0.817 0.846 0.824 0.810 0.827 0.837 0.873 0.866 
    2 3 0.059 0.059 0.059 0.049 0.056 0.052 0.056 0.052 0.042 0.042 
    3 1 0.059 0.069 0.062 0.049 0.059 0.065 0.056 0.052 0.046 0.046 
    4 2 0.062 0.059 0.062 0.056 0.062 0.072 0.062 0.059 0.039 0.046 


 download 

Version 6.614 | 22 May 2016

posted May 22, 2016, 11:02 AM by Steve Borgatti   [ updated May 22, 2016, 1:51 PM ]

  • Changed Network|Compare aggregate proximity matrices|partition to be able to handle missing values
  • Changed the CLI's IPF routine to default to treating diagonal values as valid. If you want the diagonal ignored you must add the word IGNORE to the command. E.g.,
    ->dsp ipf(padgm ignore)
 download 

Version 6.613 | 18 May 2016

posted May 18, 2016, 7:35 PM by Steve Borgatti   [ updated May 18, 2016, 7:38 PM ]

  • Added ability to sort columns in matrix editor via double-clicking column header
  • Added new version of node level regression under tools|testing hypotheses|node level|regression
  • Fixed bug in CLI version of regression which was multiplying classical p-values by df/(df-1). 
  • Fixed recently added bug in tools|testing hypotheses|dyadic|relational cross tabs which was printing incorrect correlation values
 download 

Version 6.611 | 8 May 2016

posted May 8, 2016, 1:21 PM by Steve Borgatti   [ updated May 8, 2016, 1:22 PM ]

  • Adds countcombinations routine to the CLI. Given a matrix of 1s and 0s, it counts up the number of combinations of column items found across all rows of the matrix. For example, if applied to the 2-mode Davis data, it would tell you how often each different combination of events occurs in the data. Syntax is CountComb(<dataset>). Example:

    ->dsp countcomb(davis)
Frequencies
                                         1     2 
                                      Freq  Prop 
                                     ----- ----- 
    1     E2, E3, E4, E5, E6, E8, E9     1 0.056 
    2         E1, E3, E5, E6, E7, E8     2 0.111 
    3     E2, E4, E5, E6, E7, E8, E9     1 0.056 
    4                     E3, E4, E7     1 0.056 
    5                     E3, E5, E8     1 0.056 
    6                     E5, E6, E8     1 0.056 
    7                         E6, E9     1 0.056 
    8                     E5, E7, E8     1 0.056 
    9                E7, E8, E9, E12     1 0.056 
   10               E8, E9, E10, E12     1 0.056 
   11          E8, E9, E10, E13, E14     1 0.056 
   12      E7, E8, E9, E10, E12, E14     1 0.056 
   13 E6, E7, E9, E10, E11, E12, E13     1 0.056 
   14          E7, E8, E10, E11, E12     1 0.056 
   15                         E8, E9     1 0.056 
   16                        E9, E11     2 0.111 

In these data, only two combinations of events occur more than once. 


Version 6.610 | 25 April 2016

posted Apr 25, 2016, 7:34 PM by Steve Borgatti   [ updated May 7, 2016, 1:39 PM ]

  • Fixed bug in matrix editor that was causing it to save matrices with more rows and columns than were actually present
  • (Also, a bug was fixed a few versions ago that was causing "argument out of range" errors, but the fix was not previously reported)
Update 7 May 
  • Updated the installer to include latest version of NetDraw (2.159), which fixes a bug in reading 2-mode data

 download 

Version 6.609 | 18 Apr 2016

posted Apr 18, 2016, 2:14 PM by Steve Borgatti

  • Made change in Tools|Cluster|Hierarchical Clustering. Now, when the user chooses "Newman community detection", the program symmetrizes the data via the maximum method, and also dichotomizes. 

Version 6.608 | 16 Apr 2016

posted Apr 16, 2016, 7:35 AM by Steve Borgatti   [ updated Apr 16, 2016, 8:04 AM ]

  • Fixed bug in Network|Cohesion|Density|Density by Groups, which was failing to match the attribute dataset with the network dataset when the nodes in the two datasets were in different order. 
 download 

Version 6.607 | 11 Mar 2016

posted Apr 11, 2016, 3:49 PM by Steve Borgatti   [ updated Apr 11, 2016, 4:09 PM ]

  • Fixed bug in CLI's ForFiles command to allow handling files that are not ucinet system files. For example, suppose you want to convert several excel files to ucinet. If the excel files are all named with the same prefix (such as 'team1.xlsx', 'team2.xlsx', etc, you can type:

    ->forfiles team*.xlsx $$ = loadexcel($$)
    ->dsp centrality(team1)

    Previously, UCINET assumed that the files represented by $$ were ucinet files. 
 download 

Version 6.606 | 4 Apr 2016

posted Apr 4, 2016, 4:21 PM by Steve Borgatti   [ updated Apr 4, 2016, 5:28 PM ]

  • Added option to the Regress routine the CLI. Now, you can add YPERM as the last parameter; the program will generate significance using a Y permutation test on the t-statistics. Syntax is ->regress <yvar> <xvar1> <xvar2> ... [yperm]. For example,

    ->regress betweenness(campnet) cols(campattr "gender" "role"), yperm

    Regression statistics for 1

                           Value 
                       --------- 
                  Nobs        18 
              R-Square   0.08894 
          Adj R-square  -0.03254 
                F(0,0)   0.73213 
              Prob > F   0.49730 


    Regression coefficients

                         Coef        SE         T       Sig 
                    --------- --------- --------- --------- 
          Intercept   9.62500  21.48261   0.44804   0.70440 
             Gender  -7.62500  13.39676  -0.56917   0.29040 
               Role  19.37500  16.01219   1.21002   0.13500 


    Note, the results are not very different from the classical significance test (recalling that permutation tests are conducted 1-tailed, so the classical p-values should be halved for comparison):

    ->regress betweenness(campnet) cols(campattr "gender" "role")

    Regression statistics for 1

                           Value 
                       --------- 
                  Nobs        18 
              R-Square   0.08894 
          Adj R-square  -0.03254 
                F(0,0)   0.73213 
              Prob > F   0.49730 


    Regression coefficients

                         Coef        SE         T       Sig 
                    --------- --------- --------- --------- 
          Intercept   9.62500  21.48261   0.44804   0.65978 
             Gender  -7.62500  13.39676  -0.56917   0.57669 
               Role  19.37500  16.01219   1.21002   0.24283 




1-10 of 199