Homepage  |  Publications  |   Software  |   Indicators  |   Animation  |   Geo  |   Blog  |   Search website (Google)

Annex to:

The measurement of “interdisciplinarity:”

A note on operationalizations and routines

 

Lin Zhang[2] and Loet Leydesdorff*[1]

 

Abstract

A variety of measures for interdisciplinarity has been proposed in the literature. In this brief communication, we introduce two measurement instruments which provide these indicators (e.g., Rao-Stirling diversity) for the column vectors of any data matrix. Additionally, inequality (e.g., Gini) and diversity (e.g., Simpson) values are provided. We envisage to make this freeware accessible to a wider audioence. The routines are introduced step-by-step. The various indicators are contextualized in the accompanying paper (in preparation; see also (Leydesdorff, Wagner, & Bornmann, 2018; Leydesdorff, Wagner, & Bornmann, 2019).

 

Introduction to the software

 

The program interd_vb.exe can be downloaded from

http://www.leydesdorff.net/software/interdisc.2020/interd_vb.exe ; it computes diversity measures along the column vectors of a data matrix saved in the .csv format (comma-separated variables; for example, in Excel). The program is a variant of mode2div.exe which reads network data in the Pajek format. This routine is much faster (in the case of large sets) because the heavy computation is done by a routine interdis.exe called by this program and written in Visual Basic. However, move2div.exe is not limited in the number of units to be compared, whereas this routine is (for various reasons) limited to 1023 variables in a single run.

 

Both the two programs – interd_vb.exe and interdis.exe -- and the data file (csv)  have to be in the same folder “C:\temp\”. When running the program, one is first prompted for the name of the data file. Provide the name without extension, but the file has to be named with the extension .txt. 

 

For example, the data can be in a comma-separated-variables file “toy.txt” like this:

 

0,0,3,0

0,6,0,4

9,0,0,3

4,4,0,5

0,3,4,0

 

 

One is advised to start the program from the c-prompt by typing at C:\temp\ as follows: “interd_vb” so that one obtains: “C:\temp\interd_vb”. Hit “Enter”. When prompted for a file name, one should type “toy”—or whatever the name of the file is—and then the number of column variables (“4” in this case).  Vector-based indictors such as the Gini index are computed over the column vectors. Note that disparity uses also the row vectors (for computing the distances). One can transpose the matrix if one is interested in the interdisciplinarity over the rows. Distances are computed using (1 – cosine).

 

First, interd_vb.exe, interdis.exe, and the data file (*.txt) have to be  made local in the folder “C:\temp\”. This location is assumed and therefore obligatory. (The advantage of running the routines from the C-prompt and not from the desktop is that error messages are visible if something goes wrong.) Interd_vb.exe calls interdisc.exe without user control. However, the user has to confirm “run” when prompted; but only one single time. The results are written into the files interdis.dbf and interdis.csv.

 

If a file “labels.dbf” is available in the same folder (C:\temp\labels.dbf), the names in the field “label” will be used for each column in the output of the routine. Otherwise, these labels will be named “v1,” “v2,” etc.

 

In the above case of the file “toy.txt”, the results are:

 

 

 

The advantage of this routine above mode2div.exe (at http://www.leydesdorff.net/software/mode2div ) is the speed. Thus, one can proceed larger sets within a short time. However, mode2div.exe reads Pajek .net-files.

 

Unlike syn3_vb.exe (at http://www.leydesdorff.net/software/synergy.triads ), this routine (interd_vb.exe) does not provide synergy values in order to save time.

 

Leydesdorff, L., Wagner, C. S., & Bornmann, L. (2019). Interdisciplinarity as Diversity in Citation Patterns among Journals: Rao-Stirling Diversity, Relative Variety, and the Gini coefficient. Journal of Informetrics, 13(1), 255-264.