کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
1129306 | 955245 | 2013 | 7 صفحه PDF | دانلود رایگان |

There have been two distinct approaches to two-mode data. The first approach is to project the data to one-mode and then analyze the projected network using standard single-mode techniques, also called the conversion method. The second approach has been to extend methods and concepts to the two-mode case and analyze the network directly with the two modes considered jointly. The direct approach in recent years has been the preferred method since it is assumed that the conversion method loses important structural information. Here we argue that this is not the case, provided both projections are used together in any analysis. We illustrate how this approach works using core/periphery, structural equivalence and centrality as examples.
► Full structural information of two mode network data is nearly always contained in the projections.
► Using both projections together usually means there is no loss of structural information.
► Projections have the potential to allow us to use methods which uncover additional structural information in two mode data.
Journal: Social Networks - Volume 35, Issue 2, May 2013, Pages 204–210