On the off chance that someone is interested, I've posted a *.pdf copy of my dissertation online.  I completed it about 18 months ago.  If anyone is interested in discussing it, making suggestions on the research/analysis, or doing more with the data, please let me know.  Given my advanced years :) , I undertook what I (and certainly my committee) thought was a risky dissertation.  Little to their surprise, I did not get robust positive support for all of my hypotheses.  I think that an important part of the problem was that I did not get enough variance in the self-reports of task interdependence.  My field observation was that people felt a need to be more interdependent than their tasks actually called for.  In addition, I may not have used the most appropriate analytical techniques, given that I was the only person at North Texas who had any experience with social network analysis.
In any event, the dissertation can be found at http://engleassociates.com/social_capital.htm on the bottom of the webpage.
The abstract follows:

Engle, Scott L.,  Structural holes and Simmelian ties:  Exploring social capital, task interdependence, and individual effectiveness.  Doctor of Philosophy (Organization Theory and Policy), December 1999, 175 pp., 15 tables, 13 figures, references, 167 titles.

Two contrasting notions have been put forward on how social capital may influence individual effectiveness in organizations.  Burt (1992) sets forth the informational and control advantages that are possible by building an open network characterized by large numbers of structural holes.  In contrast, Krackhardt (1996), Coleman (1990), and Simmel (1950) have suggested that network closure, exemplified by large numbers of Simmelian ties, enables actors to develop trust, cohesiveness, and norms which contribute to effectiveness.  Simmelian ties are strong, reciprocal ties shared by three actors.

It is proposed that an actor’s network cannot be dominated by both structural holes and Simmelian ties.  Thus, this study examines whether a moderating variable is at work.  It is proposed that the actor’s task interdependence in the workplace influences the relationship between network closure and individual effectiveness.  Actors in less task interdependent environments will benefit especially from the information and control benefits afforded by a network characterized by structural holes.  Conversely, actors in highly interdependent environments will benefit especially from the creation of trust and cooperation that result from large numbers of Simmelian ties.

Data was collected on 113 subjects in three organizations.  Subjects were asked to rate the strength of their relationship with all organization members and their own level of task interdependence.  Contrary to expectations, nearly all subjects reported high levels of task interdependence.  Raters in each organization provided individual effectiveness measures for all subjects.  Hypotheses were tested using hierarchical set regression and bivariate correlation.  The results indicated support for the hypothesized relationship of Simmelian ties with task interdependence.  When examining all cases, no support was found for the hypothesized relationship of structural holes and Simmelian ties with individual effectiveness and of structural holes with task interdependence.  Nonetheless, additional analyses provided some indication of an association between Simmelian ties and individual effectiveness.  Task interdependence did not moderate the relationships between either Simmelian ties or structural holes and individual effectiveness.


Scott L. Engle, Ph.D.
Engle Associates
5201 Southern Hills Dr.
Frisco, TX  75034
ph: 972-370-9161
fax: 360-242-2794
e-mail: [log in to unmask]