Report R3.3 Networks: An Emerging Paradigm in Management Research Seminar Leader: Prasad Balkundi, SUNY Buffalo, USA bprasad95[at]yahoo.com PART 1 Social network analysis (SNA) is a relatively new approach in management. Any human network with two or more nodes could be called a social network. The pattern of interaction in the network has implications for the nodes and for the network as a whole. The seminar leader affirmed Martin Kilduff’s view that one paradigm can support another and hoped that the social network paradigm could supplement the other paradigms in organisational and management research. A new paradigm brings with itself a new language and a new set of researchers who tend to specialise in some field. The seminar leader sought to explore the following questions during the course of discussions. What exactly is social network analysis? How is it different from the other approaches in the field of organisational behaviour? Is it a metaphor or theory? How is it different from traditional research? An example of SNA was presented. It was seen that the pattern of interaction led to the formation of power structures. Giving examples of “stars” and “brokers” in the network, the seminar leader emphasised the need to align the formal structure with the informal network in an organisation, to increase its performance. It was also seen that the size of the network could affect its structure. SNA has applications in diverse fields such as epidemiology (HIV control), rural sociology (diffusion of hybrid corn), and strategy (terrorist networks). Through SNA, one aims to predict the behaviour of the nodes and the network. In organisations, there are informal links not captured by the formal organisational structure. The more the informal links and the formal structures align, the higher is the probability of better organisational performance. SNA analysts assume that social interaction drives individual behaviour while it is the other way round for researchers who follow the psychological approach. Using standard statistical tools, such as regression, poses a difficulty for SNA. The difficulty arises because of interdependence of variables. Strictly dependent and independent variables are difficult to specify in a social network, because everything seems to depend on everything else. Some specialised methods such as quadratic assimilation procedure have been developed to deal with this problem. But, what is more common in SNA is the use of softwares based on "graph theory" from mathematics, such as the UCINET software. Models drawn from physics and mathematics are increasingly influencing social network analysis. Surveys are commonly used to collect network data, but network surveys cannot be anonymous. Therefore, there is a constant obligation not to harm the respondents. Network surveys require a high rate of response, say around 80%. The roots of SNA could be traced back to the work of J. L. Moreno (1889-1974) [who founded The American Society of Group Psychotherapy and Psychodrama (ASGPP), 1942, see http://www.asgpp.org/]. Moreno's work at a young women's penitentiary was discussed. There were cases of repeated violence within a penitentiary. Moreno studied the friendship network among the inmates, by focusing on who they had lunch with, etc. Based on his observations, Moreno suggested a restructuring of the dormitories and reorganising the environment in order to avoid frequent interaction among members who hated each other. It considerably reduced the cases of violence in the penitentiary. PART 2 The seminar leader shared his experiences as a doctoral student. He gave valuable insights on the doctoral programme at The Pennsylvania State University, USA (http://www.psu.edu/). He talked about choosing a topic for research, issues regarding the choice of sample, importance of publishing papers, and the structure of the doctoral programme as a whole. His dissertation asked the question: Why are some leaders more influential than others? He used SNA as a method for his research. He concluded that networks drive charismatic actions, which transform into performance and that leadership arises out of expectations and perception of subordinates. The seminar leader advised doctoral students to work on a topic different from the principal interests of their advisors/ guides to create an identity of their own. Care should be taken while choosing the topic for dissertation, for research has to be interesting and relevant for it to make an impact. It could be noted that individual choices should be well considered because they have a bearing on the rest of one’s career. Publication in “quality journals” is an indicator of the quality of the doctoral work. This emphasises the need for good academic writing where each paragraph needs to be cogent and each sentence must reflect the progressive development of thought. Researchers should aim to make their doctoral studies a rewarding experience. Reported by Adwaita Govind Menon, with inputs from D. P. Dash, Jacob D. Vakkayil, and C. D. Kuruvilla (31 July 2005). Copyleft Xavier Institute of Management, Xavier Square, Bhubaneswar 751013, India Research World (ISSN 0974-2379) http://www1.ximb.ac.in/RW.nsf/pages/Home |
