• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of procamiaLink to Publisher's site
AMIA Annu Symp Proc. 2005; 2005: 1052.
PMCID: PMC1560595

Applying Organizational Network Analysis Techniques to Study Information Use in a Public Health Agency

Jacqueline Merrill, RN, MPH, DNS(c),1 Suzanne Bakken, RN, DNS,1,2 Michael Caldwell, MD, MPH,3 Kathleen Carley, PhD,4 and Maxine Rockoff, PhD2,5

Abstract

We are applying organizational network analysis to explore information use in a public health department. The technique is grounded in social science theory, and employs calculations derived from graph theory to generate statistical and graphical models from relational data matrices. For this pilot we will develop network models to characterize how information use in routine work contributes to the agency’s performance goals. This is the first step in ongoing work that will extend network analysis with computational methods, to predict the potential effects that improving information use has on the agency’s performance.

Introduction

“Advances in social network theory, cognitive science, computer science and organization theory have led to new understanding of organizations as socio-technical systems that behave in complex ways.”1 Organizational network analysis is a quantitative descriptive technique for modeling organizational systems as interlocking networks people, knowledge, resources, tasks, and groups. The method is applied to understand performance dynamics in complex systems. It uses relational datasets organized into adjacency matrices where rows and columns represent individual actors or resources. Within each cell of a matrix numbers represent the presence or absence of a direct relation, or the frequency or strength of a relation.2

Methods

Relational data will be collected from all workers in a local public health agency (n ~ 157) using standard social network questions to capture demographics, patterns of individual information use, expertise and skills, and knowledge of colleagues’ skills. Analysis will be performed using Organizational Risk Analyzer (ORA), which generates formatted reports in computer screen displays or log files, and includes tools for graphically visualizing data and performing regressions to compare network characteristics. In ORA network measures use equations derived from graph theory, where a network N is comprised of two sets of nodes, U and V, and a set of edges E [subset or is implied by] U×V. An element e = (i,j) in E indicates a tie or a relationship between nodes where i [set membership] U and j [set membership] V.

A measure is a function that maps one or more networks to Rn. Measures are scalar (n = 1), or vector valued where n = |V| or n = |U|.3 Node sets are defined as agent to agent, agent to knowledge, agent to resource, agent to task, knowledge to knowledge, knowledge to resource, and so on.

Results

The analysis will yield both diagrams of network relations and descriptive measures that convey the implications of the relational data. For example, cohesion measures are counts and ratios of clustering, density and centralization that show how information is distributed. Structural equivalence measures show correlations between node pairs with similar roles that reveal unidentified information. Prominence measures show influence such as who is in demand in the network and where there are new opportunities for communication, as well as unrealized or blocked communication.

Discussion

Our study will describe how patterns of information use influence agency performance. Public health makes intense use of specialized information, yet public health information needs are not well met.4 Our work will help PH managers understand where improved information is needed and justify the commitment of public resources required to support improvements to information use in PH work. This pilot project lays the groundwork for extending our analysis with computational techniques that can generate predictive models of the potential impact changes in how information is used could have on performance.

Acknowledgements

Jacqueline Merrill is a doctoral candidate supported by P20 NR 007799. This research is supported by NLM N01-LM-1-3521.

References

1. Carley KM. Computational organizational science: A new frontier. In: Proc National Academy of Science; 2002. p.7259.
2. Wasserman, S. Social Network Analysis. Cambridge University.Press; 1994
3. Carley KM, Reminga J. ORA: Organization Risk Analyzer. Pittsburgh, PA: Carnegie Mellon University, CASOS; 2004.
4. Rambo N, Zenan JS, Alpi KM, Burroughs CM, Cahn MA, Rankin J. Public Health Outreach Forum: lessons learned. Bulletin of the Medical Library Association. 2001;89(4):403–6. [PMC free article] [PubMed]

Articles from AMIA Annual Symposium Proceedings are provided here courtesy of American Medical Informatics Association
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...