David Krackhardt

Redundant governance structures : An analysis of structural and relational embeddedness in the steel and semiconductor industries

Network researchers have argued that both relational embeddedness—characteristics of relationships—and structural embeddedness—characteristics of the relational structure—influence firm behavior and performance. Using strategic alliance networks in the semiconductor and steel industries, we build on past embeddedness research by examining the interaction of these factors. We argue that the roles relational and structural embeddedness play in firm performance can only be understood with reference to the other. Moreover, we argue that the influence of these factors on firm performance is contingent on industry context. More specifically, our empirical analysis suggests that strong ties in a highly interconnected strategic alliance network negatively impact firm performance. This network configuration is especially suboptimal for firms in the semiconductor industry. Furthermore, strong and weak ties are positively related to firm performance in the steel and semiconductor industries, respectively. Copyright

Informal Networks and Organizational Crises: An Experimental Simulation

This paper argues that organizations with a particular social network structure are more effective than most organizations in responding to crises. Further, it is argued that the effective structure does not occur naturally, but must be designed consciously and carefully. A theory is developed based on well-founded principles of social science, most notably work on formal structure, conflict, friendships, and organizational crises. The paper concludes with an experimental test of one of the four propositions deduced from the theory. Six trials of the experiment found significant support for this proposition.

PREDICTING WITH NETWORKS: NONPARAMETRIC MULTIPLE REGRESSION ANALYSIS OF DYADIC DATA *

Abstract This paper argues that the quadratic assignment procedure (QAP) is superior to OLS for testing hypothesis in both simple and multiple regression models based on dyadic data, such as found in network analysis. A model of autocorrelation is proposed that is consistent with the assumptions of dyadic data. Results of Monte Carlo simulations indicate that OLS analysis is statistically biased, with the degree of bias varying as a function of the amount of structural autocorrelation. On the other hand, the simulations demonstrate that QAP is relatively unbiased. The Sampson data are used to illustrate the QAP multiple regression procedure and a general method of testing whether the results are statistically biased.

Cognitive social structures

There are problems within the area of network analysis that can be fruitfully explored with cognitive social structures (CSS). Such structures can be modeled as three-dimensional (N × N × N) network structures. A definition of such structures is presented, along with a review of some of the problems CSS might address. Three types of aggregations of CSS - Slices, Locally Aggregated Structures (LAS), and Consensus Structures (CS) - are proposed to reduce CSS to a tractable two dimensions for analysis. As an illustration, the CSS of a management team of a small manufacturing firm is analyzed comparing all three types of aggregations.

On the robustness of centrality measures under conditions of imperfect data

An analysis is conducted on the robustness of measures of centrality in the face of random error in the network data. We use random networks of varying sizes and densities and subject them (separately) to four kinds of random error in varying amounts. The types of error are edge deletion, node deletion, edge addition, and node addition. The results show that the accuracy of centrality measures declines smoothly and predictably with the amount of error. This suggests that, for random networks and random error, we shall be able to construct confidence intervals around centrality scores. In addition, centrality measures were highly similar in their response to error. Dense networks were the most robust in the face of all kinds of error except edge deletion. For edge deletion, sparse networks were more accurately measured.

Sensitivity of MRQAP tests to collinearity and autocorrelation conditions

Abstract Multiple regression quadratic assignment procedures (MRQAP) tests are permutation tests for multiple linear regression model coefficients for data organized in square matrices of relatedness among n objects. Such a data structure is typical in social network studies, where variables indicate some type of relation between a given set of actors. We present a new permutation method (called “double semi-partialing”, or DSP) that complements the family of extant approaches to MRQAP tests. We assess the statistical bias (type I error rate) and statistical power of the set of five methods, including DSP, across a variety of conditions of network autocorrelation, of spuriousness (size of confounder effect), and of skewness in the data. These conditions are explored across three assumed data distributions: normal, gamma, and negative binomial. We find that the Freedman–Lane method and the DSP method are the most robust against a wide array of these conditions. We also find that all five methods perform better if the test statistic is pivotal. Finally, we find limitations of usefulness for MRQAP tests: All tests degrade under simultaneous conditions of extreme skewness and high spuriousness for gamma and negative binomial distributions.

When Friends Leave: A Structural Analysis of the Relationship between Turnover and Stayers' Attitudes

David Krackhardt and Lyman W. Porter It is argued in this paper that macro and micro perspectives can each benefit from the other. To demonstrate this, a current research issue in micro organizational behavior is analyzed with the help of theories in psychology, social psychology, and sociology. The specific question is: What effect does turnover in an organization have on the attitudes of those who remain in the organization? A longitudinal investigation of three fast-food restaurants explored this relationship against the background of the social network structures in each site. Among the findings was that the closer the employee was to those who left, the more satisfied and committed he or she became. The results underscore the importance of the structural context in studying micro phenomena, while at the same time they demonstrate the richness of micro theory in understanding why these phenomena occur.*

Cognitive inconsistencies and non-symmetric friendship

Abstract Non-reciprocated relationships, such as all workers knowing the president of the company but only a few of the workers being known by the president, and non-symmetric relationships, such as workers thinking that they know the president and thinking that the president does not know them, are endemic to most social situations. While such inconsistencies may be expected in relationships such as giving advice and lending money, they are rarely expected to occur in seemingly symmetric relationships such as friendship. Nevertheless, they do. We suggest that research in this area has been hampered by the confused language used for describing ‘symmetries’ and ‘non-symmetries’. We present a framework for thinking about these relations that clearly distinguishes cognitive inconsistencies and non-symmetric and non-reciprocated relations. Then, we employ this framework and constructural theory to suggest that owing to cognitive inconsistencies, any interaction-based relationship, including friendship, can potentially be non-symmetric. We examine a series of hypotheses concerning interaction and interaction-based behaviors that derive from this theory using friendship relations. We find that we are able to predict both who is friends with whom, non-symmetry in friendship, and non-reciprocities in the expectation for and recall of friendship.

Structure, culture and Simmelian ties in entrepreneurial firms

This article develops a cultural agreement approach to organizational culture that emphasizes how clusters of individuals reinforce potentially idiosyncratic understandings of many aspects of culture including the structure of network relations. Building on recent work concerning Simmelian tied dyads (defined as dyads embedded in three-person cliques), the research examines perceptions concerning advice and friendship relations in three entrepreneurial firms. The results support the idea that Simmelian tied dyads (relative to dyads in general) reach higher agreement concerning who is tied to whom, and who are embedded together in triads in organizations.

Network Conditions for Organizational Change

Understanding the overall network structure of organizations can help managers to support change. This article describes three different network theories of change, exploring the underlying assumptions and implications of each model. First, the E-I model predicts that cross-departmental friendship ties will help generate positive response to change in organizations by fostering trust and shared identity. The viscosity model predicts that introducing controversial (not clearly good or bad) change into the periphery of an organization and carefully regulating the interaction of innovators and nonadopters provides the best chance that it will diffuse successfully. Finally, the structural leverage theory presents a mathematical model that supports broad diffusion of clearly superior change, informing as many people as possible about the change.