Despite strong empirical data demonstrating the association between network variables and organizational outcomes, there is a lack of causal explanation linking an individual's network position to individual-level outcomes. This project examines the relationships between individual actors and network performance to develop valid models of measuring the formation of networks and their causal impact in facilitating or inhibiting success in organizational environments. The project moves beyond traditional organizational network research involving a single organization to investigate the social structure among more than 500 teachers and administrators from 25 schools in a mid-size public school district. The primary objective is to better understand the antecedents and the consequences of network structure. The network antecedent models will be designed to investigate the processes by which networks cohere. Personal construct psychology and the role repertory grid will be utilized to establish a set of personal constructs presumed to be associated with dyadic choice. The network consequence models will be developed to understand the mechanisms by which network variation influences individual and collective performance. Multilevel modeling and social network analysis will be combined to capture the interaction effects between nodal and whole network measures of teachers nested in schools.
It is anticipated that the network antecedent and network consequence models together will clarify the complex interdependencies between personal constructs, individual attributes, network position, and performance. The project also will improve understanding of how teachers interact to explore their craft and exchange knowledge. Workshops will be held with all interested schools in the district to discuss the research findings and policy implications, and the network models and methods derived from this research as well as the corresponding R code will be made available online. As a Doctoral Dissertation Research Improvement award, support is provided to enable a promising student to establish a strong, independent research career.
This study was designed to better understand how networks emerge in complex organizations (network antecedents) and to identify the effect of network structure on individual beliefs and performance (network consequences). Past research has focused on the structural determinants of social networks, such as the propensity for individuals to form reciprocal relations, as well as the effects of individual attributes like age and gender. This study sought to add a third component - interpersonal perception. Scholars have argued that interpersonal perceptions are the cognitive drivers of social behavior, yet such factors have not been systematically identified and used in developing models of how networks cohere in organizations. Regarding the consequences of networks, measuring the extent to which individuals improve their performance within the context of an organization has been historically difficult to assess. In instances where individual outcomes are emphasized, these outcomes are not direct measures of performance and tend to be highly subjective. To overcome the aforementioned limitations, this study incorporated a novel methodological approach to capture interpersonal perceptions and calculated an objective measure of individual performance. Data was collected from 521 teachers and principals in 21 schools and consisted of information on: (i) teacher social networks - the informal advice seeking connections and friendship ties formed within a school, (ii) survey variables ranging from organizational commitment to instructional leadership, and (iii) interpersonal perceptions. Findings from the investigation into the antecedents and consequences of networks include the following: Teachers rely on a consistent set of personality traits to compare their coworkers and to determine with whom they forge a social connection. Utilizing the role repertory grid technique from cognitive psychology, eleven personality traits were consistently identified by public school teachers as the key constructs used to cognitively differentiate their peers. Of the eleven traits, the most predictive of both friendship and advice tie formation was a teacher's perception of his or her peer's trustworthiness. Self-efficacy is shaped by the social structure of one's work environment. Statistical analysis revealed that a teacher's self-efficacy was increased when he or she was connected to peers with high self-efficacy. Social networks are predictive of teacher performance. This study calculated an objective measure of value-added performance based on four years of student test score data. Of more than a dozen potential predictors, two variables were significant predictors of a teacher’s value-added score: reflective dialogue and peer organizational commitment. Reflective dialogue captures the amount of time a teacher engages in thoughtful discourse with coworkers regarding student learning. Peer organizational commitment measures the organizational commitment of the individuals the teacher is connected with through his or her social network. Individuals with high organizational commitment are more likely to share tacit knowledge and extend additional efforts to help others and thus may be more valuable members of one's social network. The size of one's social network is less important than the quality of ties that comprise the network. Regarding both one's perception of their capacity to perform successfully and one's actual objective performance, the attributes of the individuals a teacher was connected to in the social network were found to be more important than the number of connections. Competition reduces collaboration. Teachers who perceived their work environment as being more competitive were found to be less likely to engage in advice and knowledge sharing with their peers. These findings have applications and implications for the schools participating in the study, education policy, and the broader study of organizations. All of the results were supplied back to the participating district and a critical discussion of the findings ensued. The district plans to use the data to make assessments as to why certain schools formed more cohesive networks and to develop district-wide strategies to facilitate a more collaborative culture among the teachers. The results strongly suggest the continued need for individuals to engage with their peers through in-depth discussions about their work and craft. Strong advice networks are associated with not only higher perceptions about one's ability to be successful, but also one's actual performance. These finding support district and school pursuit of collaborative policies like mentoring programs and professional learning communities. Furthermore, because competition may have a negative effect on collaboration, states interested in merit pay programs and other competitive policies should be aware that such programs may have unintended consequences on the schools social structure. The adaptability of the repertory grid for different organizational contexts makes it an ideal method for capturing the role of interpersonal perception in network formation. The use of the repertory grid along with statistical network models allows the analyst to measure and test structural, demographic, and cognitive factors related to network formation. Such sophisticated models can be used identify both individual and organizational factors critical for collaboration in complex organizational settings.
