Attitudes toward new organizational technology: network proximity as a mechanism for social information processing

Administrative Science Quarterly,  June, 1991  by Ronald E. Rice,  Carolyn Aydin

• E-mail
• Print
• Link

Attitudes toward New Organizational Technology: Network Proximity As a Mechanism for Social Information Processing

This study explicates and tests three network-based mechanisms whereby individuals' attitudes toward an integrated health information system may be influenced by the attitudes of proximate sources of social information. Estimates of the attitudes of generalized others are not convincingly associated with the actual attitudes of specified others. Overall, specific others' actual attitudes have only a small effect on one's attitude. Social information operates positively through communication and work-unit mechanisms (especially when weighted by importance), and negatively through the mean attitude of one's structurally equivalent position, but not through overall organizational proximity or spatial proximity. Differences in attitudes among occupational groups may have been due to implementation policies.

INTRODUCTION

Research on the adoption and implementation of organizational information systems is fairly consistent on at least one point: user attitudes toward the innovation are crucial factors in the success of such systems (Lucas, 1981). Hence, understanding how these attitudes develop should be a central concern in innovation and implementation research. One factor that has received recent attention is the extent to which these attitudes are influenced by others' attitudes. Strong claims have been made for the importance of social information processing in understanding the use and evaluation of new organizational media (Fulk et al., 1987). However, prior empirical tests of the influence of social information processing on attitudes toward new technologies provide contradictory results. Svenning (1982) found slight evidence of the influence of perceived referent-group attitudes on intentions to use a proposed videoconferencing system. A study of a census of 46 employees in a government office found that the similarity of perceptions of electronic mail was related to organizational culture and norms (Nosek, 1989). However, Davis, Bagozzi, and Warshaw (1989) found that MBA students' attitudes toward a software package were not influenced by subjective peer norms, and Pease (1988) found no influence of perceived referent-group attitudes on individuals' actual use of the same video -conferencing system studied by Svenning. Considerably stronger results have been found in studies that used network measures to identify groups of similar users or to operationalize the influence process. Examples include studies of the use and effects of a health information system (Anderson and Jay, 1985; Williams, Rice, and Rogers, 1988: chap. 11), a voice messaging system (Rice and Shook, 1990a), and electronic mail systems (Rice et al., 1990; Schmitz and Fulk, 1991). The present research reviews several fundamental components of and several persistent problems with social influence models, making specific reference to social information processing theory. It then applies network analysis constructs to ask three fundamental questions: (1) Are respondents' estimates of others' attitudes a valid indicator of those others' actual attitudes, and thus social information? (2) Are one's attitudes influenced by the attitudes of those who are relationally, positionally, or spatially proximate? and (3) Are one's attitudes influenced by the attitudes of all other individuals according to their proximity or by the simple mean of some bounded subset of others?

Social Information Processing in Organizations

According to traditional studies of organizational information systems, individual perceptions of an information system are likely to be somewhat influenced by objective characteristics of the system, individual differences such as one's past experiences with similar systems, extent of use of the system, occupational demands and procedures, and level in the organization. For example, to the extent that one actually uses the system, one is more likely to understand and have experience with the system and therefore have stronger attitudes toward the system; for successful systems this implies more positive attitudes (Lucas, 1981). To the extent that different occupations in health organizations have different norms about the role of technology and play different parts in the implementation process, occupational membership should also influence one's attitudes (Kimberly and Evanisko, 1981; Aydin, 1989). One's level in the organization would influence one's attitudes in accordance with the information requirements and access to participation in the implementation at that level (Lucas, 1981; Rice and Shook, 1990b). The social information processing model, developed in reaction to the failure of individual attributes or objective task measures to sufficiently explain reactions to workplace phenomena, brought theories of social influence to the organizational setting (Salancik and Pfeffer, 1978). Proponents of this theory would argue, then, that individual perceptions are also likely to be influenced by the opinions, information, and behaviors of salient others. Social information processing, as with social influence in general, is especially likely in ambiguous or indeterminate situations or for newcomers to the setting (Festinger, 1954; Moscovici, 1976: 25, 32; Thomas and Griffin, 1983; Sheppard, Hartwick, and Warshaw, 1988). However, even when not uncertain about the phenomenon, an individual may still be influenced by sources on the basis of affiliation, resources, or authority, or by individuals who are mediators to sources of information about the phenomenon (Moscovici, 1976). Normative influence may also be more likely in cooperative situations, regardless of the level of uncertainty. One review concluded that, while methodological and conceptual problems cloud the issues, both objective task characteristics and social information cues do appear to influence individuals' perceptions of their tasks (Thomas and Griffin, 1983). Thus social information processing theory postulates that individuals may be influenced by cues from others about what to attend to, how to value the salient dimensions of workplace phenomena, and how others evaluate the same phenomena (Salancik and Pfeffer, 1978). Central to this proposition is, first, the assumption that individuals must be proximate to the attitudes, information, or behavior of others to be exposed to social information (Salancik and Pfeffer, 1978; Hackman, 1983: 1458-1459; Dean and Brass, 1985). Proximity, in this context, is defined as the extent to which one could be exposed to social information in a given social system. Second, the individual must value the source others in some way for the social information to have an effect. The "value" of the source other, however, may be embedded in or confounded with the strength of proximity. In terms of social information influencing attitudes toward a new information system, we hypothesize: