Title: | THE IMPACT OF ROLE VARIABLES ON INFORMATION SYSTEM PERSONNEL OUTCOMES: AN EMPIRICAL INVESTIGATION |
Authors: | Baroudi, Jack J. |
Issue Date: | Sep-1984 |
Publisher: | Stern School of Business, New York University |
Series/Report no.: | IS-84-71 |
Abstract: | This study examines the antecedents of job satisfaction, commitment and turnover intentions for 229 information systems development personnel (ISDP) employed by nine companies within several industries. The antecedents studied include boundary spanning, role ambiguity and role conflict. A model of these variables is built and tested via path analysis. A secondary analysis is performed to explore the impacts of task differences on the study variables. The task differences include analytic and programming tasks. The analyses revealed the following. Systems analysts span more boundaries than programmers. The major hazard faced by systems analysts when they span boundaries is role conflict which negatively impacts their job satisfaction, commitment and intention to quit. The overall effect of boundary spanning is to reduce ISDP intention to quit despite its positive relationship with role conflict. Systems analysts and programmers are both intolerant of role ambiguity. Role ambiguity is very detrimental greatly reducing ISDP job satisfaction, commitment and increasing ISDP intention to quit. Role ambiguity is not related to boundary spanning nor is it related to analytic or programming duties. Programmers are less committed than systems analysts and are more likely to express intentions to leave. The above information is used to make recommendations to IS management. Finally, recommendations and directions are suggested regarding future research. |
URI: | http://hdl.handle.net/2451/14570 |
Appears in Collections: | IOMS: Information Systems Working Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
IS-84-71.pdf | 4.93 MB | Adobe PDF | View/Open |
Items in FDA are protected by copyright, with all rights reserved, unless otherwise indicated.