PERSONNEL PSYCHOLOGY m2.55 MlNllMlZlNG TRADEOFFS WHEN REDESIGNING WORK EVIDENCE FROM A LONGITUDINAL Q UAS I-EXPER I M ENT FREDERICK F! MORGESON The Eli Broad Graduate School of Management Michigan State University MICHAEL A. CAMPION Krannert School of Management Purdue University Although cross-sectional job design research highlights a tradeoff be- tween motivational and mechanistic work design, the redesign litera- ture is more equivocal. We develop a work redesign process that sug- gests the tradeoffs can be minimized if both motivational and mecha- nistic approaches are explicitly considered when work is designed and the ultimate outcomes of the design effort (e.g., satisfaction, efficiency, or both) are taken into account when work is redesigned. In a longi- tudinal quasi-experiment, we examined how jobs can be differentially changed in terms of their motivational and mechanistic properties. Re- sults showed at least partial support for all expected relationships. This suggests that the tradeoffs previously considered inherent in job design may not always occur, particularly if conceptual and methodological consideration is given to their minimization. Although work design enjoys a long and diverse history in the applied realm, the bulk of research in industrial-organizational (1-0) psychol- ogy has not reflected this diversity, instead focusing almost exclusively on motivational issues (e.g., job characteristics theory; Hackman & OId- ham, 1980). This motivational approach is limited, however, because it only considers a narrow set of work design factors and ignores mecha- nistically oriented approaches that have focused on such things as work simplification and specialization. This omission is all the more glaring because much of modem work design practice is largely based on mecha- nistic design principles (Morgeson & Campion, in press; Wall &. Martin, 1987). Thanks to Murray Barrick and David Hofmann for their comments on an earlier ver- sion of the manuscript. Their input resulted in many meaningful improvements. Correspondence and requests for reprints should be addressed to Frederick F! Morge- son, Eli Broad Graduate School of Management, Michigan State University, N475 North Business Complex, East Lansing, MI 48824-1122; [email protected]. COPYRIGHT 0 2WZ PERSONNELI'SYCX3OWGY. INC 589590 PERSONNEL PSYCHOLOGY Because of the relatively parochial nature of work design research in 1-0 psychology, only recently have fundamental tradeoffs between dif- ferent approaches to job design been acknowledged (Campion, 1988; Edwards, Scully, & Brtek, 1999). This research has found that motiva- tional and mechanistic approaches are at odds. For example, if the mo- tivational properties of work are improved, satisfaction with the work increases, but the efficiency with which work is performed (an impor- tant outcome of mechanistic approaches) tends to decrease. Other re- search has found these tradeoffs, but finds that they do not always occur (Campion & McClelland, 1993) or only occur in certain circumstances (Edwards, Scully, & Brtek, 2000). Unfortunately, no research pas directly tested whether these trade- offs can actually be minimized 6r whether satisfaction and efficiency- oriented outcomes can be independently influenced. We address these gaps in the literature by suggesting that the tradeoffs can be minimized if jobs are redesigned in certain ways, both motivationdand mechanistic approaches are explicitly considered when work is designed, and the ulti- mate outcomes of the design effort (e.g., satisfaction, efficiency, or both) are taken into account when work is redesigned. We investigated this possibility by redesigning computer information systems jobs according to one of three design goals: (a) increasing both motivational and mech- anistic aspects of work; (b) increasing only motivational aspects of work; and (c) increasing only mechanistic aspects of work. We then examined changes over time in these groups across several dependent measures in a 2-year longitudinal, quasi-experimental study at a large pharmaceuti- cal company. This study contributes to the work design literature in four ways. First, it is the only redesign study that simultaneously examines both mo- tivational and mechanistic work design approaches. Previous redesign research has tended to focus on motivational or mechanistic redesign, but not both. Second, the vast majority of work design research has been cross-sectional in nature. The quasi-experimental design of the present study adds value because it shows how actual changes to jobs impact a variety of outcome measures. Third, the redesign research that has been conducted has generally used relatively short-term evaluations (e.g., 6 months). The current research investigates redesign changes over a 2-year period. Finally, the jobs in the current study are high- skill, knowledge-work jobs. As such, they are reflective of the shift away from manufacturing-based organizations where goods are produced us- ing physical labor to knowledge-based organizations where services are provided through mental effort. This is different from the bulk of the work redesign literature that has primarily focused on manufacturing and entry-level jobs.MORGESON AND CAMPION 591 Understanding Motivational and Mechanistic Design Tradeofi Perhaps the first attempt to systematically design jobs utilizing sci- entific principles occurred in the early part of the 20th century through efforts such as those by Taylor (1911) and Gilbreth (1911). These mech- anistically oriented approaches focused on principles such as specializa- tion and simplification as a means of easing staffing difficulties and low- ering training requirements. Partly as a reaction to the reductionistic na- ture of mechanistic job design, and partly as an acknowledgment of hu- man potential and higher order needs, organizational theorists described job characteristics that could enhance worker satisfaction and provide for intrinsic needs (e.g., Herzberg, Mausner, & Snyderman, 1959; Lik- ert, 1961; McGrkgor, 1960). This became manifested in a motivationally oriented approach that identified a set of core job characteristics that determined the motivational nature of jobs (e.g., Hackman & Oldham, 1975; Hulin & Blood, 1968). Interestingly, this motivational approach has tended to completely ignore mechanistic approaches and the conse- quences of focusing solely on motivational