Designing Promotion Ladders to Mitigate Turnover of IT Professionals

• 成果类型: Article
• 署名作者: MacCrory, Frank; Choudhary, Vidyanand; Pinsonneault, Alain
• 署名单位: Fordham University; University of California System; University of California Irvine; McGill University
• 刊物名称: INFORMATION SYSTEMS RESEARCH
• ISSN/ISSBN: 1047-7047
• DOI: 10.1287/isre.2016.0652
• 发表日期: 2016
• 页码: 648-660
• 关键词: information-technology professionals; VOLUNTARY TURNOVER; personnel turnover; systems personnel; EMPLOYEE TURNOVER; UNFOLDING MODEL; intentions; determinants; metaanalysis; tournaments

摘要

Chronic excessive turnover among information technology (IT) professionals has been costly to firms for decades with annual turnover rates as high as 24% even among Computerworld's 100 Best Places to Work in IT. Prior information systems literature has identified two key factors affecting turnover: boundary-spanning roles and low promotability in one's current firm. We draw on tournament theory, which is primarily concerned with inducing effort in employees, to decompose promotability into two distinct constructs: the likelihood of promotion and benefit from promotion, and demonstrate that each has a distinct role in affecting turnover rates. Our key result is that a job ladder motivating IT professionals with large, infrequent promotions will lead to higher turnover than a job ladder with smaller, more frequent promotions. We describe the conditions under which rearranging the job ladder creates economic value for the firm. We also offer an explanation for the observation that jobs characterized by boundary-spanning activities have higher turnover, and show that such jobs are more sensitive to the effect of likelihood of promotion on turnover. We test our hypotheses on a detailed data set covering 5,704 IT professionals over a five-year period. We confirm that likelihood of promotion has the predicted effects on turnover of IT professionals. A one standard deviation increase in likelihood of promotion decreases turnover by over 99%, consistent with our prediction. The empirical analysis also confirms the predicted effects of boundary spanning activities.