Absence of conflict of interest.
Bonet, R. (2014). High-involvement work practices and the opportunities for promotion in the organization. Industrial Relations, 53(2), 295-324. doi:10.1111/irel.12057
- The study’s objective was to examine the impact of employees’ participation in high-involvement work practices (HIWP) on their expected promotions.
- The author used regression methods in a nonexperimental analysis to estimate impacts of participating in HIWP, drawing on self-reported data from a survey of for-profit company employees.
- The study found that employee participation in more HIWP was associated with a greater likelihood of expected promotions.
- The quality of causal evidence presented in this report is moderate because it was based on a well-implemented nonexperimental design. This means we are somewhat confident that the estimated effects are attributable to participation in HIWP, but other factors might also have contributed.
High-Involvement Work Practices (HIWP)
Features of the Intervention
Historically, companies have had an inflexible hierarchical structure with top-down decision making and narrow roles for employees. In recent years, companies have increasingly delegated and decentralized decision making to employees with broader job definitions, resulting in higher employee involvement. When participating in HIWP, employees acquire new skills through formal training and informal on-the-job learning. In this study, employees reported the extent to which they engaged in five HIWP: (1) working in self-managed teams with shared decision making, (2) working in any team or group, (3) rotating jobs to learn additional skills, (4) negotiating work-related issues with coworkers, and (5) making suggestions to management to improve productivity.
Features of the Study
Drawing on self-reported data from employees who did and did not participate in HIWP, the study used regression methods to estimate the impact of participation on expected promotions. The statistical model controlled for employee-level characteristics, including age, gender, race, and previous promotions. It also controlled for firm-level characteristics, including number of employees, employment growth, and percentage of female employees.
Data came from the National Employer Survey, a nationally representative survey of for-profit employers and employees, conducted in 1997 and 2000. The final sample included 571 employees at 74 establishments covering a variety of sectors. Thirty-five percent of establishments were in the manufacturing sector, and 58 percent operated at multiple locations. The average age of employees in the study was 42 years old. Fifty-two percent of study participants were female, and 14 percent were members of a minority racial group.
- Employees who reported participating in a larger number of HIWP were more likely to expect promotions.
- Participating in two HIWP in particular—self-managed teams and coworker negotiations—was positively and significantly related to expected promotions.
- Participating in two other HIWP—teamwork and job rotation—was not significantly related to expected promotions.
- Participating in one HIWP—making suggestions to owners and managers—was negatively and significantly related to expected promotions.
Considerations for Interpreting the Findings
The statistical model included an extensive list of control variables, including past promotions, to account for differences in observed characteristics between HIWP participants and nonparticipants. As with any nonexperimental design, however, it is possible that unobservable factors, such as individual motivation, might have contributed to the estimated effects.
HIWP might have varied substantially at different employers. Employees and employers developed their own practices independently of the research and each other. Therefore, the actual intervention implemented likely varied substantially from one employer to another.
Causal Evidence Rating
The quality of causal evidence presented in this report is moderate because it was based on a well-implemented nonexperimental design. This means we are somewhat confident that the estimated effects are attributable to participation in HIWP, but other factors might also have contributed.