Let's reflect on processes : task uncertainty as a moderator for feedback effectiveness

Traditionally, when introducing feedback interventions, the focus of performance indicators used as the basis for feedback has mainly been on the final results of tasks. This stemmed from the general contention that providing employees with information about their performance on the final results of their work unconditionally increases their performance. Only over the last few decades have researchers come to realize that findings regarding feedback effectiveness have not been consistent. More recent research has already suggested that several characteristics of feedback and task might act as significant moderators. However, until now, these moderating conditions remained poorly understood. The research reported in this dissertation aimed to contribute to the understanding of feedback effectiveness and examined the until now underexplored combined moderating effect of task uncertainty, type of feedback (outcome versus process feedback) and feedback reflection on feedback effectiveness. The main research question, ‘Dependent on the level of task uncertainty, what type offeedback should employees be provided with for feedback to be effective?’, was addressed in three separate, yet closely related studies. In all studies, the well validated ProMES method (Productivity Measurement and Enhancement System) was used for the development and provision of performance feedback. In the first study (Chapter 2), a meta-analysis, the combined moderating effect of task uncertainty, type of feedback, and feedback reflection on the effectiveness of ProMES feedback was examined using performance data from 83 field studies from a wide variety of organizational settings. Results indicated that when task uncertainty is low, employees higher on reflection on feedback outperform employees lower on reflection on feedback, irrespective of the type of feedback they receive. However, when task uncertainty is high, employees higher on reflection on feedback outperform employees lower on reflection on feedback when the proportion of process feedback is higher. Moreover, the reverse is true when the proportion of process feedback is lower. Thus, this study showed through meta-analysis over a large variety of tasks, teams, organizations, and industries that the effectiveness of feedback on performance is not at all straightforward. Task uncertainty, type of feedback, and reflection on feedback appear to be important moderating conditions for feedback effectiveness, where some combinations of these variables can lead to very large positive effects and others can actually lead to negative effects on performance. In the second study (Chapter 3), a quasi-field experiment, it was examined whether task uncertainty influences the type of performance indicators participatively developed by practitioners from the field. For this purpose, a task uncertainty framework was defined. Then, 50 care providing employees divided over 8 medical rehabilitation teams varying on task uncertainty participated in the development of performance feedback systems using the ProMES method. Results indicated that teams higher on task uncertainty develop relatively more process indicators (compared to outcome indicators) than teams lower on task uncertainty. Moreover, in line with the task uncertainty framework, process indicators developed by teams higher on task uncertainty are more of a problem solving nature, whereas process indicators developed by teams lower on task uncertainty are more of a procedural nature. Thus, this study showed through quasi-field research in health care that the level of uncertainty employees are dealing with during care provision determines which types of indicators are regarded as helpful with the successful fulfillment of their tasks. In the third study (Chapter 4), a quasi-field experiment, it was examined whether task uncertainty, type of feedback, and feedback reflection have a moderating effect on performance with the sequential introduction of outcome and process feedback. For this purpose, 107 care providing employees, belonging to 4 medical rehabilitation teams varying on task uncertainty, periodically received performance feedback through ProMES feedback reports, which were discussed in feedback meetings. Results indicated that a three-way interaction exists between the level of task uncertainty, the type of feedback, and the time spent on reflection on feedback during feedback meetings, such that only with higher levels of task uncertainty and higher levels of reflection process feedback results in higher performance than outcome feedback. In addition, longitudinal questionnaire data from arepeated measures design with three time waves were used in this study to examine the combined effect of task uncertainty and type of feedback on factors enabling the development and use of task knowledge, such as coping with task uncertainty, task information sharing, role clarity, and empowerment. The results indicated that with higher levels of task uncertainty, only the introduction of process feedback (compared to outcome feedback) has a positive effect on these supposedly performance-enhancing factors. Thus, this study showed through a quasi-field experiment in health care that the effectiveness of feedback is dependent on the level of task uncertainty, the type of feedback, and the level of reflection on feedback. Through reflection on process feedback, employees dealing with higher levels of task uncertainty are presented with the opportunity to develop and use appropriate task knowledge. In conclusion, the research presented in this dissertation confirmed through meta analysis and empirical field research that the effects of feedback are not always the same. Instead, moderating conditions such as task uncertainty, type of feedback, and reflection on feedback play important roles in the effects of feedback, both on performance and on underlying psychological factors such as coping with task uncertainty, task information sharing, role clarity, and empowerment. Ignoring these findings when designing and implementing performance feedback systems could be harmful for organizations.