New research highlights the critical impact that three factors — target, distance and valence — have on how people receive and react to peer performance reviews
In Brief: Employee feedback meetings are part of almost all management models in use today; however, exactly what makes feedback effective is not always well understood. New research finds that three feedback components — target, distance, and valence — determine how feedback is received and the likelihood it will lead to increased employee performance.
Analysis: Providing employees (and in some cases peers and managers) feedback on their performance is such a ubiquitous part of business life that it’s easy to assume that we understand how it works. However, the reality is that employee feedback is a complex issue and thus a topic of continued research focus. Indeed, a recent study from a team at UCLA makes a strong contribution to this field by breaking down the process of feedback into three dimensions and then asking how each of these dimensions impacts the efficacy of the feedback process.
As most managers understand, when we give feedback to an employee, we’re typically communicating at least three things: the reference point (e.g., average or high performance), the employee’s status against that reference, and the gap from actual to target performance. The authors of this new study note that even this simple scenario presents a manager with a set of interesting decisions:
Suppose you are a manager wanting to motivate your employees to work harder. You have heard from your colleagues at other firms that they use a quarterly leaderboard to motivate their staff through peer comparisons, so you decide to do the same. Along with providing individual performance feedback in a report to each employee, the report also compares the employee’s performance relative to their peers. But which peers should you choose for this report? For example, should you compare each employee to the median team performance? Or should you compare each employee to a group of high performers from the team, such as the top 20%? If you provided the median comparison, someone in the 52nd percentile would be two points ahead of the reference group. But if you provided the top-performer comparison, that very same person would be 28 percentile points behind the reference group.
Since most managers want feedback to improve performance, the situation noted above led the researchers to wonder how the manager’s selections would influence employee reaction and subsequent performance. Put another way, the authors wanted to understand what feedback would be the most effective in increasing employee performance?
To better understand how feedback works, the team adopted a model that separates feedback into three parts:
Target, e.g., average versus exemplary performers
Distance, i.e., being near versus far from a reference group
Valence, e.g., being better or worse than a reference group
The team then conducted a series of studies with hundreds of participants in different settings. One study asked households to conserve energy during a multi-month period, during which they were provided regular feedback on their conservation performance. In this study, one group was compared to the average household while another was benchmarked against the top performers. In another study, participants were asked to download an app to track their daily steps and were, again, provided with regular detailed information against benchmarks that were either closer to average or to the best performers. Furthermore, in a third study, different participants were asked to complete transcription assignments, again with feedback varying to test the impact of various feedback strategies.
Interestingly, though the participants and measured tasks varied, the overall results fit consistent patterns, and the team’s general findings are outlined below.
1: Positive feedback motivates more than negative, but target selection is critical. When positive feedback is given, most people will try to improve their performance. Overall, note the authors, “we see preliminary evidence that the Valence of feedback—independent of the specific Target or the Distance of that feedback—may play a significant role in shaping future behavior.” Put another way, the more positive feedback that people receive, the more likely they may be to continue trying to increase their performance. Moreover, “receiving negative social comparison information reduces people’s sense of their ability to match the Target benchmark, which in turn reduces their post-feedback performance.”
2: Target identity matters. While the conclusion noted above is generally true, it is not valid when the target presented is seen as either false/irrelevant or impossible to reach. This conclusion is especially true when the target defined is a group of “high performers” and employee views of that group are not positive. As the authors note:
…response to normative feedback is shown to depend on people’s attitudes toward and beliefs about the reference group. If people do not admire or identify with the higher-performing reference group, or if they believe the high-performing group is exceptional and therefore less relevant to themselves, a social comparison against a high-performing reference group may have little influence on them, and thus they may not increase subsequent effort.
With this finding in mind, managers should be confident that employees held up as role models are seen as worthy by others. Simply demonstrating high performance against company standards is not enough to ensure that other employees will see the HiPo group as a valid target and thus worthy of emulation.
3: The greater the distance to the target, the lower the chance that performance will improve. This is not a new finding, as previous work has established that the further someone is from a peer, the more likely it is that attempts to close the gap will diminish. However, this study provides an evolution of this idea in the feedback setting, where, the authors note, “a reference group that lies closer to one’s current performance may be more motivating than a reference group that is farther away.” In other words, asking a good “B player” to be the best B player may be more effective than asking her to reach the level of the best A player. Indeed, in the energy consumption study, comparison to average neighbors led to conservation, but comparison to the most efficient neighbors demonstrated increased energy use. Likewise, a true A player is more likely to be motivated by being slightly behind the top performer than by being far ahead of the average.
4: Feedback can make temporary performance permanent. This finding highlights a serious problem, which is that “when people find out that others are performing better than they are on a given task, they start to view the task as less important to their self-definition, which in turn causes them to exert less effort in that domain.” In other words, when feedback consistently shows gaps in performance against the desired standard, many people internalize that momentary state as an indication of permanent potentiality. Thus, it beholds managers to think carefully about when and how often performance is given to those struggling with reaching the desired state. As the authors explain:
Someone who constantly receives similar feedback may quickly develop a stable identity as an under- or top-performer, leading to long-run behavior change. If, however, frequent feedback typically vacillates between positively and negatively valenced messaging, this could have an overall backfiring effect if it undermines a person’s feelings of self-efficacy and control.
After considering all of their findings, the researchers conclude that “peer comparison feedback is not uniformly helpful but depends on the specific Target, Distance, and Valence for each feedback recipient.” Moreover, “whenever choice architects intent on providing normative feedback have the option to select a reference group, they should do so carefully.” This is because “in choosing a reference group, the choice architect is also choosing to selectively motivate some and demotivate others based on the proportion of people who will receive negatively valenced feedback and how far from the Target people will be on average.”
The findings from this series of studies are insightful, and they suggest other areas worthy of analysis. For example, the frequency of feedback sessions is a topic this research does not address. Modality — in person, virtual, or written, for example — would also be interesting areas for further research. Nonetheless, the authors have done an admirable job of providing a more nuanced and carefully calibrated way for managers to think about peer feedback design. Indeed, managers should remember that absent careful design, comparison feedback can backfire significantly. Therefore, recommend the authors, managers “might instead consider implementing individually tailored social comparisons whereby different people, depending on what they care about and what their baseline performance is, are compared against different benchmarks.” This approach would undoubtedly make feedback models more challenging to create and implement, but that effort may reward leaders with significantly better overall performance.
Jonathan E. Bogard, Magali A. Delmas, Noah J. Goldstein, I. Stephanie Vezich. Target, distance, and valence: Unpacking the effects of normative feedback. Organizational Behavior and Human Decision Processes, Volume 161, Supplement, 2020, Pages 61-73, ISSN 0749-5978. https://doi.org/10.1016/j.obhdp.2020.10.003