Direct Answer
Leniency bias is the tendency for raters to assign scores that are higher than the person's actual performance warrants. It is one of the most persistent and well-documented problems in workplace evaluation — whether in performance appraisals, letters of recommendation, or employment references. When leniency bias is present, ratings cluster at the top of the scale, making it difficult or impossible to distinguish between truly exceptional performers and merely adequate ones.
Why It Matters
Consider a manager reviewing performance ratings for a team of ten people. If nine of them receive "exceeds expectations" and one receives "meets expectations," the ratings tell you almost nothing about who is actually performing well. The manager may have a genuinely strong team — or they may be displaying leniency bias, inflating ratings to avoid difficult conversations, maintain team harmony, or simply because they default to positive evaluations when no structure guides them otherwise.
This is not just a measurement nuisance. Leniency bias undermines the validity of any evaluation system built on ratings. If the underlying scores are inflated, then any decisions based on those scores — promotions, hiring, development plans — are being made on distorted information.
The Science Behind It
Leniency has been studied for over a century in organizational psychology, and the findings are consistent: it is pervasive, stable, and driven by identifiable psychological factors.
Bernardin et al. (2009) defined leniency as "the tendency on the part of raters to assign ratings that are higher than are justified by the actual performance level" (p. 300). In a study of peer ratings under low-accountability conditions, they found that rater personality predicted leniency: Agreeableness was positively correlated with rating level (r = .18, p < .05), while Conscientiousness was negatively correlated (r = -.20, p < .05). Raters who were both more agreeable and less conscientious produced the most lenient and least accurate ratings.
Bernardin et al. (2015) extended this work to a field study of Fortune 500 retail managers (N = 125), finding that leniency was stable across different rating situations — the same raters tended to be lenient regardless of whether they were rating subordinates, peers, or managers. Performance management competence independently predicted rating accuracy, and the most lenient raters were consistently more agreeable, less assertive, and less competent at performance management.
Randall and Sharples (2010) demonstrated that conflict avoidance is a key motivator of leniency. In their experimental study (n = 230), three independent factors predicted inflated ratings of poor performance: high rater Agreeableness, inflated self-ratings from the ratee, and the prospect of future collaboration with the person being rated.
Common Misconceptions
People often assume leniency is a deliberate choice — that raters consciously decide to inflate scores. The research suggests otherwise. Leniency is largely driven by unconscious social motivations: the desire to avoid conflict, the discomfort of delivering negative feedback, and the wish to maintain positive relationships (Randall & Sharples, 2010). This is precisely why it is so difficult to eliminate through instructions alone. Telling raters to "be honest" does not address the underlying psychological pressures that produce inflation.
How This Connects to Better Hiring
Leniency bias explains why unstructured references and open-ended letters of recommendation produce ratings that are overwhelmingly positive and virtually indistinguishable across candidates. The solution is structural: behaviorally anchored questions, consistent rating scales, and formats that require raters to make specific comparisons rather than global judgments. Structure does not eliminate leniency entirely, but it reduces its impact by channeling raters toward more precise, differentiated assessments.