What Is a Behaviorally Anchored Rating Scale (BARS)?

The first time I saw a manager rate two employees with identical output differently, I realized our review system was broken. One manager gave a “5 out of 5” to anyone who showed up on time and didn’t cause problems. Another gave “3 out of 5” to the top performers because she believed no one deserved a perfect score. Same company, same rating scale, different interpretations.

That experience led me to explore structured rating methods, and behaviorally anchored rating scales (BARS) proved to be the most effective solution. BARS links each rating level to specific, observable behaviors rather than leaving interpretation to individual managers.

This guide explains what BARS is, how to build one, when it works best, and where the trade-offs are. If your current review system suffers from inconsistency, BARS might be the fix your organization needs.

Behaviorally Anchored Rating Scale Definition

A behaviorally anchored rating scale is a performance appraisal tool that combines elements of narrative assessments and quantitative ratings. Each point on the scale is anchored to a specific behavioral example that describes what performance at that level looks like in the context of a given job.

Unlike a standard 1-to-5 rating scale, where “3” means something vague like “meets expectations,” a BARS defines “3” with a concrete behavior. For a customer service role, that might be: “Responds to customer inquiries within the SLA window and resolves routine issues without escalation, but misses opportunities to proactively address related concerns.”

The method was developed by Smith and Kendall in 1963 and has been refined since then. It’s grounded in the idea that performance ratings become more reliable when raters evaluate observable behavior rather than inferred traits. Instead of asking a manager to rate an employee’s “communication skills” on a scale, BARS asks the manager to identify which behavioral description best matches the employee’s actual performance.

BARS is used in performance management processes at organizations that prioritize evaluation consistency, particularly in regulated industries, government agencies, and companies with large management teams where calibration across raters is difficult.

How BARS Works in Practice

The manager reads these descriptions and selects the one that best matches the employee’s observed behavior over the review period. Some BARS formats allow managers to select between anchors (e.g., a “4” if behavior falls between the 3 and 5 descriptions).

The specificity of these anchors is what makes BARS effective. Two managers evaluating the same employee should arrive at similar ratings because they’re matching behavior to descriptions rather than interpreting abstract criteria. According to a ResearchGate study, BARS can reduce rating errors, including leniency bias, central tendency, and halo effect, compared with graphic rating scales.

• 7 (Excellent): Proactively identifies client needs before they arise, maintains regular contact outside of sales cycles, and consistently receives unprompted positive feedback from accounts.
• 5 (Proficient): Maintains scheduled touchpoints with key accounts, responds to client requests within 24 hours, and resolves concerns without needing manager involvement.
• 3 (Adequate): Contacts clients primarily during renewal periods, responds to inquiries within SLA timelines, but rarely initiates proactive outreach.
• 1 (Poor): Fails to respond to client inquiries within the SLA, misses scheduled check-ins, and receives recurring complaints about responsiveness.

Steps to Create a BARS

Building a BARS requires more upfront investment than a standard rating form, but the resulting instrument is more reliable. Here is the process I’ve used to develop BARS at two organizations.

Step 1: Identify performance dimensions

Start by defining the key performance dimensions for the role. These should represent distinct, important aspects of the job. For an HR coordinator, dimensions might include: benefits administration accuracy, employee communication, compliance documentation, cross-functional coordination, and problem resolution.

Keep the number of dimensions manageable. Five to eight dimensions work for most roles. More than that creates evaluation fatigue and reduces the quality of ratings.

Step 2: Generate critical incidents

Gather examples of actual behaviors that represent effective, average, and ineffective performance on each dimension. This is called the critical incident technique. The best sources are experienced managers and top performers who can describe real-world situations.

For each dimension, collect 15 to 20 behavioral examples. These should describe specific actions, not personality traits. “Responds to employee benefits questions within four hours” is a behavioral incident. “Is helpful” is not.

Step 3: Classify and scale the incidents

Have a separate group of subject matter experts (not the ones who generated the incidents) sort each behavioral example into the dimension they believe it belongs to. Incidents that are consistently placed in the same dimension are retained. Those with inconsistent placement are rewritten or discarded.

Next, have the experts rate each retained incident on the performance scale (e.g., 1 to 7) based on the effectiveness of that behavior. Calculate the mean rating and standard deviation for each incident. Incidents with high agreement (low standard deviation) become the final anchors for their respective scale points.

Step 4: Build and validate the instrument

Assemble the final BARS by placing the selected behavioral anchors at their corresponding scale points for each dimension. The result is a structured evaluation form in which every rating level for every dimension has a concrete behavioral description.

Before rolling it out, pilot the instrument with a small group of managers. Have them rate the same employees and compare results. High inter-rater agreement indicates the BARS is working as intended. Low agreement means the anchors need refinement.

Advantages of BARS

BARS addresses the fundamental problem with most performance rating systems: subjectivity. When I implemented BARS at a 200-person company, the variance in ratings between managers dropped by 35% in the first review cycle. Here’s why it works.

• Reduced rating bias: Behavioral anchors force managers to evaluate what employees do rather than how they feel about them. This reduces halo effect, leniency bias, and recency bias.
• Clearer expectations: Employees can read the BARS before the review period begins. They know exactly what behavior corresponds to each performance level, which makes the system transparent and motivating.
• Better feedback conversations: When a manager rates someone a “3” on a BARS, both parties can discuss the specific behaviors described at that level and the behaviors needed to reach a “5.” The conversation becomes concrete rather than abstract.
• Legal defensibility: BARS ratings tied to documented, job-related behaviors are more defensible in employment disputes than generic rating scales. Courts and agencies view behavior-based evaluations as more objective.
• Standardized evaluations: BARS creates consistency across managers, departments, and locations. This is particularly valuable in organizations with many managers who have different standards and evaluation styles.

Limitations and Challenges

The best approach is hybrid: use BARS for job-specific competencies and supplement with other evaluation methods (goal achievement, project outcomes, peer feedback) for dimensions that BARS doesn’t capture well.

• Development cost: Building a valid BARS requires significant time and effort. The critical incident technique, expert panels, and validation testing can take weeks for a single role. For organizations with dozens of distinct job families, the investment is substantial.
• Maintenance burden: Jobs evolve. When responsibilities change, the behavioral anchors become outdated. BARS requires periodic review and updating, which many organizations neglect after the initial implementation.
• Role specificity: A BARS for a software engineer won’t work for a marketing manager. Each distinct role (or role family) needs its own instrument. Standard rating scales can be applied across the organization. BARS cannot.
• Doesn’t capture everything: BARS focuses on observable behavior. It can miss important performance factors like strategic thinking, innovation, and leadership potential that don’t always manifest in specific, recurring behaviors.

BARS vs. Other Rating Methods

Understanding how BARS fits alongside other common appraisal methods helps organizations choose the right approach for their needs.

BARS vs. graphic rating scale

Graphic rating scales are the most common appraisal format. They use generic descriptors (“exceeds expectations,” “meets expectations,” “below expectations”) applied across all roles and competencies. They’re cheap and easy to implement, but suffer from inconsistent interpretation.

BARS replaces those generic labels with specific behaviors. The trade-off is development time for evaluation quality. If your organization has the resources to build BARS, it will produce more reliable and actionable ratings.

BARS vs. BOS (behavioral observation scale)

BOS, the behavioral observation scale, is a related method that asks managers to rate how frequently an employee exhibits specific behaviors. While BARS asks “which description best fits,” BOS asks “how often does the employee do this.”

BOS can be easier to develop because it doesn’t require the scaling and validation steps that BARS demands. However, BOS can produce inflated ratings because managers tend to over-report positive behaviors. BARS forces a choice between fixed descriptions, which constrains rating inflation.

BARS vs. MBO (management by objectives)

MBO evaluates employees based on whether they achieved predefined goals or objectives. It measures outcomes rather than behaviors. MBO works well for roles with clear outputs (sales targets, project deliverables) but struggles with roles where performance is more process-oriented.

BARS and MBO can complement each other. Use BARS to evaluate how an employee works (behaviors and competencies) and MBO to evaluate what they achieved (results). Together, they provide a complete picture of performance.

Behaviorally anchored rating scales solve a real problem: the inconsistency that plagues most performance review systems. When every manager interprets “meets expectations” differently, the review process becomes unreliable, and employees lose trust in it.

If your organization struggles with rating consistency, has managers who rate differently for similar performance, or faces legal scrutiny on employment decisions tied to performance reviews, BARS is worth the development investment. Start with your highest-volume roles, build the instruments carefully, and plan for ongoing maintenance. The reliability gains are significant when the implementation is done right.

FAQ

Here I answer the most frequently asked questions about behaviorally anchored rating scales.

What is a behaviorally anchored rating scale?

A behaviorally anchored rating scale (BARS) is a performance appraisal tool that uses observable behavioral examples as anchors for each point on a rating scale. Each rating level is defined by a concrete description of what performance at that level looks like for a particular job dimension.

When should an organization use BARS?

BARS is most valuable when rating consistency across managers is a priority, when performance evaluations have legal implications (promotions, terminations), when the organization has the resources to develop role-specific instruments, and when existing review systems suffer from rating inflation or inconsistency.

How long does it take to develop a BARS?

Developing a BARS for a single role takes 4 to 8 weeks, including the critical incident collection, expert classification, scale validation, and pilot testing. Organizations with many distinct roles may need several months to develop instruments across the full job family.

Can BARS be used for all types of jobs?

BARS works best for roles with recurring, observable behaviors that define performance, such as customer service, healthcare, manufacturing, administrative support, and similar positions. It’s less effective for roles where performance is measured through creative output, strategic decisions, or long-term project outcomes.

What is the difference between BARS and a standard rating scale?

Standard rating scales use generic descriptors (excellent, good, average, poor) applied across all roles. BARS replaces those generic labels with specific behavioral descriptions tailored to each job dimension. This specificity reduces subjective interpretation and yields more consistent ratings.