Rating scales best practices

Contents

Choosing the right rating scale for your 360 review questions is crucial for gathering meaningful, actionable data. While open-ended questions provide rich qualitative feedback, rating scales offer a way to collect quantitative data that can be analyzed objectively.

This combination gives a comprehensive view of performance, allowing you to see trends, compare feedback from different sources, and make data-informed decisions. A well-designed rating scale reduces ambiguity and provides clarity for reviewers, leading to more accurate and useful results.

Types of Rating Scales

When building your review template, you’ll have the option to introduce quantitative, close-ended questions that are responded to using a rating scale. You will be able to select the anchors on your rating scale. This means you’ll not only write the question, but you will also determine what rating scale to use (e.g., Will it be an agreement scale? A frequency scale?), and how each value on your scale is defined. The most common types or rating scales for performance reviews are Likert scales. The type you choose should align with the specific question you’re asking.

Likert scales are ideal for measuring attitudes, agreement, or satisfaction. They typically use a range of options from one extreme to another.

_ Agreement Scales are best for questions like, “This person effectively communicates their ideas.” or “This person consistently demonstrates our company’s core values.” Example:

  • Strongly disagree
  • Disagree
  • Neutral
  • Agree
  • Strongly agree

_ Effectiveness Scales are best for questions like: “How effective is this person at resolving conflicts?” or “How effective is this person at leading team meetings?” Example:

  • Not Effective
  • Slightly Effective
  • Moderately Effective
  • Very Effective
  • Extremely Effective

Frequency scales are another type of Likert scale and can used to measure how often a behavior or action occurs.

_ Behavioral Frequency is best for questions like: “This person provides constructive feedback to their peers.” or “This person is proactive in seeking out new challenges.” Example:

  • Never
  • Rarely
  • Sometimes
  • Often
  • Always

_ Time-based Frequency is best for questions like: “How often does this person share updates on their project progress with the team?” Example:

  • Never
  • Once a month
  • Once a week
  • Several times a week
  • Daily

Rating Scale Anchors

Anchors are the labels you use to define each point on your rating scale. Clear anchors are essential for ensuring consistency and accuracy in feedback.

Avoid using only numbers (e.g., 1 to 5). Instead, use words that clearly define each point. For example, a scale with anchors like “Strongly Disagree,” “Disagree,” “Neutral,” “Agree,” and “Strongly Agree” is far more effective than just “1,” “2,” “3,” “4,” “5.”

The labels should represent a balanced range of options from one extreme to the other. A scale with “Bad,” “OK,” “Good,” and “Excellent” is asymmetrical and can skew responses towards the positive end. A better, more balanced scale would be “Poor,” “Fair,” “Good,” and “Excellent.”

For certain contexts, it can be helpful to provide a brief description for what each anchor means in the context of your organization. For example, “Exceeds Expectations” could be defined as “Consistently performs above the required level, demonstrating exceptional skill and initiative.”

The language of your anchors should be easy for all reviewers to understand without requiring them to consult a reference guide. Avoid jargon or overly technical terms.

Using Not Applicable or NA

At times it might make sense to include an “NA” or “Not Applicable” option. This can improve the quality of your feedback because it can prevent reviewers from having to give a forced rating when they don’t have enough information to do so accurately. Providing an NA option isn’t always appropriate, but it’s useful for questions that might not be relevant to everyone’s working relationship with the reviewee. For example:

  • “This person is effective at resolving conflicts between team members.” A peer might not have witnessed this if they don’t work on the same projects or are not in the same meetings.
  • “This person actively mentors junior employees.” A direct manager would likely be able to answer this, but a peer who is not on the same team might not know.

Without an NA option, a reviewer who doesn’t have an opinion might choose “Neutral,” “Average,” or even a random rating, which can skew your data. The NA option allows them to skip the question entirely, ensuring the data you do collect is based on actual observations and experience.

The Number of Points on your Scale

A key decision when creating your scale is how many points to include. The most common options are odd-numbered scales (like 5 or 7 points) and even-numbered scales (like 4 or 6 points). Each has its own pros and cons.

Odd-Numbered Scales (e.g., 5-point, 7-point)

  • Pros:
    • A neutral option: Having a middle point (like “Neutral” or “Neither Agree nor Disagree”) allows reviewers to indicate a lack of strong opinion, a lack of experience with the behavior, or a moderate stance. This can prevent forced answers and provide a more accurate picture.
    • Intuitive and widely used: The 5-point scale is a familiar standard in surveys and reviews, making it easy for reviewers to understand and use.
  • Con: The “Neutral” trap: Some reviewers may overuse the middle option to avoid providing a definitive rating, potentially diluting the value of the feedback.
  • When to use: A 5-point scale is generally recommended for most 360 review questions, especially when you want to allow for a neutral or undecided response. Use a 7-point scale for more nuanced topics where greater distinction between ratings is desired.

Even-Numbered Scales (e.g., 4-point, 6-point)

  • Pro: Forces a choice: By removing the middle “neutral” option, reviewers are pushed to take a stance, either positive or negative. This can lead to more decisive feedback and a clearer understanding of strengths and weaknesses.
  • Cons: Forced answers: Reviewers might feel pressured to choose a side, even if they don’t have a strong opinion, which can lead to inaccurate or misleading data.
  • When to use: A 4-point scale is useful when a clear opinion is necessary and you want to avoid ambiguity. This can be particularly effective for questions about critical competencies where “neutral” isn’t a helpful answer.

While a 1-10 scale may seem to offer more granularity, it can lead to problems. The difference between a “7” and an “8” can be highly subjective and hard to define consistently across different reviewers. This can introduce noise into your data and make it difficult to interpret. For performance reviews, a scale with fewer, clearly defined points is generally more effective.

Choosing a Rating Scale

A good rating scale is one that is appropriate for the type of feedback you are seeking and provides clear, unambiguous options for the reviewer. Here are some best practices in implementing rating scales:

Match your question to your scale based on the type of feedback you need. Is your question asking about frequency (how often a behavior occurs) or agreement/effectiveness (the extent to which someone agrees with a statement or is effective at a task)?

  • If your question is a statement like, “This person manages their time effectively”, use an Agreement Scale: Strongly Disagree, Disagree, Neutral, Agree, Strongly Agree.
  • If your question asks “How often…?” use a Frequency Scale: Never, Rarely, Sometimes, Often, Always.
  • If your question asks “How effective…?” use an Effectiveness Scale: Not Effective, Minimally Effective, Somewhat Effective, Very Effective, Extremely Effective

Clearly define each point on the scale: Don’t just label your scale with numbers (e.g., 1 to 5). Instead, provide clear, descriptive labels for each point (e.g., 1 = Strongly Disagree, 5 = Strongly Agree). This ensures everyone interprets the scale in the same way.

  • 5-point scales are generally recommended. They’re intuitive, widely used, and include a neutral option (“Neutral” or “Neither Agree nor Disagree”) which allows reviewers to opt out if they don’t have enough information to provide an opinion. This can prevent forced answers and lead to more accurate data.
  • 4-point scales are useful when you want to force a choice. By removing the neutral option, reviewers must lean one way or the other. This can be helpful for critical competencies where you want a clear positive or negative rating.
  • 7-point scales provide more granularity and can be used for more nuanced topics, but may be more difficult to distinguish between points.

Pair quantitative with qualitative: Include a space for open-ended comments after a rating scale question. This allows reviewers to provide context and elaborate on their ratings, giving depth to the numerical data.

Keep it consistent: For questions within the same category, use the same rating scale. This makes the review process smoother for the user and the data easier to analyze. For example, if you’re asking about “leadership skills,” use the same “Effectiveness Scale” for all questions in that section.

Additional resources

Creating a review template Documentation Creating custom 360 feedback questions Documentation Best practices when completing an employee review Education Having successful review conversations Education

To inform us of a typo or other error, click here. To request a new feature, click here.