When grading responses in Vervoe, it's important to note that each answer and its corresponding grade are ultimately up to the employer's discretion. This guide serves as a helpful framework, but it's essential to tailor your understanding of what constitutes "Poor," "Average," and "Excellent" to the specific needs and expectations of your organization.
Vervoe's grading scale falls into three categories from 0-10; poor, average, and excellent. The machine learning will prompt employers to grade responses and provide a suggestion to what category it think the answer falls into. Manual grading overrides AI grades and helps to train the machine learning for what each answer category looks like.
While this guide outlines general characteristics of responses within each range, it’s crucial to define how these levels look for your unique roles and business. What one company views as excellent may differ from another, so use this guide as a starting point, but always evaluate based on your company's standards.
Here's a framework for what each category means:
1. Poor (0-3)
- The candidate completely misunderstood or misinterpreted the question.
- The answer lacks substance and is not actionable or useful.
- The response shows an inability to meet the minimum job requirements.
2. Average (4-7)
- The candidate answered the question but missed opportunities to provide deeper insights or solutions.
- The response meets the basic expectations but does not exceed them.
- The candidate demonstrates knowledge but not mastery.
3. Excellent (8-10)
- The candidate not only answers the question but provides insightful, well-supported solutions.
- The response shows mastery of the subject matter and exceeds the standard for the role.
- The candidate provides exceptional detail, demonstrating critical thinking, advanced skills, or experience.
Your 10/10 answer should closely match or align with the ideal response you've provided as a correct answer sample. Correct answer samples are designed to represent the best possible answer you're seeking from candidates and help the machine learning understand what an excellent answer looks like.
Why use 0-10 grading?
Greater Precision in Scoring
- A 0-10 scale offers a broader range of options for assessing responses compared to a more limited scale (e.g., 1-5 or pass/fail). This allows evaluators to make more nuanced distinctions between candidates, especially when multiple responses fall within the same general category but differ in quality.
2. Flexibility for Different Skill Levels
- The scale accommodates a wide variety of skills and experience levels, offering flexibility in scoring responses that range from basic understanding to expert-level insight. This is particularly helpful when assessing candidates with varying degrees skill.
3. Customizable to Organizational Standards
- The 0-10 scale can be adapted to meet the unique needs of your organization. For example, an "excellent" response (8-10) in one company might require more specific technical detail, whereas another company may prioritize creative problem-solving. Having a range of scores allows employers to adjust their expectations to match their organizational standards.
4. Clear Differentiation Between Responses
- The scale makes it easier to differentiate between candidates who perform slightly above or below expectations. A candidate who scores a 6 (average) can be clearly distinguished from one who scores an 8 (excellent), ensuring a more accurate reflection of their potential fit for the role.
5. Consistency in Evaluation
- With more gradation in scoring, employers can develop a more consistent and systematic approach to evaluating candidates. This reduces the risk of overly subjective ratings and helps ensure that different evaluators can use the same criteria for more consistent results.
6. Improved Feedback for Candidates
- A detailed 0-10 scale provides more insightful feedback for candidates. Rather than only receiving general comments like "good" or "poor," candidates can be informed about where they stand within a more specific range, helping them understand areas of strength and improvement.
7. Alignment with Performance Levels
- The scale aligns with real-world performance levels. In many cases, people don’t simply fall into “good” or “bad” categories but lie somewhere on a spectrum. A 0-10 scale helps reflect the continuum of candidate performance, which is more representative of actual work situations.
For further information on Vervoe grading and machine learning, check out the following article.