Home → General Knowledge Book → Important Information for 2026 → Enhanced Scoring Quality Indicators in the New Scoring System
1.5. Enhanced Scoring Quality Indicators in the New Scoring System
We want to share an update on recent enhancements to the scoring quality indicators and clarify our performance expectations for the ClearSight project.
Enhanced Scoring Quality Indicators
The color-coded scoring quality indicators you see during scoring are designed to provide actionable feedback to help you improve the quality of your scoring. These indicators—Accuracy, Drift, and Rate—reflect how closely and consistently your scores align with established scoring standards, based on your performance on validity responses.
Previously, a single set of indicators was used to reflect overall scoring quality. We learned, however, that this approach provided insufficient—and in some cases misleading—feedback for adjusting scoring on individual traits. Because Holistic and Conventions scoring often show different patterns, we have updated the interface so that you now see separate indicators for each trait. This allows you to fine-tune your scoring for Holistic and Conventions independently.
How to Use the Indicators
Drift is the most common quality issue and should be addressed first. If your Drift indicator is orange or red, this signals a systematic scoring trend. A negative value indicates severity (scores tending too low), while a positive value indicates leniency (scores tending too high). Adjust your scoring until the indicator returns to green or blue. Reviewing training materials and paying careful attention to scoring lines can be especially helpful in correcting drift.
Accuracy may also require attention. If your Accuracy indicator is orange or red, this suggests inconsistent application of the scoring criteria. In this case, carefully review the training materials—particularly anchor responses and the “things to look for” guidance—to reinforce your understanding of the rubric.
Because indicators are based on validity responses, they update approximately every 10–20 responses you score. Please avoid over-correcting during this period. Also note that indicator values are specific to the item you are currently scoring.
Quality Expectations
We believe these enhanced, trait-specific indicators provide more useful feedback for improving scoring quality. If scoring quality issues—indicated by persistent orange or red indicators—are not corrected in a timely manner, you may be prohibited from scoring items on which you are experiencing difficulty. If no other qualified items are available, this may result in release from the ClearSight project.
Thank you for your continued hard work. We hope these updates support your success and clarify MI’s expectations. As always, please contact HelpSpot if you have any questions.
This page was: Helpful |
Not Helpful