Overview of Constructed-Response Grading
Constructed-response grading is the process used to evaluate examinee responses that require written analysis, application of legal principles, and development of a work product. Unlike selected-response questions, which are machine scored, constructed responses are evaluated by trained human graders using standardized scoring materials and defined performance criteria.
The grading model is criterion-based. Responses are evaluated against established expectations for performance on each item as documented in the scoring guides and represented by example responses, rather than in comparison to other examinees. This approach is designed to support consistency, fairness, and alignment with the skills being measured.
Constructed-response grading is a core component of the overall scoring process and is conducted following completion of exam administration and post-exam reconciliation activities.
Operational Model
Constructed-response grading is operationalized through a structured, multi-step process designed to ensure consistency, quality control, and visibility across the grading lifecycle.
For each constructed-response item, scoring materials are developed that define the evaluation criteria, describe expected performance characteristics, and provide guidance on how scores should be assigned. These materials serve as the foundation for all grading decisions.
Graders are selected and onboarded by jurisdictions in advance of grading. Graders new to NextGen complete question-type training designed to familiarize them with the format of the constructed-response exam questions. After each exam administration, all graders complete question-specific training, called grader workshops, for each question they are assigned to familiarize them with the scoring materials, the purpose of the task, and the application of scoring standards. Calibration activities are conducted prior to and during grading to establish and maintain consistency across graders.
Grading is conducted within a dedicated digital grading platform. The platform supports assignment of responses, entry of scores, monitoring of grader activity, reconciliation, and management of quality-control workflows.
Each constructed response is independently evaluated and scored by two graders. If the difference between assigned scores exceeds an established threshold, the response is routed for additional review and resolution in accordance with defined reconciliation procedures. These thresholds differ by item type and the total number of points available for an item.
Throughout the grading process, progress and scoring activity are monitored to ensure timely completion and adherence to established standards.
Grading Workflow
The constructed-response grading process follows a consistent sequence of steps:
- Scoring guides are developed for each constructed-response item, including scoring criteria and performance expectations.
- Graders are selected, onboarded, and assigned to jurisdiction grading teams.
- Graders complete both general and item-specific required trainings on scoring materials and grading expectations.
- Calibration activities are conducted to establish scoring consistency across graders.
- Examinee responses are assigned to graders within the grading platform.
- Each response is independently scored by two graders.
- Scores are evaluated against defined thresholds for agreement.
- Responses requiring additional review are routed through reconciliation workflows.
- Grading progress and completion are monitored through the platform.
This process is designed to support consistent application of scoring standards while maintaining oversight and control across the grading operation.
Supervisor Awareness and Responsibilities
While supervisors are not responsible for grading constructed responses unless they are separately assigned as graders, they should understand how the grading process fits within the overall exam lifecycle.
Supervisors should be aware that constructed-response grading occurs after exam administration and may extend over a defined grading window. They should also understand that grading is conducted centrally within a controlled platform environment and that all responses must be successfully captured and transmitted at the conclusion of exam administration to be included in scoring.
Any issues affecting response capture, transmission, or completeness must be documented and escalated in accordance with established procedures, as these may impact the grading process.