The Custom Item report job is a versatile process within Item Bank Health Check (IBHC) that enables you to execute personalized queries against a selected Collection of Items.

Unlike standard IBHC jobs that apply predefined analytical frameworks, this functionality allows you to define unique criteria and generate tailored reports using your own prompts, to address specific questions about your assessment content.

When to use this job

The Custom Item report job is particularly effective when you need to:

• Audit unique standards: Analyze Items against criteria specific to your institutional curriculum or quality benchmarks.
• Run specialized analyses: Conduct one-off evaluations that fall outside the scope of other available IBHC job types.
• Synthesize multiple dimensions: Evaluate several analytical categories in a single pass across your Item bank.

Running the job

• Open the Learnosity Console and navigate to Item Bank Health Check.
• Create a Collection or select an existing one with the Item bank you want to analyze.
• Click on Create and select Custom Item report from the menu of available job types.
• Initiate the process by clicking Run job.

The time required for completion depends on the volume of content within the Collection.

Writing effective queries

The utility of your report depends on the clarity of the query expression. Consider these guiding principles:

• Maintain high specificity: Rather than general quality checks, define exact attributes like word counts or grammatical components.
• Establish clear benchmarks: Explicitly describe any standards against which Items should be measured.
• Manage scope: Narrow your parameters to ensure the resulting report remains actionable.
• Provide examples: Describing characteristics of representative sample Items can improve AI detection accuracy.

Reviewing results

Following the analysis of the Collection, the report is available for evaluation and download.

The structure of the findings will mirror the parameters of your query. If the results do not meet your requirements, refine the query terminology and re-run the job.