Package index • GradingAccuracy

Data Cleaning and Preparation

These functions are for preparing the original exports for accuracy computations, by deidentifying identifiable information and making any adjustments to rubric items.

read_evals(): Read Gradescope Evaluations CSV
validate_metadata_json(): Validate Metadata JSON
update_scores(): Update student scores based on metadata rubric items
update_scores_in_metadata(): Update Scores in metadata JSON
deidentify_graders(): Deidentify Graders
deidentify_gradescope_evals(): Deidentify Gradescope Evaluations
generate_rubric_texts(): Generate Rubric Texts
normalize_full_credit(): Normalize Full Credit

Accuracy Functions

These functions are for computing the accuracy rates between different graders.

generate_metadata_table(): Generate Metadata Table
generate_results_table(): Generate Results Table
generate_gt_results_table(): Generate Results Table in GT Format
expert_updates_table(): Calculate Differences in Expert Grades After QA/QC
export_grading_differences_xlsx(): Export Differences between Two Graders as Excel Spreadsheet
export_grading_differences_gt(): Export Differences between Two Graders as GT Table
find_differences_wrt_AI(): Find Differences with respect to AI Table
find_differences_wrt_students(): Find Differences with respect to Students Table
compute_mae_and_isp(): Calculate MAE and ISP
isp(): Calculate Proportion of Identical Scores
rubric_mae(): Mean Absolute Error of Rubric Items