Back to Features

Data Cleaning

Automated validation and quality assurance

Data quality is critical for meaningful insights. Our platform implements multiple layers of validation and cleaning, both during data entry and after collection, to ensure the highest quality data reaches your analysis systems.

Real-time Field Validation

As data is entered, the platform performs immediate checks to prevent common errors.

  • Required field enforcement
  • Format constraints (dates, numbers, text)
  • Skip logic enforcement
  • Range validation
  • On-spot error prompts

Automated Cleaning Rules

Post-collection rules that detect and flag data quality issues.

  • Inconsistency detection
  • Outlier identification
  • Cross-field validation
  • Sequence verification
  • Duplicate detection

EGRA/EGMA Standard Practices

We work closely with statisticians to document and formalize standard EGRA and EGMA data-cleaning practices typically applied by researchers, systematically encoding these as automated rules within the platform.

Assessment Sequence Checks

Flags when "last word read" is selected before "last error marked" in oral reading fluency tests.

Score Consistency

Validates that calculated scores match the marked responses and timing data.

Timing Anomalies

Detects unusually fast or slow assessment completions for review.

Configurable Rules

Administrators can define custom cleaning rules based on assessment design.

Administrator Review Workflow

1

Flag

System identifies problematic records

2

Queue

Flagged items added to review queue

3

Review

Admin reviews with full context

4

Action

Correct, exclude, or approve

Specification Requirements Addressed

Real-time validation during data entry
Skip logic enforcement
Required field validation
Format constraints
On-spot error prompting
Configurable validation rules
Automated data cleaning rules
Inconsistency detection
Outlier flagging
Administrator review workflow
EGRA/EGMA standard practices encoded