Data Quality

The Data Quality dashboard allows you to assess the completeness and consistency of data collected in the system.

Access

Menu: Monitoring & Evaluation → Data Quality

Overview

The dashboard displays quality indicators for the main system entities, enabling rapid identification of gaps and anomalies.

---

Main Statistics

Summary Cards

Indicator	Description
Overall Score	Average quality score across all entities
Households	Household form completeness score
Beneficiaries	Beneficiary form completeness score
Anomalies	Total number of detected anomalies

Score Color Coding

Score	Color	Interpretation
≥ 80%	Green	Good quality
60-79%	Orange	Average quality, improvements needed
< 60%	Red	Insufficient quality, action required

---

Filters

Filter by Region

Select a region to see only that area's data:

• Oio
• Bafatá

Tip

Analyze each region separately to identify areas needing more attention.

Refresh

Click the Refresh button to reload data in real time.

---

Member Statistics

This section displays metrics about household members:

Indicator	Description
Total Members	Total number of registered members
Households with Members	Households with at least one member
Households without Members	Households with no registered member (anomaly)
Average Members/Household	Average ratio

Attention

Households without members represent an anomaly to correct. Each household should have at least one member (the household head).

---

Scores by Category

Completeness scores are calculated by field category:

Evaluated Categories

Category	Description
Identification	Household code, location
Household Head	Name, gender, age, contact
Composition	Size, age groups
Housing	Type, materials, equipment
Geolocation	GPS coordinates

Score Calculation

For each category:

• 100% = All required and recommended fields are filled
• Partial = Some fields are empty
• 0% = No fields filled

---

Anomalies

Anomaly Types

Type	Severity	Description
missing_fields	Medium	Required fields not filled
invalid_data	High	Inconsistent or invalid data
duplicate	High	Potential duplicates detected
orphan_record	Critical	Record without required relations
out_of_range	Medium	Values outside acceptable limits

Severity Levels

Severity	Color	Priority
Critical	Red	Immediate action
High	Orange	Priority action
Medium	Blue	To fix
Low	Gray	To improve

Anomaly Distribution

Two visualizations are available:

1. By severity: Number of anomalies by criticality level
2. By type: Distribution by problem type

---

Anomaly List

The detailed table lists individual anomalies:

Columns

Column	Description
Severity	Criticality level
Type	Anomaly category
Entity	Code of concerned household/beneficiary
Message	Problem description

Filtering

The first 20 anomalies are displayed. To see more:

• Export data to Excel
• Use region filters

---

Use Cases

Identify Low Quality Areas

1. Access the Data Quality dashboard
2. Compare overall scores of different regions
3. Identify the region with the lowest score
4. Analyze problematic categories
5. Plan corrective actions

Fix Critical Anomalies

1. View the anomaly list
2. Filter by "Critical" severity
3. For each anomaly:
   • Note the code of the concerned entity
   • Access the household/beneficiary form
   • Correct the erroneous data
4. Refresh to verify correction

Improve Completeness Score

1. Identify categories with lowest scores
2. For each category:
   • List missing fields
   • Plan supplementary collection
3. Update concerned forms
4. Track score evolution

Check Households Without Members

1. Check the "Households without Members" statistic
2. If > 0, export the anomaly list
3. Identify concerned households
4. Add missing members
5. Verify each household has at least the household head

---

Best Practices

Regular Monitoring

• Check the dashboard at least once a week
• Track score evolution over time
• React quickly to critical anomalies

Prioritization

• Address critical anomalies first
• Focus on low-score categories
• Target most problematic regions

Prevention

• Train enumerators on required fields
• Validate data before import
• Use quality controls during data entry

---

Integration with Other Modules

Households

Household-related anomalies can be corrected directly from the Household Management module.

Beneficiaries

Beneficiary anomalies are accessible via the Beneficiaries module.

Reports

Quality indicators feed into Reports and DLI calculations.