Evidence Harmonization

Analysis and synthesis of Social Welfare Services clustering, Decision Tree, and Apriori outputs.

Locality Filter All Mindanao records

Showing 4500 of 4500 respondent records. Predictive model scores remain the full-sample reference unless retraining is explicitly run.

Finding 1

Citizen segment evidence

Available

The clustering result identifies citizen segments and shows whether any segment has lower awareness, lower availment, lower satisfaction, or higher need for action.

Finding 2

Classification evidence

24 indicator models

The decision-tree classification uses citizen segment membership and profile variables as predictors. Most current indicator results are grouped as needs more predictors, which guides how strongly the result may be interpreted.

Finding 3

Association-rule evidence

Lift 1.09

The strongest repeated service-stage pattern is: Respondents who were aware of the social_welfare, and availed the social_welfare had a 97% likelihood of also were satisfied with the social_welfare (Lift: 1.09). This pattern was observed together 96.7% of the time for the eligible records.

What do the analytical results collectively show?

For Social Welfare Services, the harmonized evidence identifies this service-stage concern: Need for Action in Older Persons / Senior Citizens Program (60.7%). The recurring factor to check is MCA Dim1 (MCA profile dimension) appears in 17 model(s) across 4 stage(s). The repeated service-stage pattern is Satisfaction pattern: 96.7% confidence, lift 1.09. Together, these findings show whether the service-stage, citizen-segment, classification, and association-rule evidence point toward the same analytical interpretation.

Evidence agreement: strong consensus. The findings should be checked against local service records and field context before they are used for planning. Proposed actions, responsible offices, and monitoring indicators are presented separately in the Decision Support module.

This page harmonizes descriptive, clustering, classification, statistical, and association evidence for examination and validation. Proposed LGU actions are presented separately in the Decision Support module.

Consensus Feature Analysis

Evidence Source Agreement

4 sources
Consensus evidence checks whether different analytical outputs point toward the same interpretation. It helps avoid relying on only one model or one metric.
Evidence Sources 4
Evidence Agreement Strong consensus
Evidence Source Main Signal Interpretive Value Status
Descriptive service-stage review Need for Action in Older Persons / Senior Citizens Program (60.7%). Identifies the service-stage issue that should be checked first. Available
Chi-square / statistical support Sex shows strong relationship by Chi-square (effect 0.543, p=<0.001). Shows which profile variable meaningfully separates citizen segments before planning targeted action. Available
Decision Tree recurring variable MCA Dim1 appears across 17 model(s) and 4 stage(s). Identifies a repeated predictor that can guide validation and targeting. Available
Association rule mining Satisfaction pattern with 96.7% confidence and lift 1.09. Shows which service-stage responses repeatedly appear together and should be validated. Available
Social Welfare Services has converging evidence from descriptive review, statistical support, classification, and association rules. The combined result supports a strong integrated analytical interpretation.
Recurring Predictor Evidence

Recurring Variables Across Models

15 variables
This review counts which predictors repeatedly appear with positive importance across the saved indicator-level Decision Tree models. It helps identify citizen characteristics that consistently support interpretation across service-stage outcomes.
Models Reviewed 24
Recurring Predictors 23
Most Repeated Predictor MCA Dim1
Main Predictor Group MCA profile dimension
Recurring Variable Evidence Group Model Coverage Stage Coverage Analytical Interpretation Evidence Relevance
MCA Dim1 MCA profile dimension 17 models Appears in 17 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This is a combined profile signal from MCA. It should be read as a summary of related respondent characteristics, not as one survey question. Review the underlying profile traits behind this MCA signal before translating it into a specific program action.
MCA DimMagnitude MCA profile dimension 17 models Appears in 17 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This is a combined profile signal from MCA. It should be read as a summary of related respondent characteristics, not as one survey question. Review the underlying profile traits behind this MCA signal before translating it into a specific program action.
MCA Dim2 MCA profile dimension 16 models Appears in 16 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This is a combined profile signal from MCA. It should be read as a summary of related respondent characteristics, not as one survey question. Review the underlying profile traits behind this MCA signal before translating it into a specific program action.
Highest Educational Attainment Socio-demographic profile 15 models Appears in 15 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This respondent profile characteristic may help explain differences in awareness, service use, satisfaction, or need for action. Use the profile characteristic to refine target groups for information campaigns, eligibility review, or monitoring.
Age Group Socio-demographic profile 12 models Appears in 12 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This respondent profile characteristic may help explain differences in awareness, service use, satisfaction, or need for action. Use the profile characteristic to refine target groups for information campaigns, eligibility review, or monitoring.
Source of Information Housing and living conditions 12 models Appears in 12 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. Household conditions or information access may be linked with how citizens know about and use LGU services. Check whether household access conditions affect service reach, communication channels, or ability to avail services.
Source of Drinking Water Housing and living conditions 11 models Appears in 11 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. Household conditions or information access may be linked with how citizens know about and use LGU services. Check whether household access conditions affect service reach, communication channels, or ability to avail services.
Employment Status Socio-demographic profile 10 models Appears in 10 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This respondent profile characteristic may help explain differences in awareness, service use, satisfaction, or need for action. Use the profile characteristic to refine target groups for information campaigns, eligibility review, or monitoring.
4Ps Beneficiary Housing and living conditions 9 models Appears in 9 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. Household conditions or information access may be linked with how citizens know about and use LGU services. Check whether household access conditions affect service reach, communication channels, or ability to avail services.
Relationship to HH Head Socio-demographic profile 9 models Appears in 9 indicator-level Decision Tree model(s). 3 stages Covers 3 CSIS stage(s): Awareness, Availment, Satisfaction. This respondent profile characteristic may help explain differences in awareness, service use, satisfaction, or need for action. Use the profile characteristic to refine target groups for information campaigns, eligibility review, or monitoring.
HH Toilet Type Housing and living conditions 8 models Appears in 8 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. Household conditions or information access may be linked with how citizens know about and use LGU services. Check whether household access conditions affect service reach, communication channels, or ability to avail services.
House Ownership Housing and living conditions 7 models Appears in 7 indicator-level Decision Tree model(s). 3 stages Covers 3 CSIS stage(s): Awareness, Availment, Need for Action. Household conditions or information access may be linked with how citizens know about and use LGU services. Check whether household access conditions affect service reach, communication channels, or ability to avail services.
Place of Work Socio-demographic profile 6 models Appears in 6 indicator-level Decision Tree model(s). 4 stages Covers 4 CSIS stage(s): Awareness, Availment, Satisfaction, Need for Action. This respondent profile characteristic may help explain differences in awareness, service use, satisfaction, or need for action. Use the profile characteristic to refine target groups for information campaigns, eligibility review, or monitoring.
MCA DimQuadrant Low Dim1 / High Dim2 MCA profile dimension 4 models Appears in 4 indicator-level Decision Tree model(s). 3 stages Covers 3 CSIS stage(s): Availment, Satisfaction, Need for Action. This is a combined profile signal from MCA. It should be read as a summary of related respondent characteristics, not as one survey question. Review the underlying profile traits behind this MCA signal before translating it into a specific program action.
Civil Status Socio-demographic profile 4 models Appears in 4 indicator-level Decision Tree model(s). 2 stages Covers 2 CSIS stage(s): Availment, Satisfaction. This respondent profile characteristic may help explain differences in awareness, service use, satisfaction, or need for action. Use the profile characteristic to refine target groups for information campaigns, eligibility review, or monitoring.
A recurring predictor is decision-support evidence, not proof of causation. Use it together with the indicator-level metrics, decision-tree example, clustering results, and service delivery review.
Association-Rule Evidence

Repeated Service-Stage Patterns

6 rules
Association-rule evidence shows which CSIS service-stage responses commonly appear together. Use these repeated patterns to decide which service-stage link needs validation by the LGU.
Patterns Reviewed 6
Top Result Area Satisfaction
Top Confidence 96.7%
Top Lift 1.09
Rule Result Area Repeated Pattern Support Confidence Lift Meaning Interpretive Relevance
Rule 1 Satisfaction Citizens who were aware of the social welfare services, and availed the social welfare services had a 97% likelihood of also were satisfied with the social welfare services (Lift: 1.09). 27.1% seen in eligible records 96.7% observed together 1.09 Strong This rule is about citizen experience after service use. The linked response is observed together 96.7% of the time and appears more often than expected by chance. Use this pattern to review service quality, frontline experience, and satisfaction drivers for Social Welfare Services.
Rule 2 Satisfaction Citizens who were aware of the social welfare services, and felt a need for further action, and availed the social welfare services had a 94% likelihood of also were satisfied with the social welfare services (Lift: 1.06). 12.6% seen in eligible records 94.3% observed together 1.06 Strong This rule is about citizen experience after service use. The linked response is observed together 94.3% of the time and appears more often than expected by chance. Use this pattern to review service quality, frontline experience, and satisfaction drivers for Social Welfare Services.
Rule 3 Satisfaction Citizens who availed the social welfare services had a 92% likelihood of also were satisfied with the social welfare services (Lift: 1.04). 35.6% seen in eligible records 91.9% observed together 1.04 Strong This rule is about citizen experience after service use. The linked response is observed together 91.9% of the time and appears close to the ordinary baseline rate. Use this pattern to review service quality, frontline experience, and satisfaction drivers for Social Welfare Services.
Rule 4 Satisfaction Citizens who felt a need for further action, and availed the social welfare services had a 89% likelihood of also were satisfied with the social welfare services (Lift: 1.00). 16.3% seen in eligible records 89.0% observed together 1.00 Strong This rule is about citizen experience after service use. The linked response is observed together 89.0% of the time and appears close to the ordinary baseline rate. Use this pattern to review service quality, frontline experience, and satisfaction drivers for Social Welfare Services.
Rule 5 Awareness Citizens who availed the social welfare services had a 70% likelihood of also were aware of the social welfare services, and were satisfied with the social welfare services (Lift: 0.97). 27.1% seen in eligible records 69.9% observed together 0.97 Weak This rule is about awareness and information reach. The linked response is observed together 69.9% of the time and appears close to the ordinary baseline rate. Use this pattern to validate where Social Welfare Services information campaigns or barangay communication may need strengthening.
Rule 6 Awareness Citizens who felt a need for further action, and availed the social welfare services had a 69% likelihood of also were aware of the social welfare services, and were satisfied with the social welfare services (Lift: 0.95). 12.6% seen in eligible records 68.6% observed together 0.95 Weak This rule is about awareness and information reach. The linked response is observed together 68.6% of the time and appears weaker than the ordinary baseline rate. Use this pattern to validate where Social Welfare Services information campaigns or barangay communication may need strengthening.
Association rules show co-occurrence, not causation. A high-confidence rule should still be checked against service records, citizen feedback, and field context before implementation.
Next Step: Decision Support

After examining and validating the harmonized evidence, continue to the Decision Support module to review planning-oriented recommendations and their traceability paths.

Open Decision Support
Showing 4500 of 4500 respondent records for All Mindanao records.