Quantifying the Dynamics: How Intersecting Conditions Shape Household Sensitivity

In one of our previous blogs, The Rurbanise Reality, the team shared the “lived realities” of mothers navigating muddy roads and coastal communities building "fragile frontlines" against the tide. Those Focus Group Discussions (FGDs) provided the "human face" of climate change.

But how can those powerful stories turn into actionable data?

Project RURBANISE spent months building the next phase: the Household (HH) Survey. The household survey was designed to investigate and characterize the different conditions and factors (intersectionality) that drive how vulnerability differs (differential vulnerability) within informal households exposed to climate-related hazards.

The household surveys were led by the University of the Philippines School of Urban and Regional Planning (SURP) through its research arm Planning and Research Development Foundation, Inc. (PLANADES). HPFPI-PACSII performed a critical role in managing logistics, scheduling, and high-level coordination with local leadership, particularly across regional jurisdictions.  To gather feedback on the process, TAMPEI conducted systematic check-ins throughout the survey process, focusing on the technical nuances of survey preparation, the integrity of data collection, and structured debriefing sessions with all participants. 

The survey sought to answer the following questions:

1. How vulnerable are households (in informal communities) to climate-related hazards, specifically weather-related events like flooding or severe winds from tropical cyclones?

2. How do households differ in terms of levels of vulnerability? 

3. What social, economic, and environmental conditions shape these differences? 

4. Which intersecting conditions shape specific levels of vulnerability? 

5. How do male and female views on these vulnerability conditions differ?

Building the Blueprint

The Differential Vulnerability Framework (as shown below) and its operational definitions served as the blueprint for selecting our survey indicators. Here is how that conceptual model was refined into a functional research tool: 

• Categorizing vulnerability components: Distinct indicator categories  were established — including exposed units, social identities, tenure/informality — to measure each core component of vulnerability (exposure, sensitivity, and adaptive capacity). 

• Clustering evidence-based measures: These categories guided the final selection and clustering of specific measures, which were distilled directly from Focus Group Discussions (FGDs) and technical contributions from partners.

• Integrating intersectional depth: To capture the gendered realities of climate change in urban poor settings, additional indicators were integrated based on intersectional markers. Individual-level characteristics—such as age, gender, and specific social circumstances—were also included to better describe the multiple identities that drive vulnerability. 

  
  

Expanding hazard scope: Hazard indicators were adjusted to accommodate all nine Homeowners Associations (HOAs). This ensures that data reflects a total community picture, moving beyond a singular focus on flood-prone areas to include the diverse hazards that communities face.

The survey structure and content was composed of four sections: 1) Exposure, 2) Potential Impact, 3) Sensitivity, and 4) Adaptive Capacity. Each of these sections represents a dimension of differential vulnerability. Each section encompasses thematic categories of variables, which have been operationalized into specific survey items through targeted questions, indicators, and response scales. The core terms and concepts within these items are organized alphabetically by theme and have been refined based on insights from the pre-test and technical reviews by the enumerators.

Transitioning from Design to Deployment

In August 2024, representatives from each HOA participated in an online pre-test. This allowed community members to flag concerns regarding survey length, complexity, and the presence of "sensitive" personal questions, ensuring the final tool was culturally appropriate and ethical. 

Following feedback from community partners, the tool was updated with clarified technical terminologies and sent for digital conversion. By using Kobo Toolbox, the survey was transformed into a mobile-ready format for efficient field data collection.

Despite a devastating fire in March 2024, the ASHAI community still expressed interest in participating in the survey. A sex-disaggregated list of 724 onsite members provided by HPFPI was utilized to ensure that the sampling design remained inclusive and representative. 

Utilizing the Kobo Toolbox meant involving a software specialist. In September 2024, the first version of the digital tool was demonstrated at an online conference for an open discussion. PLANADES conducted an intensive training program for 15 community-based enumerators and eight (8) field coordinators. The curriculum moved beyond basic interview protocols to include the logical structure and content of the questionnaire and video materials presenting the critical role of gender in climate adaptation. 

The surveys were planned to be administered through the use of digital format (browser-based, tablets and laptops), via interviews with audio recording, and pre-identified interview site/s per community.

Training of Survey Enumerators, September 2024

Between September and November 2024, the digital tool underwent rigorous, iterative updates based on direct field feedback from enumerators, ensuring the survey questions were optimized for local clarity and technical accuracy.  Furthermore, the tool was enhanced by incorporating technical inputs and logistical guidelines, such as protocols for the utilization of external storage. A printed version of the tool was transmitted to TAMPEI for comprehensive translation into Bisaya and Filipino.

Survey preparatory meeting with community enumerators

Following the formal communication from TAMPEI on November 21, 2024, the final study population was adjusted from eight to seven Homeowners Associations (HOAs) after APSHAI elected to withdraw due to organizational internal conflicts. Prior to this finalization, PLANADES issued a final call for the validation of the October 2024 master lists. With the exception of ULHOA, all participating HOAs confirmed or revised their listings by November 15 to establish the final master list. Despite these adjustments, PLANADES successfully maintained the statistically significant minimum sample size of 384. To ensure the robustness of the data collection phase, the master list incorporates a 20% buffer, allowing for the systematic selection of replacement respondents should any primary members decline to participate. 

With the training successfully completed and the digital tool and master list finalized, the project moved into the actual survey phase, scheduled from December 2024 to January 2025.

Survey proper at SMASH community

📝 The Survey Team The successful rollout of the household survey relies on a synchronized effort between technical experts and community leaders. Under the guidance of PLANADES-SURP, the survey team was structured into three distinct roles to ensure data integrity and community trust:
Role	Led By	Core Responsibilities
Lead	PLANADES-SURP	Study design, tool development, training, data analysis, results validation and research report, operational guidance and liquidation
Overall Technical Support	PLANADES- SURP and Consultant	Digital tool preparation
Administrative and Finance Support	TAMPEI and HPFPI RO, with support from PLANADES	Administrative needs, budgeting
Technical Field Coordinators (TFC)	TAMPEI	Technical oversight, digital tool troubleshooting, interview scheduling, and data validation
Community Field Coordinators (CFC)	HPFPI & HOA Volunteers	Onsite logistics, respondent verification, and conducting Informed Consent briefings
Enumerators	Trained Local Residents	Primary data collection via the Kobo Toolbox, interview rehearsals, and field reporting

Collecting and Processing Data

During the data collection phase, the project achieved a 99% response rate, with 382 out of 384 target respondents participating; of these, 64% were female and 36% were male. Following the successful deployment, PLANADES generated a partial summary of findings, which was presented to partners (TAMPEI, ESSC, and UoS) on March 26, pending final verification of consent forms and respondent identities.

To facilitate remote monitoring and real-time technical support, dedicated online platforms were established for each HOA via Messenger chat group. These channels allowed for immediate guidance on tool navigation, protocol adherence, and the systematic replacement of unavailable respondents. Mid-course feedback sessions, particularly with first-time enumerators from ULHOA, identified challenges regarding interview duration and linguistic barriers within the Tagalog version of the tool. To optimize tool operability and respondent comprehension, the enumerators were authorized to use mixed dialect (Tag-lish). They were also encouraged to leverage the Technical Reference Guide along the show cards to clarify complex terminologies and facilitate more efficient responses. 

Establishing the Variables

To map out the communities’ vulnerabilities with statistical precision, Project RURBANISE’s research arm, PLANADES, employed advanced econometric modeling—including LASSO regularization, polychoric principal component analysis (PCA), and Stepwise Robust regression. 

Here is a quick guide to the significant composite indices and single measures used to evaluate each household’s adaptive capacity, readiness, and recovery potential when subjected to environmental hazards:

Index	Description
Potential Impact Index	Calculates the expected degree of disruptions a household faces during a disaster, tracking past experiences with house damage, power outages, and blocked roads. The higher the score, the more severe the expected cascade of disruption.
Exposed Service Facilities Index	Identifies potential points of failure for the spatial exposure of critical service facilities (like clinics, schools, and markets) during extreme weather events.
Environmental Risk Perception Index	Aggregates community perceptions of localized stressors—such as pollution, ecological degradation, and geophysical risks—that act as threat multipliers.
Autonomous Adaptation Index	Measures a household's self-initiated strategies on social memory, risk perception, and emergency readiness across 28 different behaviors, from emergency savings to structural home reinforcement.
Household Female Intensity	Tracks the gendered demographic balance across key age brackets (0–4, 25–48, 60–65, and 65+) to capture caregiving burdens and dependency pressures. This index refers to the proportion of female household members. A higher presence of working-age women (ages 25–48) often highlights strong income-earning potential or vital caregiving capacity within the home. Conversely, a high ratio of elderly women signals higher dependency pressures during a crisis.
Social Assistance and Insurance	Tracks access to social assistance for the elderly and financial support from HOAs or civil society organizations as determinants of a household’s recovery potential and adaptive capacity. This tracks the safety nets that help families cushion the financial blow of a crisis.
Basic Amenities and Employment in the Public Sector	To account for differences in household infrastructure and employment stability, two structural controls are added: HH with Own Electrical Connection: A quick proxy for basic housing quality and structural security.  HH member employed in the government: A marker indicating relative job security and stronger institutional ties.

Analyzing the Survey Data

Beyond the numbers and technical data, the survey provides eye-opening information on the condition of the households and the communities:

The Paradox in Proximity

Standard development narratives suggest that when critical facilities like schools, medical clinics, barangay halls, and markets are placed close to households, a community becomes inherently more functional and secure. However, the findings from the Exposed Service Facilities Index disrupts this assumption. Although proximity undeniably streamlines day-to-day access to services, embedding these critical facilities within the same hazard-prone areas as the residential zones creates an unintended vulnerability loop. When climate hazards hit, their shared vulnerability triggers a rapid cascade of risk that simultaneously paralyzes infrastructure and compromises the safety of local service users. 

The Accuracy of Anxiety

The survey reveals that risk awareness is deeply grounded in experience, not imagination. As the Environmental Risk Perception Index increases, a household's potential disaster impact climbs significantly. This statistical association shows that when residents express high concern over co-existing environmental threats, they are accurately reflecting their physical surroundings. Their awareness is a direct byproduct of lived vulnerability, shaped by past disruptions and the objective realities of living in high-exposure zones.   

The Severity of Self-Help

Another significant finding from the survey presents the strong positive link between the Autonomous Adaptation Index and the Potential Impact Index. The analysis indicates that autonomous adaptation within informal settlements functions primarily as a reactive mechanism rather than a marker of preventive readiness. In plain terms, families who frequently engage in self-driven adaptive practices— making minor home repairs, seeking alternative livelihoods during crises, stocking up on basic supplies, or organizing informal neighborhood coping mechanisms—are almost always the ones already facing the highest physical exposure. It emphasizes a key takeaway: without large-scale, systemic infrastructure and institutional support, localized coping mechanisms remain uneven, exhausting safety nets that the urban poor are forced to build just to survive. 

The Interplay of Inequality

The survey data reveals that a household's demographic composition plays a decisive role in its disaster vulnerability, specifically tracking how age and gender interact. Households with a higher proportion of working-age women (ages 25–48) or elderly women (ages 60 and above) face a significantly higher potential impact during crises. For middle-aged women, this vulnerability stems from a heavy burden of simultaneously balancing income-earning roles and intensive family caregiving during crises. For older women, the risk is driven by long-term income insecurity and health-related dependence.

Interestingly, the data revealed a unique contrast: households with a higher proportion of young girls (ages 0–4) actually showed a decrease in potential disaster impact. This suggests that families with infants and toddlers often trigger stronger kinship networks, receiving more external community support and shared carework during a crisis. Ultimately, gendered vulnerability is not uniform. It intensifies with age and the weight of caregiving responsibilities, reflecting the deeply layered nature of household sensitivity.

The Empowerment of Energy

The survey confirmed that energy infrastructure acts as a powerful shield. Households with their own legal, metered electrical connections recorded a significant decrease in potential disaster impacts. This proves that energy autonomy functions as a vital buffer against systemic disruptions, granting families a measure of adaptive independence. Having a reliable, self-managed energy source doesn't just provide physical security; it offers the psychological assurance of staying in control when a disaster strikes, whereas any disruption to power reduces a household's functionality.

The Trap in Tenure 

Households with at least one government-employed member actually showed a higher potential disaster impact. While public sector work suggests a steady, formal income, the reality is government wages remain insufficient to cover baseline living expenses, let alone emergency costs for informal settlers. Furthermore, it shows mobility constraints as workers frequently endure costly, rigid commutes from densely populated, high-exposure areas, worsening their vulnerability during a crisis. 

The Asymmetry of Assistance

Contributory systems like SSS and GSIS stabilize households through steady income replacement, effectively buffering them against moderate hazard exposure and economic shocks. Conversely, needs-based assistance—such as PhilHealth, Pag-IBIG, or solo-parent aid—successfully identifies and targets structurally fragile segments, but lacks the resource depth to protect them from the compounding, intersecting crises that follow a disaster. 

The Protection in Partnerships

The survey data reveals that active membership in a Homeowners’ Association (HOA) statistically lowers a household’s potential disaster impact. While HOAs typically lack the financial capital to distribute direct cash assistance, their value lies in localized, in-kind support and collective mobilization. This proves that organized community solidarity acts as a powerful safety net, building genuine household resilience when an individual household’s adaptive capacity is insufficient.

Navigating Layered Vulnerabilities

Ultimately, the survey findings paint a clear picture: vulnerability in informal settlements is not a single, isolated problem, but a deeply layered structure. Households that are exposed to fragile critical infrastructure, driven by high caregiving demands from middle-aged or elderly women, and reliant on limited social safety nets inevitably face the highest risk of disaster impacts.

Conversely, families that leverage proactive, self-initiated risk management practices and maintain active linkages within their communities display much lower sensitivity when a crisis hits.

This leads to a pivotal insight for Project RURBANISE: rurban resilience depends far less on raw income alone, and much more on a household's specific position within overlapping systems of physical exposure, local self-help, and formal institutional support.

—

Research supported by CLARE is bridging critical gaps between science and action: developing new tools and supporting partner governments, communities, and the private sector to use evidence and innovation to drive effective solutions to the climate challenge, whilst building capacity of both those carrying out the research and those using the resulting evidence.

Through long-term commitments and partnerships worldwide, and needs-driven, action-focused research, CLARE links up short-and long-term issues, enabling long-term, sustainable, and fair economic and social development in a changing climate whilst supporting early action to reduce impacts of climate variability whilst providing a better understanding of the risks associated with climate.

---

About the Author

Janina Salubo is a Knowledge Management and Development Communications Volunteer for the RURBANISE project, specializing in translating complex resilience research into accessible insights for funders and grassroots communities.