Comprehensive Exam — Thesis Proposal
Sensors reduce downtime detection, but does monitoring alone improve outcomes? Or only when embedded in responsive institutions?
CBM critiqued extensively. Hybrid professional-community models show promise but lack rigorous comparative evaluation in ASAL contexts.
Uptime gets attention; microbial quality at POU remains understudied. Relationship between O&M programs and water quality is poorly understood.
Validated 12-item cross-cultural instrument, but few studies link HWISE outcomes to specific programmatic interventions in ASAL settings.
DCE and best-worst scaling rarely applied in N. Kenya, where public rural water companies introduce new governance dynamics.
Performance-based financing is emerging but empirical evidence on effectiveness in ASAL contexts is limited.
Drought Resilience Impact Platform — Functionality, Uptime, Network Diagnostics, and Intelligence
200 boreholes · Sep 2023 – Mar 2025 · Layered with RAPID+ and STAWI Mashinani
60 boreholes · May – Sep 2025 · Continued RAPID+ integration
Partial effectiveness → ideal case for studying determinants of success and failure
DRIP FUNDI's integrated model will produce statistically significant improvements: >92% uptime, ≥15% increase in dry-season water access (HWISE), and ≤10 CFU/100mL E. coli at intervention vs. non-intervention sites.
Effectiveness is primarily driven by trusted management structures, performance-linked financing, and responsive O&M with water treatment. Households prefer local WMC over public water companies at equivalent service levels.
A consistent suite of determinants will predict sustained water security across climate-stressed contexts, yielding transferable design and policy principles for ASAL regions.
Descriptive cross-sectional quantitative design with mixed-methods qualitative component, supported by historical data analysis (2017–2026).
| Level | Instrument | Method | Sample | Duration |
|---|---|---|---|---|
| Household | Survey + HWISE + DCE | Quantitative | 300 HH | 25 min |
| Operator | Semi-structured KII | Qualitative | Each borehole | 15–20 min |
| Community | Focus group discussion | Qualitative | 6–10/group | 20–25 min |
| Committee | Focus group discussion | Qualitative | 6–10 WMC | 20–25 min |
| Institutional | Semi-structured KII | Qualitative | County/NGO | 15–20 min |
| Comparative | KII (ANTHC, Alaska) | Qualitative | ANTHC reps | — |
| Infrastructure | Observation checklist | Quantitative | Each borehole | — |
| Water Quality | Aquagenix CBT | Quantitative | All boreholes | — |
| Historical | Virridy sensor data | Quantitative | 2017–2026 | N/A |
Structured 25-minute survey administered via mWater on handheld devices in local languages.
Worry, interruption, clothes, plans, food, hands, body, drink, angry, sleep, none, shame. Scored 0–36 (4-point frequency, 1-month recall).
Local WMC (same tariff) vs. public water company (doubled tariff) at equivalent service. Tests governance preference and WTP for management change.
Repair organization, financing models, monitoring, what limits sustained functionality
Breakdown causes, repair chain, delay factors (parts, technicians, funding, security, climate)
Indigenous governance, cultural dimensions, climate resilience in Arctic communities
Herders & farmers, 6–10/group, gender-segregated. Reliability trust, quality trust, WTP, program perceptions.
No mention of DRIP FUNDI unless respondents raise it (bias reduction)
6–10 WMC members. Step-by-step repair process mapping, financing adequacy, management challenges.
Composite 0–1 index from z-score normalized uptime, HWISE, and microbial safety with expert-validated weights. Internal consistency via Cronbach's α.
Community FGD guide prohibits mentioning DRIP FUNDI unless respondents raise it. Prevents leading questions and social desirability bias.
FGDs gender-segregated (male, female, youth). Same-gender facilitators and note-takers. Accommodates cultural and religious norms.
All instruments in Somali, Borana, Turkana, Rendile, Samburu. 20 locally recruited enumerators. Verbal consent in local language.
Quantitative HH data + qualitative FGD/KII + observation + water quality lab results + 9 years of historical sensor data.
| Phase | Period | Activities |
|---|---|---|
| IRB & Preparation | Apr–May 2026 | Ethics approval, instrument piloting, enumerator training |
| Data Collection (Wave 1) | Jun–Aug 2026 | HH surveys, KIIs, FGDs, water quality testing |
| Preliminary Analysis | Sep–Oct 2026 | Data cleaning, descriptive analysis |
| Data Collection (Wave 2) | Nov 2026–Jan 2027 | Second drought-season data collection |
| Full Analysis | Feb–May 2027 | Statistical analysis, qualitative coding |
| Synthesis (Study 3) | Jun–Aug 2027 | Literature synthesis, comparative analysis |
| Writing & Revision | Sep 2027–Feb 2028 | Dissertation writing, committee review |
| Defense | Spring 2028 | Dissertation defense |
| Category | Item | Cost (USD) |
|---|---|---|
| Salary | 20 enumerators (3 months) + research assistant | $20,000 |
| Equipment | Aquagenix test kits (500) + tablets for mWater | $8,000 |
| Supplies | Sample materials, printing, communication | $3,500 |
| Travel | 5 counties × 2 waves + conferences | $16,000 |
| Other | Software, transcription, translation | $4,500 |
| Total | $52,000 | |
Note: Virridy sensor infrastructure and DRIP FUNDI program costs covered by existing USAID BHA and Coca-Cola Foundation funding.
University of Colorado Boulder
PI: Karl Linden
Co-PI: Evan Thomas
Co-PI: Denis Muthike
Millennium Water Alliance
Program implementation
Field coordination
Community engagement
University of British Columbia (Canada)
Oslo School of Architecture (Norway)
Oxfam (South Sudan)
Lytton First Nation (Canada)
ANTHC (Alaska)
Funded by National Science Foundation, Virridy, and Millennium Water Alliance. Data not shared between institutions — results compared across sites.
Questions & Discussion