Research Glossary
Key terms and definitions from applied economics research on food security, transit equity, education policy, and causal inference methods.
Food Security
Food Desert
A low-income census tract where at least 500 people or 33% of the population live more than 1 mile from a supermarket (urban) or 10 miles from a supermarket (rural), as defined by the USDA Economic Research Service.
Analysis of 6,613 grocery stores found zero food deserts in Santa Clara County by this federal definition, yet SNAP participation varied 4.7x across similar communities, suggesting the food desert framework may be insufficient for understanding food insecurity in metropolitan areas.
Related: The Food Desert Myth, Mobility Desert, Vulnerability Paradox
Mobility Desert
A neighborhood where grocery stores exist within 1 mile but transit access is inadequate (no stop within 0.5 miles or fewer than 2 stops nearby), making stores practically unreachable for residents without cars.
12% of California census tracts (1,086 neighborhoods) are mobility deserts, representing approximately 150,000-250,000 transit-dependent residents. Los Angeles metro (412 tracts) and San Diego (216 tracts) account for 58% of all mobility desert tracts statewide.
Related: Hidden Mobility Deserts, Food Desert
Vulnerability Paradox
The observed pattern where communities with better geographic access to grocery stores show worse food security outcomes, suggesting economic barriers outweigh proximity advantages.
Majority-minority tracts in California average 1.03 miles to grocery stores versus 1.78 miles for majority-white tracts, with half the food desert rate (21.9% vs 39.9%), yet show higher vulnerability scores (0.337 vs 0.279) due to poverty and housing instability.
Related: Food Access Vulnerability Paradox, Extended Vulnerability Index
SNAP Enrollment Gap
The difference between the number of people eligible for the Supplemental Nutrition Assistance Program and those actually enrolled, reflecting administrative barriers, stigma, or information gaps.
In Santa Clara County, SNAP participation rates range from 12% to 56% of eligible households across census tracts with similar poverty rates and demographics, a 4.7x variation that cannot be explained by distance to grocery stores alone.
Related: The Food Desert Myth, Diverging Trajectories
Extended Vulnerability Index
A validated composite measure combining economic (55%), geographic (20%), mobility (15%), and density (10%) factors to predict food security risk at the census tract level.
Tested across 408 Santa Clara County tracts, the index shows strong criterion validity (r = 0.83 with SNAP participation) and robust rankings across five alternative weighting schemes (all Spearman r > 0.91).
Related: Extended Vulnerability Index, Vulnerability Paradox
Transit Equity
Distance-Transit Paradox
The pattern where majority-minority neighborhoods have closer grocery stores but worse transit access, meaning geographic advantage does not translate to accessibility advantage.
Majority-minority tracts in California average 0.72 miles to grocery stores versus 0.85 miles for majority-white tracts, yet have higher mobility desert rates (13.2% vs 11.1%). The 1.8 percentage point disparity persists after controlling for income and density.
Related: Transit Equity and Race, Mobility Desert
Cal-ITP (California Integrated Travel Project)
A statewide GTFS data aggregation maintained by Caltrans that compiles transit data from 200+ California agencies via data.ca.gov, providing comprehensive coverage for transit accessibility research.
Using Cal-ITP data (64,060 stops from 200+ agencies) corrected the mobility desert rate from 49% to 12%, compared to incomplete data from 8 major agencies (24,421 stops) that captured only 38% of California's actual transit infrastructure.
Related: Data Quality: 49% to 12%
Education Policy
Maintenance of Effort (MOE)
A federal requirement under IDEA that school districts must spend at least as much on special education as the previous year to continue receiving federal funding, creating a legal floor that prevents spending reductions.
During the Great Recession, MOE requirements prevented California districts from cutting special education, forcing general education to absorb all budget reductions. Special education's budget share rose from 16.7% to 21.3% (2008-2012) mechanically, not intentionally.
Encroachment
In education finance, the practice of diverting general education funds to cover special education costs when special education expenses exceed designated special education revenue.
Encroachment averages 3-4% of general fund budgets in California, but 7% of districts experience encroachment exceeding 10% of general education resources, with very high-impact districts averaging 13.2%.
Related: The Unfunded Mandate, Maintenance of Effort
Fiscal Cliff (Education)
The sudden budget shortfall experienced by school districts when temporary federal stimulus funding (ARRA) ended in 2011, producing a steeper decline than the recession itself.
California K-12 funding dropped 8.7% in a single year when ARRA funds ended (from $53.8 billion to $49.1 billion), compared to the 7.2% boost the stimulus had initially provided.
Econometric Methods
Event Study
An econometric method that estimates the dynamic effect of a treatment or policy change by comparing outcomes at multiple time points before and after the intervention, testing for parallel pre-trends.
Event study designs allow researchers to visualize treatment effects over time and test whether treated and control groups were on similar trajectories before the policy change, a key assumption of causal inference.
Related: Difference-in-Differences, Parallel Trends Sensitivity
Difference-in-Differences (DiD)
A quasi-experimental method that estimates causal effects by comparing changes over time between a treatment group and a control group, removing time-invariant confounders through double differencing.
Used in the COVID differential impact analysis to estimate that high-vulnerability tracts diverged by +2.35 percentage points (p = 0.012) more than expected based on other tracts' trends.
Related: Event Study, COVID Differential Impact
Synthetic Control Method
A statistical method that constructs a weighted combination of untreated units to serve as a counterfactual for a single treated unit, enabling causal inference when only one unit receives treatment.
Particularly useful in policy evaluation where a single jurisdiction implements a policy change and researchers need to estimate what would have happened without the intervention.
Related: Difference-in-Differences, County Comparison Methods
Get Research Updates
New analysis delivered to your inbox. No spam, unsubscribe anytime.
Subscribe for Free