Methodology & Data Principles
FoodAlt is an educational food decision tool built on transparent, reproducible nutritional mathematics. Here is how we source, calculate, and present food data.
1. Authoritative Data Sources
Every food record in FoodAlt is mapped directly to authoritative, peer-reviewed, and published food composition databases. Our primary sources include:
- USDA FoodData Central (Foundation Foods, SR Legacy, and Survey Databases)
- Regional Agricultural and Nutritional Tables for traditional crops and regional staples
- Peer-reviewed Food Chemistry Literature for verified composite botanical analyses
Each food page displays its primary source repository, dataset type, and unique record ID (e.g. USDA FDC ID) with a direct link to the underlying source record.
2. Food States & Preparation Accuracy
Nutrient concentrations change significantly depending on preparation. For example, cooking reduces water content, concentrating minerals while potentially degrading heat-sensitive vitamins (such as Vitamin C and Folate).
FoodAlt explicitly labels the preparation state of every record (Raw, Cooked, Boiled, Roasted, or Dried). When comparing foods with differing preparation states, the platform displays an explicit mismatch notice.
3. The Four Comparison Modes
A. 100 g Standard Edible Portion
The international scientific baseline for food chemistry. Comparing per 100 grams removes portion bias and reveals true chemical concentration.
B. Typical Serving Size
Reference serving portions established by dietary guidelines or common culinary consumption (e.g. 1 medium fruit, 1 cup cooked, or 1 tablespoon).
C. 100 kcal Nutrient Density
Calculated as: Amount × (100 / Calories per 100g). This answers: "How much nutrient do I get per unit of food energy?" Essential for identifying nutrient-dense, calorie-efficient foods.
D. 300 kcal Same-Energy Normalization
Normalizes two foods to identical total caloric intake (300 kcal).
Disclaimer: 300 kcal is a mathematical comparison basis, not a recommended dietary serving.
4. Missing Data vs. Verified Zero
A critical flaw in generic nutrition websites is treating unanalyzed nutrients as zero. In FoodAlt:
- 0 / 0.0 mg: Indicates the nutrient was analyzed and found to be absent or below detection thresholds.
- Not available: Indicates the specific laboratory assay did not measure this nutrient in that source record. We never default missing data to zero.
5. Alternative Ranking Algorithm
When you search for an alternative (e.g., "more Vitamin C than Orange"), our multi-factor scoring algorithm evaluates:
- Absolute Nutrient Yield: Total content of target nutrient.
- Nutrient Density: Nutrient delivered per calorie.
- Calorie Efficiency: Penalizing excessive calories unless higher energy is desired.
- Category Proximity: Ranking sensible culinary substitutions within or adjacent to the baseline food's category.
To prevent deceptive math, we apply a small baseline floor: percentage differences are suppressed when baseline amounts are negligible, avoiding misleading claims.
6. Data Verification & Corrections
We continuously audit our food records against updated USDA releases. If you notice any discrepancy between a FoodAlt value and an authoritative laboratory record, please submit a report on our dedicated corrections page.