Our Food and Recipe Rating System


Why did we rate our foods and recipes?

We rated our foods and recipes to give you an extremely fast, simple, yet highly reliable way to meet your personal nutrition needs. Are you looking for foods that are richest in zinc? Or recipes that are highest in fiber? We wanted you be able to access any nutritional information of this kind with a click of your mouse button.

Where did we start?

We began with a computerized analysis of the nutritional contents of the World’s Healthiest Foods. In other words, we started with a food like carrots, and we analyzed how much vitamin C, vitamin A, zinc, protein, etc. that food contained in one commonly-eaten serving. We repeated this process until we analyzed 30 different nutrients for each of our 125 healthiest foods.

What kind of categories did we establish for ranking foods?

We picked a simple, three-category system for ranking all foods. The three categories we chose were “good,” “very good,” and “excellent.” We decided to develop a ranking process that would let us separate “good” food sources of a nutrient from “very good” sources, and “very good” sources of a nutrient from “excellent” ones. We assumed you would not want to know about the nutritional quality of a food if it didn’t rise up to at least the “good” level.

How did we decide which foods were “good,” “very good,” and “excellent” sources of nutrients?

We developed a two-part formula to determine how “good” a food was as a source of nutrients. The first (and most important) part of this formula looked at the nutrient density of each food. Nutrient density is a key idea in nutritional science. It compares the nutritional content of a food to its calorie content. If a food is very high in one or more nutrients, but very low in calories, it can be described as “nutrient dense” because it uses up very little of your day’s calories but is still “thick” with nutrients. If a food is low in nutrients but high in calories, it is the opposite of nutrient dense – we could describe it with the words “nutrient sparse” or “nutrient thin.” A “nutrient thin” food gives you very little in terms of nutrition, but uses up a lot of your day’s calories. Many foods fall somewhere in the middle and are considered “average” in terms of density. They give you a fair number of nutrients and use up a fair number calories; or they give you a lot of nutrients and use up a lot of calories. We didn’t think you would want to know about the “nutrient thin” or average foods, but only those foods that were nutrient dense and provided you with lots of nutrients while costing you few calories.

With the categories of “good,” “very good,” and “excellent,” in hand, and our thinking focused on nutrient density, we developed specific density criteria for each of our three categories. To qualify as a “good” source of nutrients, we decided that a food should have a density value of at least 1.5. In other words, that food had to give you 1.5 times as much nutrition as calories. Or to put it slightly differently, it had to give you 50% more in nutrition than it used up in calories.

Let’s take a real-life example. One-quarter cup is a commonly-eaten snack portion of nuts or seeds, so we took one quarter cup of sunflower seeds as our starting point for ranking this food. That quarter cup of sunflower seeds was determined to have about 200 calories (205 to be exact), and those 205 calories were about 11% of a 2,000-calorie daily diet. Were sunflower seeds a “good” source of any nutrient? We used our 1.5 criterion to help us decide. Since those 205 calories’ worth of sunflower seeds provided about 11% of a day’s calories, they had to provide at least 1.5 times that much of some nutrient to qualify as a “good” source of that nutrient. Did one-quarter cup of sunflower seeds supply at least 17% of the daily requirement for some nutrient? The answer was yes! While using up only 11% of the day’s calories, it provided over 20% of the daily value for folate and vitamin B5; and over 25% for phosphorous, tryptophan, copper, magnesium, and mangenese.

We moved our criterion up to 3.4 to help us decide if a food was a “very good” source of a nutrient. In other words, we said that whatever percentage of a day’s calories a food supplied, it had to supply 3.4 times that much in a nutrient in order for it to qualify as a “very good” source. To go back to our real-life example: our quarter cup of sunflower seeds still supplied 205 calories or about 11% of a day’s calories. But how did it do as a “very good” source of nutrients? Did it provide at least 3.4 times that much of any nutrient, i.e., at least 38% of the daily value? The answer once again was yes! One-quarter cup of sunflower seeds provided over 38% of a day’s vitamin B1 need.

Finally, to help us decide if a food could be called an “excellent” source of a nutrient, we moved our criterion up again from 3.4 to 7.6. Whatever percentage of a day’s calories a food supplied, we said it had to supply 7.6 times that much in a nutrient in order for it to be considered an “excellent source.” We knew that our quarter cup of sunflower seeds provided about 11% of a day’s calories? But did they provide 7.6 that much in any nutrient, i.e., 86% of the daily value? Once again, the answer was yes! Our quarter cup of sunflower seeds provided over 90% of the daily value for vitamin E, making sunflower seeds an “excellent” source of this nutrient.

For the purpose of providing you with a basic explanation of our Food and Recipe Rating System, we’ve used the term “Daily Value” to describe our standard of comparison when determining a food’s ranking on the “good,” “very good,” or “excellent” scale. At a much more detailed level of explanation, our standards of comparison would be described as follows:

Daily Values (DVs) were actually derived in a somewhat odd and inconsistent way by the US Food and Drug Administration (FDA) beginning in the 1990s. Some DVs were based on the Recommended Dietary Allowances (RDAs) issued in 1968. Other DVs were based on RDAs set forth in 1980 or 1989. Most, but not all DVs were based on the average RDAs for men and women combined. As you can see, there was no single framework used in the setting of the DVs.

In addition to this complexity involving the nature of the DVs themselves was the complexity associated with the database we used to obtain our standards of comparison. For this purpose, we chose one of the gold standards in the field of nutrition, namely, Food Processor Version 7.60, the food composition software developed by ESHA Research in Salem, Oregon (phone: 503-585-6242), database version December 2000. The research staff at ESHA Research in Salem, Oregon incorporated a mixture of widely-accepted nutritional standards when designing the Food Processor software. In many cases, the DVs were indeed used as standards of comparison for ranking the nutrient content of foods. In other cases, however, RDA values or AI (Adequate Intake) values were used instead of the DVs. In still other cases, where no RDA, or AI, or DV standards were available, the research staff at ESHA actually calculated their own standard of comparison based on a review of the nutritional research. Only a few standards were set through this last method, however. In short, our standards of comparison - while largely related to the DVs established by the FDA - also include RDA and AI standards, as well as a few standards devised by the research staff at ESHA Research

How did we compare foods to each other?

In order to compare foods to each other, we developed a formula for adding together all of the “good” and “very good” and “excellent” ratings that a food had received for its nutritional value. Our sunflower seeds, for example, turned out to be a “good” source of 8 nutrients, giving them 8 “goods.” In addition, sunflower seeds were a “very good” source of 1 nutrients, giving them 1 “very good.” Finally, sunflower seeds were an excellent source of 1 nutrient, giving them 1 “excellent.” In our formula, we awarded 1 point for each “good,” giving sunflower seeds 8 x 1 or 8 points for their “goods.” For “very goods” we awarded 2 points, since our density criterion of 3.4 for “very good” was roughly twice as high as our 1.5 density criterion for “good.” For sunflower seeds, this decision meant multiplying 1 “very good” by 2 points each, for a total of 2 additional points. Finally, we awarded 4 points for each ranking of “excellent,” since our density criterion of 7.6 for excellent was roughly twice our 3.4 density criterion for “very good.” This decision earned sunflower seeds an additional 4 points, since it achieved 1 ranking of “excellent” for its vitamin E content. When we added the “good” points (8) and “very good” points (2) and “excellent” points (4) together, we arrived at a total score for sunflower seeds of 14. We used this same procedure to arrive at a single nutritional score for each of our 125 foods.

Why turnip greens, mustard greens, and tofu are ranked as better sources of calcium than dairy products

For example, our food ranking system shows turnip greens, mustard greens, and tofu to be better sources of calcium than dairy products. We arrive at this conclusion since our ranking system looks not only at the amount of calcium in food but also at the caloric content of food and how many calories of a food are necessary to provide the desired amount of calcium. Dairy foods - which are often listed as excellent sources of calcium - turn out to be very good sources of calcium in our ranking system (just one step below turnip greens, mustard greens, and tofu) because of their lower nutrient density.

Should I only eat foods with the highest scores?

No! Even though foods with high scores are the most nutrient dense and give you more nutrients for fewer calories, all of the foods featured on this website have unique nutritional value. Many foods with lower scores can complement each other perfectly to round out your total nourishment. At the same time, however, if you find yourself eating exclusively at the lower end of the ranked foods, you may want to consider experimenting with recipes that can bring more highly ranked foods into your diet.


  • . Food Processor for Windows [Nutrition Analysis Software]. Version 7.60. Salem, OR: ESHA Research; 2000 Dec; c2000.
  • Erbersdobler HF. [Evaluation of sugar and sugar-containing foods as to their significance for energy and nutrient supply]. Z Ernahrungswiss 1990;29 Suppl 1:16-20.
  • Lowik MR, Westenbrink S, Hulshof KF, et al. Nutrition and aging: dietary intake of "apparently healthy" elderly (Dutch Nutrition Surveillance System). J Am Coll Nutr 1989 Aug;8(4):347-56.
  • O'Dea K. Cardiovascular disease risk factors in Australian aborigines. Clin Exp Pharmacol Physiol 1991 Feb;18(2):85-8.
  • Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 1986 Jul;124(1):17-27.

This page was updated on: 2004-12-02 17:21:27
© 2002 The George Mateljan Foundation