Methodology

EWG’s Shopper’s Guide to Pesticides in Produce™ ranks popular fresh fruits and vegetables based on levels of pesticide contamination and pesticide toxicity. 

To create the guide, EWG analyzes pesticide residue data for tens of thousands of samples that the Department of Agriculture collects through its Pesticide Data Program. EWG then adds pesticide toxicity information published by government agencies.

Selection of sampling years and sample preparation

Rather than test all crops each year, the USDA selects a subset of fruits and vegetables to test for pesticide residues. The USDA typically releases the results of sampling efforts two years after samples have been collected. 

In 2025, for instance, new data were available for 2023 collection of apples, avocados, blackberries, celery, grapes, mushrooms, onions, plums, potatoes, sweet corn, tomatillos, tomatoes and watermelon.

This year’s guide incorporates data from a total of more than 53,000 samples of 47 fruits and vegetables. EWG uses USDA data for non-organic samples of fruits and vegetables from the most recent sampling periods, which typically spans one to two years for each item. 

For example, to analyze residues on spinach, we used 1,295 samples the USDA collected between 2015 and 2016, as that’s the most recent data range for that type of produce. But for apples, we used only the 334 samples published for the most recent sampling period, which began in October 2023, and will continue through 2024, though only data for 2023 samples are currently available.

Washing and preparing samples

Before the USDA analyzes samples for hundreds of pesticide residues, it prepares each piece of fruit or vegetable in the same way many people do at home. This includes washing samples under running water for about 15-20 seconds.

Produce with edible peels, like apples, are rinsed under cold water and drained before they are tested. Items with inedible peels, like bananas, are peeled. Produce with inedible ends is trimmed then about three to five pounds of a type of produce are blended together to create a sample. The sample then provides a good indication of consumers’ exposure to pesticides.

Details about how the USDA prepares each sample before processing can be found on the Pesticide Data Program website. The list of produce items included, sampling years selected, and numbers of samples analyzed can be found here. 

Determining fruit and vegetable rankings

Starting in 2025, EWG refined our methodology to consider four measures of pesticide contamination. Each of these factors captures unique information about how consumers may be exposed to pesticide residues on foods:

• Percent of samples with at least one pesticide detected.
• Average number of different pesticides detected on a single sample.
• Average total concentration of pesticides found on a single sample.
• Overall toxicity of pesticides on a crop.

The metrics convey unique but equally relevant information about pesticide levels on produce respectively: abundance, diversity, intensity and toxicity.

EWG calculated the raw values for the four factors above for each of the 47 fruits and vegetables and ranked them. We then normalized, or standardized, the raw values so that they could be combined into a single indicator of overall pesticide load. 

The values were normalized on a scale of 1 to 100 with 100 representing the highest calculated value for each metric. For each type of produce, the normalized values were added together to calculate a total score, out of 400, since there were four metrics.

The Shopper’s Guide full list shows how all 47 fruits and vegetables ranked, from least to most pesticide contamination, based on these total scores. The 15 with the lowest total scores are considered the Clean Fifteen™, while the 12 with the highest total scores are classified into our Dirty Dozen™ list. 

Calculating the toxicity of pesticides on a crop

Overall toxicity for pesticides is determined by comparing the concentration of a pesticide detected to a value that quantifies the toxicity of that pesticide, or a pesticide-toxicity ratio. 

First, EWG summed all concentrations for all metabolites or degradates of a known parent pesticide across all samples for each produce type. These values are reported in parts per million or milligrams of a parent pesticide per kilogram of fruit or vegetable. Next, for each type of produce, pesticide to toxicity ratio values were calculated by dividing the average summed concentration values, by a toxicity reference value for the parent pesticide.

EWG added the toxicity ratio values into a total value that represents the overall toxicity of all pesticides on a crop.

For each parent pesticide, EWG used the corresponding no-observed-adverse-effect-level, or the NOAEL, in milligrams of a given parent pesticide per kilogram body weight of the exposed animal as the toxicity reference value. The NOAEL is the highest dose, or pesticide exposure level, at which no harmful effects were observed following exposure to a given pesticide in animal studies.

To select the NOAELS, EWG primarily identified the values used by the Environmental Protection Agency when it reviews toxicity information for each pesticide when it is registered for use in the U.S. These values were obtained through a Freedom of Information Act submitted by the Center for Biological Diversity. No easily accessible, comprehensive database of toxicological reference values for pesticides exists.

If EWG could not identify a suitable value, other government agency risk assessment documents within either the EPA or the European Food Safety Authority were reviewed. Using the NOAELs provides a consistent source for a toxicological reference value. A full list of toxicity values and sources can be found here. 

More on EWG’s approach 

Hundreds of individual types of pesticide residues are found on produce.

But regulatory agencies evaluate pesticide toxicity one at a time when assessing them to establish allowable limits of pesticide residues in foods.

Many pesticides have not been well studied by independent scientists, and there is little to no understanding of their potential effects on human health.

Some pesticides are, however, known to be more toxic than others. For this reason, EWG incorporates metrics based on pesticide toxicity, as well as accounting for all pesticides measured on commodities. The scores for each fruit and vegetable represent the overall pesticide burden. EWG designed and updated this method to help consumers reduce their overall pesticide load.