As concerns over climate change and resource depletion grow, comparing the environmental impact of various products has become something of a pastime: Paper towels or hot-air hand dryers? Paper or plastic grocery bags? Plane, train or automobile?

Environmental scientists can give pretty good answers to those questions using a technique known as life-cycle analysis, or LCA, which is an attempt to quantify a product’s effects on the planet, from greenhouse gas emissions to acid rain. Companies order LCAs to assess and trumpet their environmental credentials, and nonprofit groups conduct them to measure the effects of various products and processes.

Although few ordinary consumers are familiar with it, LCA is at the center of a battle among manufacturers, retailers and conservationists over how to measure and express a product’s overall environmental impact. Emerging rules governing LCA will help policymakers decide which products are eco-friendly and which companies are engaged in “greenwashing” (the marketing of a product as environmentally friendly when, in fact, it’s no better than any of its competitors). And future government efforts to push businesses toward sustainability will be based on LCA.

For a conscientious consumer, it’s useful to know how LCA works as well as how it can be abused.

The first step is to select what analysts call a functional unit. “You have to make sure you’re comparing apples to apples,” says Michael Lepech, a Stanford University professor studying ways to improve the sustainability of engineering processes.

It might be hard, for example, to compare a hybrid sedan and a pickup. “Hybrids and trucks don’t always do the same things: If you’re a commuter, you’re concerned with passenger miles — the environmental impact of the car for every mile you and your family travel in it; if you’re a business owner, you might be more interested in cubic-feet-of-cargo-space miles.”

The next step will be to set boundaries on your analysis.

Classic life-cycle analyses are done on a “cradle-to-grave” basis, in which the analyst considers the environmental impacts of extracting and gathering the raw materials, assembling the product, transporting it to its user, and the disposal or recycling of it at the end of its useful life.

In some cases, however, an analyst might not want that much information. If a car manufacturer has ordered the analysis to find areas where it can improve its resource or energy consumption, it wouldn’t be interested in how much energy the steel company uses or what happens to the car after it breaks down.

This “gate-to-gate” reckoning is fine for internal purposes, but if made public, it’s one way an entity could massage LCA results to its benefit — by excluding some of the more environmentally toxic aspects of a product’s impact.

That sort of thing used to be relatively common. Today, however, the International Organization for Standardization (known as ISO, for its French name) has a set of rules that govern life-cycle analysis. To get ISO certification, an analysis must be cradle-to-grave, and the resources, processes and calculations included in the analysis must go through a peer-review process conducted by disinterested third parties.

Unfortunately, this isn’t directly useful to the American consumer at this point. You’ll rarely see an ISO certification on the label of a product, but you may be able to find a related environmental product declaration on the manufacturer’s Web site, especially for large products such as furniture.

Once the analyst has chosen the functional unit and set the boundaries, it’s time to conduct a life-cycle inventory.

This process used to be incredibly burdensome: It required the analyst to track every single part of the product back to its raw material origins.

Just consider a single stainless-steel screw. The analyst has to calculate the environmental impacts of: extracting the chromium, nickel, carbon, manganese and silicon that make the steel; transporting those materials to the steelmaker; combining the materials into stainless steel; moving the steel to the screw maker; forging the steel into a screw; and transporting the screw to the car manufacturer. Now imagine doing that for each of the 30,000 parts in an automobile.

The process is easier these days because people can use widely accepted data about many materials and products to construct their analyses. It is also possible to buy software, such as GaBi and SimaPro, to conduct an analysis, although it’s pricey and the results would be a little crude without any training in the field.

A life-cycle inventory generates mountains of data, but it has to be sorted into a form that someone can understand and use. That’s where the final step, the life-cycle-impact assessment, comes in. “At this point, we group the emissions and other impacts into buckets representing particular impacts on humans, ecosystems or resources,” explains Lepech.

Most analysts separate the impacts into 11 or 12 categories, such as greenhouse gas emissions, acidification and eutrophication potential (the tendency to spur aquatic-plant growth that can choke out other marine life). In the end, the LCA report quantifies the product’s total impact on each category.

While the ISO standards have made it more difficult to get away with greenwashing, most environmental analysts feel they are incomplete.“The ISO standard sets out general rules for life-cycle analysis, but they don’t address some of the more specific issues that analysts deal with,” says Roland Clift, a professor of environmental technology at Britain’s University of Surrey.

“Cows produce beef, milk and leather,” he says. “Somehow, you have to figure out how to distribute their environmental impact between those products.”

Another problem is reporting a product’s impact on biodiversity, which is difficult to quantify. And land use is an additional point of controversy.

“In Brazil, beef produced on long-established pasture land emits about 30 kilograms of carbon dioxide equivalents per kilogram of beef,” he says. “But if rain forest has to be converted into pasture, the figure goes to more than 700 kilograms of carbon dioxide equivalents.”

The problem, as Clift points out, is that the ISO standards are mostly silent on land-use changes. So a rancher is free to ignore the removal of carbon-sequestering trees that allowed him to create a pasturing operation, making his beef look far better environmentally than it is.