One of the planet’s largest conservation groups, the World Wildlife Fund (WWF), just published a frightening statistic: Populations of most major animal groups, including mammals, birds, and fish, have declined by an average of 69 percent in the last half century.

It’s worse for animals in certain habitats and regions. Communities of freshwater species, such as fish and frogs, have declined by an average of 83 percent, globally, whereas populations of all major vertebrate groups found in Latin America have fallen by an average of 94 percent during the same period.

WWF and the Zoological Society of London, another nonprofit, calculated these stunning figures using a popular metric called the Living Planet Index (LPI). The index is designed to measure how animal populations, on the whole, are changing through time. The idea is that this figure can provide an early warning that ecosystems are in peril, according to Rebecca Shaw, the chief scientist at WWF.

“What we care about is ecosystem health,” she told Vox. “That’s what underpins a stable climate, healthy food production, healthy water production, and, really, human health.”

WWF/Zoological Society of London

Big numbers (i.e. “a 69 percent decline”) tend to make headlines for a few obvious reasons. For one, they appear simple. They’re also extreme. And for many decades, environmental groups have relied on negative numbers to raise money for conservation.

These headline figures are undoubtedly important; they highlight a very real and very severe crisis of biodiversity loss. The problem is that they’re confusing and often misinterpreted in a big way. Even when they’re not, there’s a lot that these headline numbers leave out, including the more hopeful data and stories that demonstrate how conservation can actually work.

What the Living Planet Index is and what it isn’t

A day in the life of an ecologist often includes counting animals. They tally up bug splats on car windshields, fly drones over colonies of waterbirds, and strap camouflaged cameras to trees that snap photos when animals stroll by.

Over time, these counts reveal how wildlife populations are changing. If a group of manatees in Florida, for example, runs out of food one year, a later survey may find fewer of them, revealing a population decline, often expressed as a negative percentage.

The Living Planet Index is built upon all of this counting.

To come up with the global LPI, scientists first calculate how individual populations of birds, mammals, reptiles, amphibians, and fish have changed between 1970 and 2018 (the data always lags behind a few years). A population of, say, 1,000 manatees that has lost 500 has decreased by 50 percent. The same is true for a population of 10 that has lost just five.

Then they average up all of those changes, be they increases or decreases, to produce one number. That means the index is an average of changes in population sizes, not the average of the number of creatures lost. (Our World In Data has a very helpful explanation if you want to learn more.)

Joe Raedle/Getty Images

But let’s turn back to the headline figure from WWF’s report: 69 percent. That number suggests that if you average all the changes in wildlife populations, globally, since 1970 — a population of frogs is down, a community of gorillas is up, and so on — you’ll get a decrease of 69 percent.

That figure is broadly helpful. It helps us understand that a lot of animal populations are in decline. But — and this is crucial — that figure does not mean there are two-thirds fewer animals today compared to 50 years ago. Again, it’s not counting all the animals lost in each group and adding that up; it’s measuring the relative size of the decline in each population and averaging it.

WWF/Zoological Society of London

It’s this distinction that is confusing and most often misinterpreted. Two year ago, when WWF published the last edition of the report, which revealed a similar result — an average decline of 68 percent globally — several news outlets ran stories like, “The world lost two-thirds of its wildlife in 50 years.” That’s wrong.

“[The global index is] constantly being interpreted as we’ve lost 69 percent or 68 percent of the abundance of animals worldwide,” said Brian Leung, an associate professor of ecology at McGill University. “It’s not that. It’s not just that everything is in decline.”

WWF has tried to get ahead of misleading headlines, explaining in the report that a 69 percent decline doesn’t mean that 69 percent of individual animals have been lost. It also presents a ton of other information in the report, including other metrics that often get overlooked. “We’re going overboard this time to be really clear about what this indicator is and what it’s not,” Shaw said of the LPI. “We can’t really control the interpretations in the media.”

Even if all the media reports take pains to represent the LPI statistics accurately, there’s still a lot these figures leave out. That’s not a knock on WWF or ZSL — the global Living Planet Index is an impressive analysis. But on its own, it’s just not the whole picture.

No single metric is perfect

In recent years, scientists have pointed out other shortcomings of the LPI, such as that extreme outliers — namely, wildlife populations that have declined massively since 1970 — can bring the overall average down. Research by Leung showed that if you remove these outliers, the overall trend of the global LPI is much less dramatic (you can read the responses to that paper here; WWF addresses outliers in the supplementary information for the report). Other studies have pointed out that some statistical methods used to calculate the LPI also seem to favor a negative trend.

Then again, consider what the LPI is trying to do: summarize global biodiversity loss in a single number. That’s not easy, said Hannah Ritchie, the head of research at Our World In Data, who has written a lot about LPI. As she told Vox over email, any metric on that scale is going to have some important caveats. That’s why scientists I spoke to tended to direct their criticisms at the idea of summarizing biodiversity loss in general, not at the LPI per se.

“There’s no such thing as a perfect indicator,” said Falko Buschke, an ecologist and freelance researcher, who led a study last year that looked at how random fluctuations in animal populations affect the LPI. These headline figures are useful for advocacy, he added, but from a science perspective they’re often confusing and can raise more questions than they answer.

“It’s a noble endeavor but also a difficult endeavor,” Andrew Rypel, a professor and fish ecologist at the University of California Davis, said of big summary statistics. “You want the number to be accurate but you’re also trying to distill down a lot of complicated data.” An added challenge, he said, is that there isn’t great data for a large number of species, including those that inhabit freshwater (where LPI registered some of the largest average declines).

This might all sound a bit … pedantic. Why are we fussing over how to measure biodiversity loss when the news is bad by any metric?

In the race to reverse wildlife declines, metrics matter. They help dictate how government officials, scientists, and environmental advocates divvy up a limited conservation budget. They can also inform public policy. Researchers used LPI, for example, to measure progress against several biodiversity conservation targets under a major UN convention.

There is more, though, that policy makers can mine from this data, and they could also highlight some of the better news.

The case for good news

Headline figures like the global LPI tend to obscure the fact that populations of plants and animals are changing in vastly different ways. On the whole, yes, those changes are negative and represent declines — and there’s some evidence that readers pay more attention to bad news. But many wildlife populations are stable or even increasing. In fact, half of all vertebrate populations in the report show an increasing trend.

WWF points this out plainly in the report. The number of loggerhead sea turtles nests increased by 500 percent in Chrysochou Bay, Cyprus, between 1999 and 2015, the group writes. That’s thanks to deliberate conservation activities such as relocating nests that were close to human developments.

Etienne Torbey/AFP via Getty Images

Meanwhile, the report reveals that populations of mountain gorillas in Central Africa’s Virunga Mountains have grown from 480 individuals in 2010 to more than 600 today — again, thanks largely to conservation.

There are hundreds of examples like these (check out the Wildlife Comeback Report) and these are important stories to tell. People are tired of seeing the same negative figures that haven’t changed much in the last decade. “I feel the reticence of people to take in new information like this,” Shaw of WWF said. This data shows what’s working: that wildlife populations can recover, and that decades of conservation work haven’t been fruitless.

Indicators that clump populations together have value; they indicate, in simple terms, that ecosystems continue to trend in the wrong direction. And LPI is one of many featured in WWF’s report (there’s also metrics like “biodiversity intactness” and “mean species abundance”). But it’s important to look beyond headline figures to see what’s actually happening to individual populations.

“Of course, that doesn’t give us a snappy headline figure,” Ritchie said. “But it does give us essential insights into what populations are doing well, what ones are struggling, and what we need to do to restore them back to health.”

The Living Planet Index has a data portal that allows you to do this. It includes tens of thousands of animal populations that you can click into to see how they’re changing over time.

The new report also focuses on how to curb the decline of wildlife, offering many examples of what seems to work. This is critical, Shaw of WWF says, because more people than ever are aware that there’s a crisis. They want solutions.

“The old-school way of talking about this is to pound on the negative numbers just to get anybody to listen,” Shaw said. “But people are listening now.”