An analysis published Friday confirms the state of American gun policy science is not good, overall.

The nonprofit RAND Corporation analyzed thousands of studies and found only 63 that establish a causal relationship between specific gun policies and outcomes such as reductions in homicide and suicide, leaving lawmakers without clear facts about one of the most divisive issues in American politics.

The analysis is notable for its depth, if not for its overall finding — RAND spent two years and $1 million dollars on the project, which, beyond the analysis, also included a survey of gun policy experts and the construction of a research database on state gun laws.

For decades, social scientists and other researchers have pointed to a profound, and purposeful, lack of federal funding for gun research and a lack of federal data-gathering on guns as enormous impediments to studying gun violence. The federal government has spent much less on research into gun violence than on similarly lethal issues, such as motor vehicle crashes, liver disease and sepsis.

"Most of the effects that we were looking for evidence on, we didn't find any evidence," says Andrew Morral, a behavioral scientist at RAND and the leader of the project.

They found, for example, no clear evidence regarding the effects of any gun policies on hunting and recreational gun use, or on officer-involved shootings, or on mass shootings or on the defensive use of guns by civilians.

There were some categories with better data, however, Morral says. There is relatively strong evidence, for example, that policies meant to prevent children from getting access to firearms — such as laws that require guns to be stored unloaded, or in locked containers — reduce both suicide and unintentional injury and death.

Previous work has also found that places that require a permit (issued by law enforcement) for the purchase a firearm do reduce violent crime.

There is also some evidence that prohibitions against purchase by people who have been diagnosed with mental illness reduce violent crime, and that "stand your ground" laws, which allow citizens who feel threatened in public to use lethal force without retreating first, lead to an increase in violent crime.

In general, however, good studies were few and far between, the RAND researchers say.

To help bridge the data gap for scientists, RAND has now compiled an open-source database of state-by-state gun laws going back to 1979. A similar, unrelated database run out of Boston University was published last year.

"Given the lack of data in the firearm research field, [the RAND database] will be a welcome addition, and I'm sure it will be valuable to researchers," says Dr. Michael Siegel, the physician who leads the Boston University project. He notes that, since it launched, the Boston database has been most helpful for journalists writing about gun violence.

"It's funny because I didn't actually think about that as a purpose of the project when we started it," says Siegel, "but it seems to have become the primary use of the database so far (in addition to use by other researchers)."

The RAND team also surveyed 95 gun policy experts from across the political spectrum about what they thought the effects of 15 different gun policies would be on 12 outcomes. The policies included universal background checks, bans on the sale of assault weapons and high-capacity magazines, expanded mental illness prohibitions, minimum age requirements and required reporting of lost or stolen weapons.

Many of the policies are now being considered by state and federal lawmakers, after a man used an AR-15-style rifle to kill 17 people at a Florida high school in February.

The vast majority of the specialists RAND surveyed agreed that the primary objectives of gun policies should be reducing suicides and homicides, and that protecting privacy, enabling hunting and sport shooting and preventing mass shootings were secondary priorities.

"That was a surprise, actually," says Morral. "I think people on either side of gun policy debates think that the other side has misplaced values — or that it's a values problem, in any case. But that's not what we find. We find people prioritize the same things in the same order."

However, those surveyed varied widely in their predictions about how different policies would affect each outcome.

"Where they disagree is on which laws will achieve those those objectives. So this is a disagreement about facts," says Morral. "And the facts are sparse."

Copyright 2021 NPR. To see more, visit https://www.npr.org.