Technically, in the US, executive branch agencies do two things: (1) sponsor actual science (either directly by employees, or by grant to third parties), (2) decide on how to apply scientific results.
"The EPA" is more of a body concerned with "What do we do about scientific results?" than "What is scientific truth?"
It often conducts research, usually in areas where private industry has no incentive to do so, but that's not why it exists.
And historically the dichotomy between science and policy has been reflected in the organization: scientists were highly discouraged from advocating policy positions in their research results, and policymakers were highly discouraged from influencing science.
Technically, in the US, executive branch agencies do two things: (1) sponsor actual science (either directly by employees, or by grant to third parties), (2) decide on how to apply scientific results.
"The EPA" is more of a body concerned with "What do we do about scientific results?" than "What is scientific truth?"
It often conducts research, usually in areas where private industry has no incentive to do so, but that's not why it exists.
And historically the dichotomy between science and policy has been reflected in the organization: scientists were highly discouraged from advocating policy positions in their research results, and policymakers were highly discouraged from influencing science.