Thanks for pulling up data!
These numbers are incommensurate in a way that may not be obvious.
7% of the population doesn't tell you what population fraction is covered by such policies.
36% coverage is even harder—every child in the US is eligible for Medicaid, and such children may not always need it, or may move states after using Medicaid, in a way that makes them doubly counted.
80% of the working population is also less clear; is that 80% of policy-holders get their own policy through their own job? Or 80% of working-age people have a policy through some workplace, even if they are not working?