[monkeyshine]

they also did a supplementary sample of 125 Latino "likely voters", which were added to the Latinos within the 800

But the link above says "To allow for analysis, the main sample was supplemented to provide a total of 125 Latino likely voters

So, what am I missing?

[okie01]

My mistake. Your expression is correct.

The sub-sample totalled only 125 -- barely sufficient to "split" for analysis. But the additional Latinos to make up the 125 sub-sample would not have been counted in the 800.

Odd, though, that -- in a random sample of 800 -- they didn't find even 125 Latino respondents (15%).