We operationalize the concept of "independent" using questions pid1d and pid1r, which each ask:
*  *"Generally speaking, do you usually think of yourself as a Democrat \[Republican\], a Republican \[Democrat\], an independent, or what?" (the two questions are identical except that "Democrat" and "Republican" are swapped around).*

If the participant answered that they are an independent in either question, we consider them to be "independent" for the purposes of this study.

We operationalize the concept of whether or not a voter believes that the federal investigations are "baseless" using question russia16, which asks:
*  *"Do you think the Russian government probably interfered in the 2016 presidential election to try to help Donald Trump win, or do you think this probably did not happen?"*

There are other questions which also ask about the Russians and Mueller's investigation (muellerinv and coord16) but russia16 most directly gets at the issue of Russian interferance.

If the participant answered that they think the Russian government probably did not interfere on russia16, we consider them to believe the investigations are "baseless" for the purposes of this study.

**We assume that all observations count as "voters". If we wanted to be very specific with the operationalization, we could operationalize the concept of "voter" like this (but we have not done so in this analysis):**

Operationalize the concept of "voter" using questions turnout18, turnout18ns, turnout16, and turnout16b which ask:
*  *"In the election held on November 6, did you definitely vote in person on election day, vote in person before Nov 6, vote by mail, did you definitely not vote, or are you not completely sure whether you voted in that election?"*
*  *"If you had to guess, would you say that you probably did vote in the election held on November 6, or probably did not vote in that election?"*
*  *In 2016... did you definitely vote, definitely not vote, or are you not completely sure whether you voted?*
*  *Do you think you probably voted or probably did not vote?*

If the participant answered that they definitely voted (in person on election day, in person before election day, or by mail), or *probably* voted in either election, we would consider them a "voter" for the purposes of this study.