This one is a bit trickier, but again, it's a non-story.
Three professors at Old Virginia U have looked at a study (
CCES) that queried people BEFORE THEY VOTED on who they intended to vote for and then came back later to survey them about the election. The survey data HAVE NOT BEEN DELIVERED YET.
The pre-election survey (
CCES Pre-Election Survey, 2016 | CCES) was released in November, and covered about 84k "likely voters." There is no mention of non-citizen voters, just unregistered voters. Is this the number our esteemed professors are looking at and extrapolating from?
I have emailed the head of the study to ask about the non-citizenship data. Hopefully there is a response coming in the media, as their research is being cited as a source without it even being released yet.
(
Announcing the 2016 Cooperative Congressional Election Study | CCES)
"Schedule
The study will be fielded in October and November 2016 (pre- and post-election waves). Survey data will be delivered by March 2017, and data matched to the voter files will be delivered in July 2017."