A couple of things:

1. Maybe the respondents assigned to C2 are abandoning at a higher rate for some reason. Check your in-progress responses (if they haven't been deleted).


2. Maybe, due to changes and testing, the block counts didn't start at the same place (zero) when you launched the survey. Click on 'Edit Count' to check.



---

It does seem as though C2 was abandoned at a higher rate, though not at the stage at which the conditions differ, so I'm not sure why that would be. The count indicates equal presentation to participants.
One thing I'm worried about is that the survey timers might be glitching on participants, particularly those in C2--I have timers set so that people have to stay on certain pages for a certain amount of time to watch videos. It generally seems to be working fine, but once or twice, during previews, it hit a snag, leaving me unable to move forward even after enough time had elapsed. Have you ever encountered that?
Regardless, thanks for your help. I'm really hoping to find some concrete difference in the survey setup that could explain this--something I could fix--but I guess it could be chance at this stage. If anyone thinks they'd know what to look for in terms of survey design, I'd be happy to send over a more detailed copy of what I've got.

lcrisman ,
A couple of more thoughts:

1. The issue may be the video timers since it was a known issue going in. I don't know anything about the specifics of your videos or timers, but generally, it is better to use a video ended event rather than a timer.


2. If you collected meta info, you could take a look at that to see if there is a correlation between browser type(s) and adbandons. It could related to a incompatibility with specific browsers/devices.



---