Consultores 21 carried out 1500 in-home interviews between Nov. 26th and Nov. 30 in the country's nine largest cities (margin of error=2.6%) and compared the results with an earlier poll of 840 voters in those same nine cities (margin of error=3.5%). So this poll is not directly comparable to the previous C21 poll we reported, which was based on a nationwide (i.e., Urban+Rural) sample.
Once again, the momentum is all on one side:
![](http://1.bp.blogspot.com/_iHLs_Wf3Ohs/R1J1ik0WW0I/AAAAAAAAAw4/SxkE2I-cZOw/s400/C21.001.jpg)
It's not just that more people are planning to vote No than just a couple of weeks ago, it's that more people believe the No side will win. That conviction is important when it comes to turning out NiNis.
Nonetheless, the all-urban sample tends to overstate the No vote. Based on previous data about the Urban-Rural split, Consultores 21 corrects their results. They then go out on a limb and project a nationwide outcome: 55.7% for the "No" and 44.4% for the "Sí".
![](http://3.bp.blogspot.com/_iHLs_Wf3Ohs/R1J1jE0WW1I/AAAAAAAAAxA/RM122yecB9A/s400/C21.002.jpg)
If anything, they conclude the gap between the "No" and the "Sí" has widened in the last few days, and that the trend points to the "No" to increasing its lead.