Poll herding is such a PITA from a modeling perspective.
It means that if you are trying to figure out what will happen you are looking for signals that may simply not be there.
Poll herding is such a PITA from a modeling perspective.
It means that if you are trying to figure out what will happen you are looking for signals that may simply not be there.
One way to think of this is as a file drawer problem.
Essentially as we get closer to the election your poll-based metrics become _less_ reliable, not more, because you don't know what is not being published, and you can't know if most pollsters change their methodologies or weighting slightly.
@hrefna
The whole point of poll aggregation assumes the "wisdom of crowds" thing where all the estimates are independent. Otherwise it's essentially just a poll of what the pollsters think the overall consensus is, right?
GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.
All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.