and, likewise, i'm going to ignore the framing around this recent polling and just look at the data, however scant it may be. here's our trend line from campaign research, which is conducting "online research" around voting intentions in the next election:

the immediate response would be "not much movement outside the margin". however, as this "online research" does not utilize random sampling, there is no margin of error to consult.

how have the conservatives changed? well, they're down a point since february, but really pretty much flat since december - and no doubt for months before that. i'm not even sure that the second coming of christ would move the conservative numbers, at this point. their base is rock solid, entirely unreachable, but they're continuing to fail at generating interest outside of it. the data really perfectly represents the stereotype of the conservative supporter as an ideologue that you simply can't argue with it, and the party as having little appeal to much of anybody else in the 21st century.

the ndp are similarly flat. but, the liberals are trending mildly downwards.

so, if the ndp and conservatives are flat and the liberals are trending down, where are the votes going?

to their discredit, campaign did not post a trend line for the other parties, but you can at least find a snapshot of the recent "online research".

looking at this, we don't know if the undecided or the greens or the bloc went up, but the greens are performing fairly well in this poll, and the undecideds are actually fairly low compared to where they were at this point last time around.

it's easy to make a conclusion then - if this "online research" is uncovering anything, it is that people are a little bit cynical about the liberals right now, but don't like the other options. you really didn't need to conduct online research to figure that out....

comparing this to the ridiculous media framing is instructive, as it demonstrates the purpose of what the firm is doing. the headline is that the liberals are in trouble; the data suggests no such thing. but, who reads the data? people read the headline. this "online research" is not meant to measure public opinion, but to create it. so, there is a gramscian caveat to my analysis: while the issue doesn't appear to have harmed the liberals at this point, if the media continues to push the point that it has then it very well may in the end.

the liberals should be more concerned about the companies doing the research than they are in the results of the research.