What a Good Strategic Poll Looks Like in the NDP Leadership Race

24 posts / 0 new
Last post
KenS
What a Good Strategic Poll Looks Like in the NDP Leadership Race

V

V

KenS

From Pundits Guide:

What a Good Strategic Poll Looks Like in the NDP Leadership Race

I think the availability of this is an incredible opportunity for our collective self education on doing politics.

I made a dedicated thread for it because I do not want that opportunity to get mixed up with opinions on and allegiances of the leadership race itself.

No doubt I will make my own comments on where this fits in the leadership race; but I'll do in the existing threads, and please follow suit yourself.

 

I am wondering how many babble residents got this survey.

I think I remember Wilf referring to what must be this, at about the right time.

Anyone else get the survey?

KenS

When do we ever get a chance to see something like this, and discuss how it works and can work?

And in this particular case, my hunch is that we are looking at something where we are all on pretty equal footing when it comes to familiarity with the issues and the ongoing dynamic that the survey is meant to measure.

Idealistic Prag... Idealistic Pragmatist's picture

I got it, and filled it out. I remember being impressed with it at the time, and am again now that I have a chance to look at it again.

Howard

I think the poll is a good one but largely a defeatist strategy.

It focuses on individuals that are already NDP members, tries to tease out the depth of their entanglement with the NDP, and completely ignores new sign-ups.

It raises serious concerns about Topp's candidacy for me as his strategy has seemed to be to appeal to the NDP's established bases of support (particularly in BC) with little to no attention to growth for the party. Contrast that with other leadership campaigns that have spent a lot of time in the hinterlands of NDP support, or pounding the pavement in NDP bastions past.

The optics of a strategy that only appeals to the party's existing membership base are also very bad for Topp, as everyone already knows that he has the endorsements of the party establishment. One major concern this overlooks is NDP member's desire for growth and the ability to move in to government. Whether Topp feels it or not, the urgency of the membership to get Harper out and move in to power may override perceived ideological differences between voters and candidates.

ottawaobserver

Howard, dumb question, but how could you poll new sign-ups in November?

KenS

And what optics are there in poll questions most of us never see or hear about even indirectly?

FWIW, there would be an entirely different logic to the strategy of contacting new sign-ups. For one thing, first you need to do the 'big picture identification' of the ones you want to put energy into, and how much. 

Howard

ottawaobserver wrote:

Howard [snip] how could you poll new sign-ups in November?

You can't but you can poll NDP supporters or some measure of the NDP universe (i.e. those that would consider supporting the NDP, in various ways).

A really simple adjustment to Topp's poll that would have been strategic in this way, would have been to ask existing members/poll respondents if they had family or friends (or neighbours or coworkers) that were not yet NDP members but that might sign up. This would give you a sense of where growth could occur (i.e. what kind of supporters you could ask for help in signing up new members). Also, gauging which members might have more that they would be willing to give (in terms of volunteering or other forms of donating).

There are many other things that could be done to get at the same question. You mostly have to ask yourself as a candidate: what kind of membership growth am I after and what do I think is the most efficient way to get it? Topp's poll is unidimensional in that it only goes after existing members and wastes a lot of ink in terms of trying to figure out how "pur et dur" of an NDPer they are.

Howard

KenS wrote:

And what optics are there in poll questions most of us never see or hear about even indirectly?

Topp releasing the full details of the poll and confirming he was behind it is a strategic move, to show just how strategic and clever he is. The NDP did this with Brad Lavigne's faux "memo to Layton" before the spring campaign, leaked to the media and "informing" Layton that the NDP was poised to make major gains. In a follow-up interview some weeks later, Lavigne confirmed that he had leaked the "memo" and restated its points. It was an attempt to show that the NDP was ready for an election and it was ready to bargain strongly with Stephen Harper over budgetary concessions (of course, as Robin Sears has written, Layton specifically picked a few "demands" that he knew Harper could not deliver because he needed the provinces' help [not forthcoming] among other things). So all of this is inside baseball but it is meant to contribute to all the aura of Topp being a formidable (and strategic) candidate that can take the party to victory.

This is also a trick the Conservatives have used (probably were the NDP got the idea from): leaking elements of their strategy, to seem more omnipotent and strategic.

Like I said earlier, good poll...but its myopic.

KenS

I dont think I have a handle on what a typical existing member is like.

Mostly I see people making a lot of assumptions- a good portion of which I think are likely incorrect.

Nor do I think that new members being signed up are necessarily qualitiatively different.

You dont assume they are more or less the same. And it is good to know.

But its hard to do a good survey like this. It doesnt take many attempted to permutations to turn the whole thing into junk.

The operating pronciple to first do what you can do right. After you make sure of that much, add what you can.

KenS

Howard wrote:

Topp releasing the full details of the poll and confirming he was behind it is a strategic move, to show just how strategic and clever he is. 

Have the results been released? That would be a surprise.

And confirming it is hardly a strategy to achieve something, and/or intentionaly putting it out there.

The campaign would look stupid or suspicious or something if in light of having the survey discussed they pretended it didnt exist.

It has been noted here since it happened in November.

Howard

KenS wrote:

Howard wrote:

Topp releasing the full details of the poll and confirming he was behind it is a strategic move, to show just how strategic and clever he is. 

Have the results been released? That would be a surprise.

And confirming it is hardly a strategy to achieve something, and/or intentionaly putting it out there.

The campaign would look stupid or suspicious or something if in light of having the survey discussed they pretended it didnt exist.

It has been noted here since it happened in November.

Riiiiiight.

Releasing the poll but not the results implies that Topp has great intel on the race, and great strategic chops, which is how it has been interpreted on the punditsguide post you link to.

Topp was under no obligations to release any information about the poll. Many campaigns have conducted polls of various types but only Topp is releasing information about them.

Of course, for all we know, the results of Topp's poll might have been garbage (e.g. how does the age distribution of respondents compare to the age distribution of the membership?), but these results have not been released and probably never will. Still he looks strategic (and well-informed/all-knowing).

KenS

Howard wrote:

 

Howard wrote:

Of course, for all we know, the results of Topp's poll might have been garbage (e.g. how does the age distribution of respondents compare to the age distribution of the membership?)

Thats the biggest reason that any poll gets good information about the respondent, so you can weight appropriately the aggregate respondents. If you get two times too many male respondents, you weight the resulting data just half.

ottawaobserver

Howard wrote:

ottawaobserver wrote:

Howard [snip] how could you poll new sign-ups in November?

You can't but you can poll NDP supporters or some measure of the NDP universe (i.e. those that would consider supporting the NDP, in various ways).

A really simple adjustment to Topp's poll that would have been strategic in this way, would have been to ask existing members/poll respondents if they had family or friends (or neighbours or coworkers) that were not yet NDP members but that might sign up. This would give you a sense of where growth could occur (i.e. what kind of supporters you could ask for help in signing up new members). Also, gauging which members might have more that they would be willing to give (in terms of volunteering or other forms of donating).

There are many other things that could be done to get at the same question. You mostly have to ask yourself as a candidate: what kind of membership growth am I after and what do I think is the most efficient way to get it? Topp's poll is unidimensional in that it only goes after existing members and wastes a lot of ink in terms of trying to figure out how "pur et dur" of an NDPer they are.

I don't know about that, Howard. At the time, the survey's sponsor was anonymous, so how would they use it to sign up new members or find volunteers? It just doesn't make sense.

Howard

ottawaobserver wrote:

Howard wrote:

ottawaobserver wrote:

Howard [snip] how could you poll new sign-ups in November?

You can't but you can poll NDP supporters or some measure of the NDP universe (i.e. those that would consider supporting the NDP, in various ways).

A really simple adjustment to Topp's poll that would have been strategic in this way, would have been to ask existing members/poll respondents if they had family or friends (or neighbours or coworkers) that were not yet NDP members but that might sign up. This would give you a sense of where growth could occur (i.e. what kind of supporters you could ask for help in signing up new members). Also, gauging which members might have more that they would be willing to give (in terms of volunteering or other forms of donating).

There are many other things that could be done to get at the same question. You mostly have to ask yourself as a candidate: what kind of membership growth am I after and what do I think is the most efficient way to get it? Topp's poll is unidimensional in that it only goes after existing members and wastes a lot of ink in terms of trying to figure out how "pur et dur" of an NDPer they are.

I don't know about that, Howard. At the time, the survey's sponsor was anonymous, so how would they use it to sign up new members or find volunteers? It just doesn't make sense.

They would use it to figure out what kind of NDP member had friends or family members (etc) that were not currently members but could potentially be members. They would use it to figure out what kind of NDP member would be willing (or best able) to help with new signups. What's more they would know what issues were important to these members and which candidates they were leaning towards (and thus more likely to help with recruiting). That is more of a growth strategy. This poll is a, how do I hold on to the support the party already has.

My biggest point is, the membership wants the leadership to aim for growth. If they play it safe and aim to keep what we have, they will miss people's desire for growth (and change). I think that is precisely what Topp and the establishment are in the process of doing.

ottawaobserver

OK, that's way more than I would read into it, but whatever.

Howard

Howard wrote:

ottawaobserver wrote:

Howard [snip] how could you poll new sign-ups in November?

You can't but you can poll NDP supporters or some measure of the NDP universe (i.e. those that would consider supporting the NDP, in various ways).

By the way, this was done by the media (with a sample size of ~300) and you criticised it. I thought it was actually a good idea (to poll NDP supporters about who they would like as leader), but agree that the sample size was too small.

I wonder if Topp or any of the other campaigns are polling NDP supporters as opposed to NDP members. I think it would be a good idea, as there has been much hay made of campaigns taking existing supporters and getting them to sign memberships.

Howard

I guess I will give Topp some credit for being forward-looking in that he asks how long members have been in the party. This might allow his campaign to get a sense of how the membership has changed over time and maybe infer what the "new wave" of supporters has been to date. His recruiting efforts could target more "new wave" type members.

It does not however, do any sleuthing about who the "next wave" of membership signups might be.

Perhaps Topp thinks the "new wave" is a better predictor of the "next wave" than polling supporters or asking questions of existing members and the non-member supporters they know.

ottawaobserver

I read it as a poll *of* NDP members, Howard.

Howard

ottawaobserver wrote:

I read it as a poll *of* NDP members, Howard.

oic. my apologies.

ottawaobserver

No problem. Now I understand your earlier points.

KenS

Back up boy.

Nobody ever releases results of campaign surveys, so the fact we are getting those means nothing. Alice Funke  said in the blog comments that she got the survey from a respondent's screen grabs.

And in my estimation, for the Topp campaign having all this information out there of what is being looked for sounds like a double edged sword. It's only guaranteed to impress the people who like him already [no gain], who have always been Topp's best constituency, albeit a limited one. You are giving us a real time demonstration of the downsides of such a close view of what is going on behind the veil.

KenS

Although this is for the leadership race, this survey is the kind of thing that is needed for the NDP to 'push the envelope' or 'move the goal posts' of electoral politics.

If you really mean to do it, and not just pay homage to the pretty words, moving the goalposts is risky business. Because by definition, it is about taking voters in your supporter universe to places and opinoins where they are not at the moment. Where they could be, but are not at the moment.

And if you are going to do that without heedlessly paying a stiff price, with not even any long term rewards, you had better go out there and test in the real world the parameters of where people are ready [and not] to move to.

In the NDP, when its just words about going to new places, we don't go there. 

"Someday."

[Maybe]

And if you try to travel to riskier territory on your hunches and gut feelings- no matter how much experience you have and how much you have talked and reflected on plans- the consequences are not going to be pretty.

The main risk is not that you end up going where you should not- although you do need also to check on that. If you do not do real research, you are going to miss both opportunities and problems in advancing the agenda you have decided you want to pursue.

Because the Topp campaign's survey is gauging NDP members, and the parameters of a leadership race, it is going to look different than testing where potential supporters might go with you. And there would be more research instruments for the latter. But the operating principles are the same.

For example: when this survey came out, Topp was already beyond the point of no return on commitment to the taxation policy element of his social equailty agenda. But the campaign still needs to know if that is resonating or likely to resonate. And if the feedback data is disapointing- what to supplement with? Etc.

 

janfromthebruce

I thought the poll was excellent and gets top marks. Being somewhat of a researcher in my past I thought it was well designed and thoughtout.