A reminder, if a poll shows a candidate being +2 or -1 since last month, it means nothing's changed

[Godhumor]

Last edited Fri Dec 11, 2015, 01:11 PM - Edit history (2)

I'm not going to go through the real way of checking the stability of numbers in a poll month over month, because it requires a fair bit of calculation across multiple polls. However, there is a white lie stand-in that is almost as good.

Basically, look at a poll's margin of error. If a candidate's change in support, in either direction, is less than or equal to the MOE, the change is not significant. If it is more than the MOE, then there is some real change occurring (though maybe not as much as you think.).

For example, A poll with a 5% MOE has Clinton support at +3 since last month, Sanders at -2, and O'Malley at -1. All three candidates have changes within the MOE range, so, the race would be said to be stable from last month, as it is possible no actual change occurred.

Another poll has an MOE of 4%. Sanders is +6 since last month, Hillary is -3 as is O'Malley. Sanders has significant change while Clinton and O'Malley don't. However, with the MOE stand-in, Sanders real change could be as low as +2 (His net improvement minus the MOE). Furthermore, HRC and MOM both have nonsignificant changes. The instinct is to say Sanders benefited from both of them equally, but, due to their insignificant changes, we don't actually know if that is true. It is possible he gained from both, or from just one or from neither (Could be he picked up some undecided voters.).

Be careful interpreting trends month over month. In reality, this race has been ridiculously stable the past two months. Almost boringly so.

On edit: if you're trying to interpret something like Reuters/Ipsos, just set your computer on fire and walk away.

On 2nd edit: The white lie had been called out, so let's add an addendum. To see what portion of a month over month change may be significant, you can take the absolute value of the MOE and subtract out the change, or vice versa. That gives you the total net change that might be significant depending on where in the MOE values actually fall. Since we don't know where in the MOE the values are, this is a dubious calculation. That is why the White lie to knowing if change is significant is if it is completely outside the range of the MOE.

[1StrongBlackMan]

1. +1 ...

But that doesn't allow us encouragement!!!

So stop!

[Godhumor]

3. Your +1 is within the OP margin of error

So I don't know if it was a real +1.

[JaneyVee]

7. Here's another +1. SURGE!

[morningfog]

5. Everything, including this OP allows encouragement.

[Manifestor_of_Light]

2. Margin of error, right.

I noticed that on MSNBC when they go over changes in stats, they will say "this is bigger than the MOE".


And a poll doesn't tell the person taking the poll what the voter will do when they are actually punching a ballot and thinking, "Do I REALLY want to vote for that person?" which I don't think is the same thing as the old phrase "hold your nose and vote".

I remember seeing Tweety saying that the margin of white voters for Obama who really put him over the top was 6 percent, or something like that.

[Godhumor]

4. Right, but they're not talking month over month

They're commenting on the difference between candidates being larger than the MOE. If the difference is less than the MOE x2, the race is considered tied.

[GusBob]

6. I really appreciate your input this election, thanks

Question 1: could you briefly explain how moe is calculated ?

Q2: what's the best way of setting my computer on fire?

[Godhumor]

8. Sure, I can answer both questions

1) Assuming how asking me about what the MOE days and his it works in polling, I'm going to refer you to a primer I wrote earlier this year:

http://www.democraticunderground.com/1251567623

If, instead, you're actually asking how to mathematically calculate one, let me know, and I'll put together a post to explain it.

2) Nuke it from orbit. Only way to be sure.

[GusBob]

9. I see it in the primer thanks

:thumbs up:

[ismnotwasm]

10. I love your OP's

Just sayin

[Godhumor]

11. Gracias

Consider this my final "naked" bump.

[NurseJackie]

12. As the number of "undecided" voters shrinks ...

... the days of big-poll-swings is likely to be over. Things are settling in now, and it's unlikely that there will be any genuine "surges" or "plummets".

The poll numbers (individual polls and trends) don't look very promising for Bernie. In spite of all the aggressive optimism and gleeful comparisons to 2008, Bernie's fans see the same numbers that everyone else does, and they know full well that Bernie no longer stands a chance at winning the nomination. (An individual state here or there, perhaps, but not the nomination.)

[JonLeibowitz]

13. This OP is mathematically false.

Actually, it means the mean, the likeliest percentage of <insert what you are measuring here> has shifted by +2 or -1. The MOE shifts with that.

Please consider self-deleting this post, it is inaccurate.

[Godhumor]

14. Hmm. No I don't think I will be deleting, but I will edit

Last month a poll shows Clinton at 49 within an MOE of +/- 3%

This month it is at 50 with same MOE:

So 46 - 52 month one
Or 47 - 53 month two.

The only results outside of each other is the extreme edge of the MOE at 46 or 53.

Using the MOE as a stand-in for whether a change is significant is standard practice for quick interpretation. But I will add a note that when the mean moves it is possible to have significant change on the far outside of the band. Even though the only way to know whether you have significant change or not is if the margin is outside the range of the MOE. Which was the point of the OP.

[JonLeibowitz]

15. That's accurate. Thank you for the update.

I still think mean shifts can be considered, but small changes relative to the MOE do deserve to be treated with a great deal of circumspection.

[Godhumor]

18. No worries. Truthfully, I didn't want to get into the grey area n/t

[redgreenandblue]

16. Here is what you are looking for:

http://elections.huffingtonpost.com/pollster/2016-national-democratic-primary

Moving averages.

[Godhumor]

17. Yes, aggregations are a different beast

I'm talking specifically about, say, PPP showing a change in support from this month to the next.