NBC/WSJ poll: Heading into conventions, Obama has four-point lead

[Dr.House]

But that's not what you said. You said a segment (I presume Dems) were oversampled.

Thats what the numbers represent, yes...

How would you ever know that? By, perhaps, taking a poll to find out?

How would you know what the population breakdown is in order to determine the proper sample?

In a RV poll, you don't get that... You have to pay more for that...

No you don't. Tell us, how would this additional money be spent to determine an accurate breakdown?

Would it, for instance, poll a thousand or so random registered voters?

The breakdown numbers in post #4 show an oversampling of democratics...

I don't know how to make that any easier for you to grasp...

Barry should be up by more than 4 with an oversampling... If you look at the trend from the same firm/client, Barry has lost 2 from the period ending 7/22...

Yikes...

 

[8537]

There is no real way to know, but any poll done over the last few years puts those who consider themselves repubs and dems at around 30 percent each.

No, a poll done at the exact same time as the one mentioned above found a breakdown identical to this poll, using the exact same random sampling process!

Why would you trust some other poll more than the one done concurrent with this one?

I'm no pollster but more polls end up being wrong then right.

virtually every poll taken in the few days prior to the 2008 election was within the MoE - and of course, we fully expect approximately one out of every 20 to be outside that MoE.

Its not a poll, its a trend of many polls done over an extended period of time. More accurate then just a single poll.

Gallup August 2012: Dems +3
Pew August 2012: Dems +12

You can't do it over "a period of time" because a poll is a snapshot of a specific day.

 

[8537]

The breakdown numbers in post #4 show an oversampling of democratics...

I don't know how to make that any easier for you to grasp...

Speaking of grasping...

How did you arrive at what the proper, actual, real breakdown of Dems and Reps should be?

Hint: the only way to figure that out would be a census.

 

[SniperFire]

AP's latest now has it as just a one point election, well within the margin of error - and that is is based on survey of 'adults' - not likely voters.

Barry the Turd is sinkin fast. LOL


AP-GFK POLL SHOWS WHITE HOUSE RACE STILL TIGHT

AP-GfK poll shows White House race still tight

 

[Leweman]

No, a poll done at the exact same time as the one mentioned above found a breakdown identical to this poll, using the exact same random sampling process!

Why would you trust some other poll more than the one done concurrent with this one?


virtually every poll taken in the few days prior to the 2008 election was within the MoE - and of course, we fully expect approximately one out of every 20 to be outside that MoE.

Its not a poll, its a trend of many polls done over an extended period of time. More accurate then just a single poll.

Gallup August 2012: Dems +3
Pew August 2012: Dems +12

You can't do it over "a period of time" because a poll is a snapshot of a specific day.

People who are strong Repub and strong Dem or who consider themselves Conservative or Liberal aren't changing those stances on a given day.
Rasmussen has Rep +1. Someone has to be wrong if not all of them.

 

[8537]

Its not a poll, its a trend of many polls done over an extended period of time. More accurate then just a single poll.

Gallup August 2012: Dems +3
Pew August 2012: Dems +12

You can't do it over "a period of time" because a poll is a snapshot of a specific day.

People who are strong Repub and strong Dem or who consider themselves Conservative or Liberal aren't changing those stances on a given day.

OK. but how do you know the percentage of people who consider themselves Strong Dems and Strong Reps is the same? You could do a poll and ask....

Rasmussen has Rep +1. Someone has to be wrong if not all of them.

Nothing has to be wrong with them. Public opinion polls are typically sampled to produce an MoE within which the poll will fall 95% of the time.

Sometimes it's the other 5%.

 

[Liability]

Wall Street Journal.

SO?

If the "sample" still is overly weighted with Dims, then adding the name of the WSJ doesn't cover it with anything credible.

How do you conclude it's overweighted with Dems?

Note the word "if."

Given the history of these polls, it's also a fair guess.

 

[Dr.House]

The breakdown numbers in post #4 show an oversampling of democratics...

I don't know how to make that any easier for you to grasp...

Barry should be up by more than 4 with an oversampling... If you look at the trend from the same firm/client, Barry has lost 2 from the period ending 7/22...

Yikes...

Speaking of grasping...

How did you arrive at what the proper, actual, real breakdown of Dems and Reps should be?

Hint: the only way to figure that out would be a census.

I only said it makes sense with an RV poll... Oversampling a group that is not Likely Voters will generally favor the oversampled demographic...

This one is trending downward for Barry...

Yikes...

 

[Dr.House]

AP's latest now has it as just a one point election, well within the margin of error - and that is is based on survey of 'adults' - not likely voters.

Barry the Turd is sinkin fast. LOL


AP-GFK POLL SHOWS WHITE HOUSE RACE STILL TIGHT

AP-GfK poll shows White House race still tight

That's another one that is trending downward for Barry since their last poll...

Yikes...

 

[8537]

SO?

If the "sample" still is overly weighted with Dims, then adding the name of the WSJ doesn't cover it with anything credible.

How do you conclude it's overweighted with Dems?

Note the word "if."

Given the history of these polls, it's also a fair guess.

So they do a random sample, but the folks at the WSJ who helped fund it request that they exclude a certain number of Republican responses?

Hang up when someone says they are a Republican?

 

[Leweman]

Gallup August 2012: Dems +3
Pew August 2012: Dems +12

You can't do it over "a period of time" because a poll is a snapshot of a specific day.

People who are strong Repub and strong Dem or who consider themselves Conservative or Liberal aren't changing those stances on a given day.

OK. but how do you know the percentage of people who consider themselves Strong Dems and Strong Reps is the same? You could do a poll and ask....

Rasmussen has Rep +1. Someone has to be wrong if not all of them.

Nothing has to be wrong with them. Public opinion polls are typically sampled to produce an MoE within which the poll will fall 95% of the time.

Sometimes it's the other 5%.

I would guess that Pew poll is in the other 5 percent. You don't KNOW anything with polls. It's why we have elections. Statistics would favor doing polls based on the long term trend in party affiliation since the last election. Those who consider themselves "Strong" aren't going to fluctuate. It's the moderates or independents that will do that. Sometimes it's not possible do a poll that is representative. You can only reach who you can reach when polling.

 

[Liability]

How do you conclude it's overweighted with Dems?

Note the word "if."

Given the history of these polls, it's also a fair guess.

So they do a random sample, but the folks at the WSJ who helped fund it request that they exclude a certain number of Republican responses?

Hang up when someone says they are a Republican?

One thing at a time, bub.

First, it is time to drop the "IF." They once again managed to conduct a random telephone poll of REGISTERED voters that DID over-sample for Dims:

QF4a Generally speaking, do you think of yourself as (ROTATE a Democrat, a Republican, an independent, or
something else? (IF "DEMOCRAT" OR "REPUBLICAN," ASK Would you call yourself a strong
(Democrat/Republican) or not a very strong (Democrat/Republican)? (IF "NOT SURE," CODE AS "NOT
VERY STRONG DEMOCRAT/REPUBLICAN.") (IF "INDEPENDENT," ASK Do you think of yourself as
closer to the Republican Party, closer to the Democratic Party, or do you think of yourself as strictly
independent? (IF "NOT SURE," CODE AS "STRICTLY INDEPENDENT.")+
Strong Democrat ............................................... 24
Not very strong Democrat ................................. 10
Independent/lean Democrat .............................. 11
Strictly Independent .......................................... 13
Independent/lean Republican ............................ 12
Not very strong Republican ............................... 9
Strong Republican ............................................. 18
Other (VOL) ....................................................... 3
Not sure ........................................................... 1
+ Results shown reflect responses among registered voters.

-- http://msnbcmedia.msn.com/i/MSNBC/S...y_Stories_Teases/August_NBC-WSJ_Int_Sched.pdf

SECONDLY, it is possible to take the over-sampling into account when analyzing the results. Was such an effort attempted?

 

[thanatos144]

Democrats are grasping at straws.

 

[SniperFire]

How do you conclude it's overweighted with Dems?

Note the word "if."

Given the history of these polls, it's also a fair guess.

So they do a random sample, but the folks at the WSJ who helped fund it request that they exclude a certain number of Republican responses?

Hang up when someone says they are a Republican?

You don't seem to understand standard poll weighting.

Why do you post so much on polls if you haven't a clue?

 

[8537]

The breakdown numbers in post #4 show an oversampling of democratics...

I don't know how to make that any easier for you to grasp...

Barry should be up by more than 4 with an oversampling... If you look at the trend from the same firm/client, Barry has lost 2 from the period ending 7/22...

Yikes...

Speaking of grasping...

How did you arrive at what the proper, actual, real breakdown of Dems and Reps should be?

Hint: the only way to figure that out would be a census.

I only said it makes sense with an RV poll...

You know we can read your prior posts, right? Let's review what you actually said:

Barry should be up by more than 4 with that level of oversampling...

Yikes...

With a sample of registered voters, an even split of R/D/I...

In a sample of registered voters, polling firms don't go that deep... You have to pay bigger bucks for that and many clients don't want to fork over the cash..

.
^some sort of mysterious additional poll they'd conduct, I guess.

RV polls are the entry level, so splitting them evenly makes as much sense as anything...

But that's not what you said. You said a segment (I presume Dems) were oversampled.

Thats what the numbers represent, yes..

 

[8537]

Note the word "if."

Given the history of these polls, it's also a fair guess.

So they do a random sample, but the folks at the WSJ who helped fund it request that they exclude a certain number of Republican responses?

Hang up when someone says they are a Republican?

You don't seem to understand standard poll weighting.

Why do you post so much on polls if you haven't a clue?

Tell me Sniperfire, how would you arrive at the proper breakdown in order to establish the weighting?

 

[Dr.House]

I don't play out of context games... You're boring now...

An oversampled poll of registered voters only shows Barry up by 4.... (trending downward, too)

Yikes...

 

[8537]

Note the word "if."

Given the history of these polls, it's also a fair guess.

So they do a random sample, but the folks at the WSJ who helped fund it request that they exclude a certain number of Republican responses?

Hang up when someone says they are a Republican?

One thing at a time, bub.

First, it is time to drop the "IF." They once again managed to conduct a random telephone poll of REGISTERED voters that DID over-sample for Dims:

QF4a Generally speaking, do you think of yourself as (ROTATE a Democrat, a Republican, an independent, or
something else? (IF "DEMOCRAT" OR "REPUBLICAN," ASK Would you call yourself a strong
(Democrat/Republican) or not a very strong (Democrat/Republican)? (IF "NOT SURE," CODE AS "NOT
VERY STRONG DEMOCRAT/REPUBLICAN.") (IF "INDEPENDENT," ASK Do you think of yourself as
closer to the Republican Party, closer to the Democratic Party, or do you think of yourself as strictly
independent? (IF "NOT SURE," CODE AS "STRICTLY INDEPENDENT.")+
Strong Democrat ............................................... 24
Not very strong Democrat ................................. 10
Independent/lean Democrat .............................. 11
Strictly Independent .......................................... 13
Independent/lean Republican ............................ 12
Not very strong Republican ............................... 9
Strong Republican ............................................. 18
Other (VOL) ....................................................... 3
Not sure ........................................................... 1
+ Results shown reflect responses among registered voters.

-- http://msnbcmedia.msn.com/i/MSNBC/S...y_Stories_Teases/August_NBC-WSJ_Int_Sched.pdf

SECONDLY, it is possible to take the over-sampling into account when analyzing the results. Was such an effort attempted?

Again, Liability: Please read my question carefully. It's an honest question:

How do you know they over-sampled Democrats?

In order to know this, you would need to know the true, accurate weighting of Dems and Reps in the US. There's only one way to achieve that - a census.

Since no one is going to fund a census, we use sampling. For instance, we could sample a couple thousand random people and get a breakdown to use - we could do that and achieve an MOE of 3% or so, with 95% of samples expected to fall within that MOE.

Try this: Let's say we want to know how the nation feels about slavery reparations. We sample all Americans. But we don't know the actual breakdown of blacks and whites for some reason, so we weight the poll: 50% whites, 50% blacks.

Would that give an accurate picture?

 

[SniperFire]

So they do a random sample, but the folks at the WSJ who helped fund it request that they exclude a certain number of Republican responses?

Hang up when someone says they are a Republican?

You don't seem to understand standard poll weighting.

Why do you post so much on polls if you haven't a clue?

Tell me Sniperfire, how would you arrive at the proper breakdown in order to establish the weighting?

I would do it the industry-standard way they are doing it.

Again, if you haven't a fucking clue, why do you post so often on the topic?

Why not just read and learn?

 

[8537]

I don't play out of context games... You're boring now...

An oversampled poll of registered voters only shows Barry up by 4.....

Lol.

I'd run from the discussion if I was so spectacularly wrong as well. Nice work.