Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
Ralston took a look at the raw numbers from the poll and found something else interesting: the CNN/Time pollster apparently didn’t interview anyone under the age of 50. In the age demographic breakdowns, all questions have “N/A” for 18-34 and 35-49 categories.
In Kentucky, the poll shows no data for 18-29YOs and non-white voters, and all the self-described liberals must have been in hiding.
Well, the good news is, no one ever quotes CNN/Times polls unless they are part of Real Clear Politic's averaging. Thank god some people in the media, even if they are web media, are finally looking at sampling methodology and data instead of just spewing "results."
I don't think I've ever seen a poll like this, that leaves off voters under 50, rural voters and those that live in "growing suburbs".
Some pollsters due this when the sub sample is quite small. So its not saying N/A because they didn't interview anyone in that particular demographic, but rather because the sub sample is very small and that translates into a large margin of error. Some pollsters will list sub samples even when they are real small and have a large margin of error, some will not.
Using the Kentucky and non-white voters for example. The state is roughly 91% white. The overall CNN/Time poll had an adult base of 1,520 which was 1,336 registered voters and 785 likely voters. So that puts the sub sample of likely non-white voters somewhere around 75 as far as people who were interviewed. Something like that is going to have a very high margin of error so they decided not to even list it.
Some pollsters due this when the sub sample is quite small. So its not saying N/A because they didn't interview anyone in that particular demographic, but rather because the sub sample is very small and that translates into a large margin of error. Some pollsters will list sub samples even when they are real small and have a large margin of error, some will not.
Using the Kentucky and non-white voters for example. The state is roughly 91% white. The overall CNN/Time poll had an adult base of 1,520 which was 1,336 registered voters and 785 likely voters. So that puts the sub sample of likely non-white voters somewhere around 75 as far as people who were interviewed. Something like that is going to have a very high margin of error so they decided not to even list it.
Would you say the sample is not representative of all voter ages so they should not have displayed the data that way and they should have been up front of what it really represents --- old voters in Nevada?
Would you say the sample is not representative of all voter ages so they should not have displayed the data that way and they should have been up front of what it really represents --- old voters in Nevada?
Perhaps or at least mentioned the % of representation of each category. Some exit polls for example will have that. It may show for example a certain demographic subsample is 10% of those who were polled, but have N/A for the subsample because the subsample was too small to have anything other than a high MOE.
One of the reasons why I think the Dems might do better than some people think, is some of these polls seem to be underpolling younger people. Younger voters are obviously going to make up a bit smaller % of the electorate than they did in 2008, but some of these drop offs really don't make any sense For example Pennsylvania, 35% of voters were under 40 in 2008, the latest Morning Call Tracking Poll had them at 12%. Now its not going to be 35% this year, its surely going to be lower than that, but no way in hell is it only going to be 12%.
Some pollsters due this when the sub sample is quite small. So its not saying N/A because they didn't interview anyone in that particular demographic, but rather because the sub sample is very small and that translates into a large margin of error. Some pollsters will list sub samples even when they are real small and have a large margin of error, some will not.
Using the Kentucky and non-white voters for example. The state is roughly 91% white. The overall CNN/Time poll had an adult base of 1,520 which was 1,336 registered voters and 785 likely voters. So that puts the sub sample of likely non-white voters somewhere around 75 as far as people who were interviewed. Something like that is going to have a very high margin of error so they decided not to even list it.
Some pollsters due this when the sub sample is quite small. So its not saying N/A because they didn't interview anyone in that particular demographic, but rather because the sub sample is very small and that translates into a large margin of error. Some pollsters will list sub samples even when they are real small and have a large margin of error, some will not.
Using the Kentucky and non-white voters for example. The state is roughly 91% white. The overall CNN/Time poll had an adult base of 1,520 which was 1,336 registered voters and 785 likely voters. So that puts the sub sample of likely non-white voters somewhere around 75 as far as people who were interviewed. Something like that is going to have a very high margin of error so they decided not to even list it.
So what you're saying is that it's not a grand conspiracy? How else to bash the "left-wing" press? I'm sure they'll find a way.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.