Map with both people and land

[Beet]

http://www.bopnews.com/archives/002306.html#2306

I thought this was quite insightful.

[Beet]

Why? The point is that the vast majority of the country is GOP, not the majority of people in it.

It speaks it for itself:

"President Bush's votes were widely dispered across the country and included many sparsely populated rural counties."

"Sen. John F Kerry's votes were concentrated largely in urban centers, which traditionally are Democratic strongholds."

[Beet]

Sure, but I think that's pretty well known

That's true. Though its very well hidden in Dave's maps through no fault of his own, I doubt he can develop 3-D imaging by himself.

[Beet]

It is unlikely that Kerry would win a large city and not win the county that it was in.

Note that not all cities exist solely in one county.  Bush for example won Richmond County, NY.  This is one of the 5 counties of New York City.

Ah, correct. My question is, what difference does showing the big Kerry votes in the cities really have to do with anything?

Nothing, except showing Kerry votes. Yes, he did get votes. Maybe you find that offensive.

I just thought it might be interesting, it certainly struck me when I first saw it, to have a graphical representation of reality. That is what the whole site is about right? Apparently some people are offended by the fact that Kerry got votes and any map that shows it must be bitched about.

[Beet]

It is unlikely that Kerry would win a large city and not win the county that it was in.

Note that not all cities exist solely in one county.  Bush for example won Richmond County, NY.  This is one of the 5 counties of New York City.

Ah, correct. My question is, what difference does showing the big Kerry votes in the cities really have to do with anything?

Nothing, except showing Kerry votes. Yes, he did get votes. Maybe you find that offensive.

I just thought it might be interesting, it certainly struck me when I first saw it, to have a graphical representation of reality. That is what the whole site is about right? Apparently some people are offended by the fact that Kerry got votes and any map that shows it must be bitched about.

I don't find it "offensive" but I do question if it gives any real and new information.  I kinda figured out that Democrats tend to do well in cities.  I expected Kerry to win large totals in large cities.  I take it you realized that as well.

I also realize that Republicans tend to do well in rural areas and win a large majority of the counties. Does that mean all the maps on Dave's site are useless to me?

[Beet]

True, and its the margins that matter in the end. This map shows the margins which a flat map does not do. It thus gives a much clearer idea of which counties played and important role.

[Beet]

Read my previous post. If I were a political strategist or analyst, the 3-D map would be definitely more informative.

[Beet]

Not really.   It shows where one  candidate ran up vote totals.  From the 3-D map, I cannot see the other sections of PA, for example, where Kerry won, other than Pittsburgh and Phila.  I couldn't tell that there was a city, called San Diego, in southwestern part of CA.  It doesn't provide any useful information.

I don't really understand what you're trying to say here. There are supposed to be 2 maps there, if your computer isn't showing it, that's unfortunate, but it's all in the same graphic.

You can see all of the counties where Kerry won and all of the counties where Bush won. It shows where one candidate ran up vote totals yes, but it also shows where the other candidate ran up vote totals. You can see where both candidates ran up totals.

You can see ALL the section of PA. The entire state is covered on both maps. You can see San Diego as well. You can see that Bush won San Diego and you can see that he won it by a much smaller margin than he won nearby Orange county. So I don't really understand what you mean when you say you cannot see this or that.

Secondly, even if you could not see those two places, I don't see what impact it has on the rest of the graphic which is there. Suppose the map maker accidentally left out San Diego and PA except for Philadelphia and Pittsburgh. You could still see every other part of the county.

And even if you could only see one candidate, you can still see the margins they gained and where they gained them from; and how their vote was distributed.

None of your objections seem to be factually accurate, at least from the way the graphic showed up on my computer. However, even if they were all accurate, your last statement would still be false.

So I guess I'm a little confused.

[Beet]

There is no real comparative value from these maps.  I'd want one map that showed the comparisons between the two, including statewide total, if I were using this.

Well, the information is the same whether it is presented on one map or two maps.

You can find statewide totals on a table or bar chart, no need to use a map.

[Beet]

You must be logged in to read this quote.

You must be logged in to read this quote.

[/quote]

As far as I can tell, each county was one by EITHER Bush OR Kerry. No county was won by both candidates simulataneously.

You must be logged in to read this quote.

The higher the bar, the higher the relative strength. The lower the bar, the lower the relative strength. That's what it means. It is also labelled.

You must be logged in to read this quote.

If that's all the map reveals, I guess the Washington Post could have just printed the sentence "KERRY WON IN CITIES" instead of creating that map? Perhaps you should e-mail the editors to enlighten them about their mistake for future elections?

Engineer-

1. the map is labelled "height shows the margin of victory"

2. You can look up that information in a table. There are over 3,000 counties. If each one was labelled on a map, it would be so impossibly cluttered that you wouldn't be able to read it easily. Hence, that is why there are tables. In fact, you can look up that information on this website.

3. Probably neither 1% OR 10%. Candidates rarely win by exact percentage points. Again, you can look that up on this website.

4. Actually, it's total number of votes, not percentages, that counts. If you win 90% in county A and just 45% in county B, in a state with only 2 counties, you have won the average percents. But you did not necessarily win the state. However, if you won more absolute votes than any other person, that generally counts as a win on the state level. Hence, the total number of votes in the county is more helpful than percentages.

5. The total number of people is also irrelevant. That in no way factors into the calculus.

6. In no county, as far as I know, did all the people vote for Kerry. But this still doesn't make total numbers of votes less important than county percentages. It's more important to win a decent percentage in a county with a lot of people than a high percentage in a county with very few people.

******

If there is any additional confusion, feel free to ask. Also, feel free to check out these sites:

How to read maps

Review of bar graphs

[Beet]

You must be logged in to read this quote.

You must be logged in to read this quote.

As far as I can tell, each county was one by EITHER Bush OR Kerry. No county was won by both candidates simulataneously.

You must be logged in to read this quote.

The higher the bar, the higher the relative strength. The lower the bar, the lower the relative strength. That's what it means. It is also labelled.

You must be logged in to read this quote.

If that's all the map reveals, I guess the Washington Post could have just printed the sentence "KERRY WON IN CITIES" instead of creating that map? Perhaps you should e-mail the editors to enlighten them about their mistake for future elections?

Wrong again.  If I look at Phila for example, I can see that had about 500,000 votes.  It's hard for me to know how many Bush had.  I don't 700,000 votes case, or 1,400,000 votes cast.   I really provides very little information that is useful.
[/quote]

Your sentences are not making any sense. I can barely make out what you're trying to say. But I'm guessing that you're trying to say it would be helpful to know how many votes Bush had in Philadelphia, for example, given the map shows Kerry won it by 500,000.

Let me ask you this: Does it change the popular vote balance in the state if Bush won 200,000 and Kerry won 700,000 or if Bush won 2,200,000 and Kerry won 2,700,000?

In both instances, Philadelphia contributes 500,000 to Kerry's margin. The results in all other counties are equal. The person who wins a positive margin wins the state. Hence, in both cases, Philadelphia county has the same impact, and it makes no difference.

[Beet]

Your sentences are not making any sense. I can barely make out what you're trying to say. But I'm guessing that you're trying to say it would be helpful to know how many votes Bush had in Philadelphia, for example, given the map shows Kerry won it by 500,000.

Let me ask you this: Does it change the popular vote balance in the state if Bush won 200,000 and Kerry won 700,000 or if Bush won 2,200,000 and Kerry won 2,700,000?

In both instances, Philadelphia contributes 500,000 to Kerry's margin. The results in all other counties are equal. The person who wins a positive margin wins the state. Hence, in both cases, Philadelphia county has the same impact, and it makes no difference.

Sorry, I'm capable of thinking faster than I can type.  :-)

It's the intracounty comparison.  The map would be useful if it would show the relative amount of the vote in each county.  For example, knowing that Kerry had about 500,000 votes in Phila is pretty meaningless unless I know how many votes Bush had there as well.  If he had 80,000 votes, that's one thing.  If he had 200,000 votes, it would mean something else entirely.

Well first thing, you don't know how many votes Kerry had in Phila. You know how many more votes Kerry had than Bush. I think that addresses your entire complaint. Thanks.

[Beet]

No...
The whole point is, at least if I'm getting it correctly,

Well, he's also made many confusing statements, such as-

1- the map provides no useful information
2- he cant tell where one candidate lost
3- two candidates may have won the same place simultaneously

You must be logged in to read this quote.

That's what I thought he was saying, which I discussed, but his last post, just before the one I replied to which you then replied to, suggest that he's saying there's a difference between Bush having 80,000 and 200,000 given that Kerry has 500,000. What you're saying is that there's a difference between Bush having 80,000 and 200,000 given that Kerry has a margin of x amount.

Regarding the difference between big margins in small counties and small margins in big counties, it really makes no difference for the candidate. What matters is how many more votes they got than the other candidate in a particulate place. It doesn't matter very much if Kerry got 1,000,000 votes in Miami-Dade if Bush got 950,000 votes. It's the same as if he got 100,000 in a smaller county if Bush only got 50,000. In terms of electoral strategy, population matters only in terms of the fact that you need smaller percentage wins in order to get bigger margins. In anthropolog or sociology, populations would be important as an end to themselves, but in political elections, it is only vote margins that matter in the end, since margins are the same thing as majorities. To show that information on a map would be redundant and force the viewer to estimate in his or her own mind the differences. It would just cause too much clutter and not be dramatic enough. That's why I think they chose it this format and not the other format.

[Beet]

You must be logged in to read this quote.

Well, yes, if the maps would show the amount that that Kerry or Bush had (or that they lost by), it would be a much more useful map.  This one doesn't.

You must be logged in to read this quote.

You must be logged in to read this quote.

It does show these things. It does NOT show absolute vote numbers.

[Beet]

You must be logged in to read this quote.

Well, yes, if the maps would show the amount that that Kerry or Bush had (or that they lost by), it would be a much more useful map.  This one doesn't.

You must be logged in to read this quote.

You must be logged in to read this quote.

It does show these things. It does NOT show absolute vote numbers.

It does show the Kerry plurality, i.e. how many votes Kerry had over Bush; I went back and looked.  It doesn't however show the amount Kerry lost by on the same map.

But that's WHY there are two maps.

You must be logged in to read this quote.

No...

You must be logged in to read this quote.

Then what does, pray?

[Beet]

It is hard to match up the two maps.  If all the information was on one map, it would be useful.

Yes.  Looking at Kerry map of PA, I cannot tell how much he lost Berks county, for example.  May just a little, or maybe enough to take a big bite out of his totals in Phila.

What does is how well Kerry ran in a lot of those red counties; you cannot tell that from the Kerry map.  Sorry, but it just isn't clear enough.

This map tells you that Kepy tended to run up large vote totals in small areas, cities.  I kinda suspected that.

Once again, populations are just a means to an ends. Having a big margin in a small county or a small margin in a big county if the margins are the same in absolute numbers comes out the same. A campaign strategist would be indifferent between the two, all other things equal. The goal isn't to see which areas have more people living in them but which areas are relatively more Republican or Democratic.

As for the rest... it just involves looking at both maps. If that's too hard, I don't know what else to say.

[Beet]

Plus, including more data would clutter it too much and dilute its impact with information that people already know; that there are big cities where Chicago, NY, etc. etc. are. It's better to get the message across in the clearest and most simple way. Again, if we were studying population, it would be a different story, but we aren't.