U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Urban Planning
 [Register]
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.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
Reply Start New Thread
 
Old 05-27-2014, 04:13 AM
 
1,709 posts, read 1,676,130 times
Reputation: 1838

Advertisements

Quote:
Originally Posted by Katiana View Post
Research!
We use logic to make sense of research!

Also, you can do all the research you want, but logic will lead you to a faster conclusion as long as your premises are sound and your conclusion follows properly. For example, I can spend my entire life researching that not breathing is correlated with death; or I could use common sense and logic to see that not breathing will lead to death.
Reply With Quote Quick reply to this message

 
Old 05-27-2014, 05:24 AM
nei nei won $500 in our forum's Most Engaging Poster Contest - Thirteenth Edition (Jan-Feb 2015). 

Over $104,000 in prizes has already been given out to active posters on our forum and additional contests are planned
 
Location: Long Island / NYC
45,990 posts, read 41,998,698 times
Reputation: 14810
Quote:
Originally Posted by Katiana View Post
Oh, knock it off! You don't understand that the U of I researcher even said CORRELATION DOES NOT EQUAL CAUSATION!
No, but how could one prove causation for this topic? Only way a study could work for this I can think would be to:

1) Remove other variable (income, other demographic variables)
2) Find a correlation
3) Use logic to attribute a cause

Even though imperfect, there is no way to find a cause directly.
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 06:00 AM
 
Location: Ft. Myers
17,628 posts, read 11,189,868 times
Reputation: 37672
Quote:
Originally Posted by Escort Rider View Post
Hmmm..... I never noticed that before, although I cannot claim you are wrong. Have you noticed that only in parking lots, or on the road as well?

Good thing this forum is anonymous, otherwise the fat lib people would be after you!
Hehe, obviously, my post was a little tongue in cheek, but I really have noticed that if I am in some parking lot, like at the grocery store, and some car comes barreling through, lots of times the driver has some chubby cheeks. I've almost been mowed down more than once by one of them.

Don
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 08:12 AM
 
Location: Anchored in Phoenix
1,942 posts, read 3,921,914 times
Reputation: 1767
I know skinny people who drive 60 miles one way to work five days a week. I know amorbidly obese guy who drives 45 miles. The skinny guy probably sleeps 4 hours a night, is the father of five kids, and goes running. The morbidly obese guy starts eating cheetos before 8am, is the father of two girls.

It all depends. I know I would be very obese if I had kids and drove 30 miles each way. But I do not have kids. I always rent. So I move to where my next job is and I make sure there is a good workout facility for my one hour swims. Time on the road takes away time for working out. Those people who are sedentary, whether they commute or not have to eat fewer calories to stay thin. It's a trade off. There are apps that are free and keep track of your calories, nutrient balance, and exercise. They demonstrate the laws of physics. If you add more fuel than you burn off you will get fat. If you burn off more fuel than you add you will lose fat. There really is not any medical excuse. I am hypothyroid and should be morbidly obese. Instead I have good definition at age 55, although yes I take synthroid
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 08:35 AM
 
Location: Pittsburgh, PA (Morningside)
12,436 posts, read 11,937,287 times
Reputation: 10542
I'd like to know how many more steps, all things considered, someone who commutes via mass transit gets than someone who uses their car.

Generally speaking, after all, people won't walk to a bus stop or train station if it's more than a 10-15 minute walk from their house. Then they need to walk to their place of work when they get to Downtown (or wherever), so at maximum, they might be adding 30 minutes of moderate walking activity daily, as opposed to the five minutes or so of walking someone who goes from garage to surface lot in an office park might have.

Obviously if you live in a fully walkable neighborhood, the amount of physical activity would potentially be much higher, as you might walk for hours in evening and on the weekends as part of shopping/socialization. But I don't see any way simply moving more people towards mass transit would help reduce obesity.
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 08:43 AM
Status: "Summer!" (set 22 days ago)
 
Location: Foot of the Rockies
87,014 posts, read 102,634,943 times
Reputation: 33082
Quote:
Originally Posted by nei View Post
No, but how could one prove causation for this topic? Only way a study could work for this I can think would be to:

1) Remove other variable (income, other demographic variables)
2) Find a correlation
3) Use logic to attribute a cause

Even though imperfect, there is no way to find a cause directly.
Exactly! That's why some of these "armchair urbanists" drive me crazy sometimes! Just as likely, there is no ONE cause for obesity. Oh, well all know it's exercise and eating, but what causes us to overeat and underexercise is multifactorial.

Quote:
Originally Posted by eschaton View Post
I'd like to know how many more steps, all things considered, someone who commutes via mass transit gets than someone who uses their car.

Generally speaking, after all, people won't walk to a bus stop or train station if it's more than a 10-15 minute walk from their house. Then they need to walk to their place of work when they get to Downtown (or wherever), so at maximum, they might be adding 30 minutes of moderate walking activity daily, as opposed to the five minutes or so of walking someone who goes from garage to surface lot in an office park might have.

Obviously if you live in a fully walkable neighborhood, the amount of physical activity would potentially be much higher, as you might walk for hours in evening and on the weekends as part of shopping/socialization. But I don't see any way simply moving more people towards mass transit would help reduce obesity.
Yes, I believe the standard is that people are willing to walk 1/4 mile to the bus stop. That's hardly 10 minutes, probably more like 5. Now of course, some walk more, but they are the exceptions. If one works downtown, one can probably get off the bus within a block of the office. I probably walk more than that AT work!
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 08:51 AM
nei nei won $500 in our forum's Most Engaging Poster Contest - Thirteenth Edition (Jan-Feb 2015). 

Over $104,000 in prizes has already been given out to active posters on our forum and additional contests are planned
 
Location: Long Island / NYC
45,990 posts, read 41,998,698 times
Reputation: 14810
Quote:
Originally Posted by Katiana View Post
Exactly! That's why some of these "armchair urbanists" drive me crazy sometimes! Just as likely, there is no ONE cause for obesity. Oh, well all know it's exercise and eating, but what causes us to overeat and underexercise is multifactorial.
Not quite what I'm saying. Obviously studies find connections all the time, but proving a direct cause is very difficult. Easier is finding a correlation and finding a mechanism. Off topic, but the lead-violent crime link is a very hard to prove connection that correlation gives a strong hint for.

Of course there's no one cause, I think people have said driving rather than walking for transportation could be one of them. But is it a significant one?

Quote:
Yes, I believe the standard is that people are willing to walk 1/4 mile to the bus stop. That's hardly 10 minutes, probably more like 5. Now of course, some walk more, but they are the exceptions. If one works downtown, one can probably get off the bus within a block of the office. I probably walk more than that AT work!
Depends on the type of transit. I'd care more about total trip time, than the walking specific portion. Express buses and trains people are willing to walk further. Here's a rather thorough source on the topic:

Human Transit: basics: walking distance to transit
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 08:56 AM
nei nei won $500 in our forum's Most Engaging Poster Contest - Thirteenth Edition (Jan-Feb 2015). 

Over $104,000 in prizes has already been given out to active posters on our forum and additional contests are planned
 
Location: Long Island / NYC
45,990 posts, read 41,998,698 times
Reputation: 14810
Here's my guess:

For walking for transportation to have any influence, it would probably be for someone who walks for most trips, not just does a short walk for a transit stop. Furthermore, you'd have to look at a sufficiently walkable place. Most walkable places people want to drive do. Or driving is still more convenient, so most still drive. In a place like NYC, the majority don't own cars, and even many of those that do still walk for most trips due to congestion and lack of convenient parking. There some other urban neighborhoods like that in the US, but not many. These would be the only place walkabilty could have a significant health impact.

Other good comparison might be for college students. In some large campuses, most students walk to class from their place, often a long walk. Others are more car-oriented. Demographics have a lot in common, so it might be a good environment for comparison, if someone thought to do a study on it.
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 11:02 AM
 
Location: Vallejo
14,072 posts, read 16,094,154 times
Reputation: 12647
Quote:
Originally Posted by nei View Post
Here's my guess:

For walking for transportation to have any influence, it would probably be for someone who walks for most trips, not just does a short walk for a transit stop. Furthermore, you'd have to look at a sufficiently walkable place. Most walkable places people want to drive do. Or driving is still more convenient, so most still drive. In a place like NYC, the majority don't own cars, and even many of those that do still walk for most trips due to congestion and lack of convenient parking. There some other urban neighborhoods like that in the US, but not many. These would be the only place walkabilty could have a significant health impact.

Other good comparison might be for college students. In some large campuses, most students walk to class from their place, often a long walk. Others are more car-oriented. Demographics have a lot in common, so it might be a good environment for comparison, if someone thought to do a study on it.
Or you could just look at correlation. Most college students walk or bicycle places, at least a far far greater proportion than the general population walk places. College students have a lower rate of obesity than the general populace. Therefore, walking/cycling places would seem to have a very high causal relationship to not being obese. Of course, that ignores some rather important things. College students tend to be more affluent than the population at large and socioeconomics is correlated to obesity, more educated people tend to be less obese than less educated people, younger people tend to be less obese than older people.

It's actually not that difficult to at least attempt to infer causal relationships rather than just correlations. There's a whole field that does just that. It's called statistics. That's what regression analysis attempts to do. You're still, of course, inferring a causal relationship by calculating the effect from each of several variables. It isn't perfect by any means. If you miss a variable or have bad data, your results are off. Still, whenever you see a study that does not even attempt to do regression analysis, you know it's basically either complete and utter garbage or just never was intended to look at causation. Not every study is attempting to show causation.

The problem is that journalists are generally completely ignorant of anything statistics and don't get that. When you dilute the pool further and remove the editors and look at bloggers, it gets even worse. That's part of what an editor is supposed to do is fact check. Nobody expects journalists to be experts on the things they are reporting stories about. That said, a basic freshman statistics course really should be mandatory for journalism/communication majors. That really wasn't the case here since the article went out of its way to explicitly point out that the study did not attempt to control for any other variables. Given it being published in The Economist, the audience is expected to understand what that means I presume.
Reply With Quote Quick reply to this message
 
Old 05-27-2014, 12:59 PM
Status: "Summer!" (set 22 days ago)
 
Location: Foot of the Rockies
87,014 posts, read 102,634,943 times
Reputation: 33082
Quote:
Originally Posted by Malloric View Post
Or you could just look at correlation. Most college students walk or bicycle places, at least a far far greater proportion than the general population walk places. College students have a lower rate of obesity than the general populace. Therefore, walking/cycling places would seem to have a very high causal relationship to not being obese. Of course, that ignores some rather important things. College students tend to be more affluent than the population at large and socioeconomics is correlated to obesity, more educated people tend to be less obese than less educated people, younger people tend to be less obese than older people.

It's actually not that difficult to at least attempt to infer causal relationships rather than just correlations. There's a whole field that does just that. It's called statistics. That's what regression analysis attempts to do. You're still, of course, inferring a causal relationship by calculating the effect from each of several variables. It isn't perfect by any means. If you miss a variable or have bad data, your results are off. Still, whenever you see a study that does not even attempt to do regression analysis, you know it's basically either complete and utter garbage or just never was intended to look at causation. Not every study is attempting to show causation.

The problem is that journalists are generally completely ignorant of anything statistics and don't get that. When you dilute the pool further and remove the editors and look at bloggers, it gets even worse. That's part of what an editor is supposed to do is fact check. Nobody expects journalists to be experts on the things they are reporting stories about. That said, a basic freshman statistics course really should be mandatory for journalism/communication majors. That really wasn't the case here since the article went out of its way to explicitly point out that the study did not attempt to control for any other variables. Given it being published in The Economist, the audience is expected to understand what that means I presume.
This bold in particular. Obesity increases with age.

I'm not impressed by economic analyses of health issues by economists or others who have no health care background. There is a very popular pregnancy book out now, written by an economist who thinks she has the Holy Grail, that says it's OK to drink a drink a day the last 6 months of pregnancy! She has done some sort of "analysis" and has come to that conclusion despite the fact that 40 years of research about Fetal Alcohol Syndrome says otherwise. I don't mean to hijack, just giving an example.
Reply With Quote Quick reply to this message
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.


Reply
Please update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:

Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Urban Planning
Follow City-Data.com founder on our Forum or

All times are GMT -6.

2005-2019, Advameg, Inc. · Please obey Forum Rules · Terms of Use and Privacy Policy · Bug Bounty

City-Data.com - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 - Top