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.
<sigh> ok, I guess I have to take a little time to write a long boring explanation. Warning: most of you will probably want to skip to the next post.
I'm sorry that you took offense at having a flaw in the idea of using cars as a measuring device pointed out. Please know that YOU are not under scrutiny, a method for data analysis is. If you don't want people to comment on methodology, don't ask them direct questions such as "What would you suggest as an alternative?"
You say that you were "reflecting aloud, challenging my own concepts, and trying to understand where the differences between mine and others came from."
That's great--but when people do that, they should expect that others will help you in your quest to challenge those concepts by pointing out flaws in logic or methodology. Don't look at such comments as personal attacks, but rather an attempt to help you answer your own questions.
You asked for my suggestion for an alternative. I'm sorry you didn't care for the suggestion I gave you. I still think my suggestion is the best option, but if you really want a way to accurately prove the frequency of pedestrians, the only other ones that I can (hesitantly) suggest is to either set up a video camera running 24-7, or, if you want to use observations posted by people on this forum, you could try asking volunteers to walk the exact same stretch that you mentioned in the OP, at different hours of the day--and to do so daily for a year.
At that point you might have enough to create a statistic that proves how frequently people walk--but it will only be good for that particular stretch of road. I'm not sure if that would really be worth the effort, but that's the only approach I can think of that would (to some degree) prove such a thing and also involve collecting observations from your fellow posters on city data.
The thing is, an internet forum like this is just not a good place to compile statistics--it's a place to compile opinions, which are also extremely valuable, but are not the same as statistics. I have this thing about statistics. Many people like to manipulate flimsily constructed "statistics" to somehow prove their opinion is a fact, therefore I think it's important to point out potential flaws so that the end result be as accurate as possible. It's not an attack on you--not at all. It's a desire to see statistics that are less susceptible to manipulation.
Please do not take the word "manipulate" to be a personal accusation--it's not aimed at you. Anyone can easily manipulate statistics, especially when the methodology results in an assortment of variables that can alter the results.
For example, watch how easily I can manipulate the observations this thread has collected so far. Warning: here's where it will probably get really long and boring (if it hasn't already, LOL).
OK, for those who are interested and still reading, let's consider a detailed examination of my commute to work. I have a variety of routes I take--if I compare cars to pedestrians, each one can be used to manipulate the data in several different directions, depending on what I might find amusing. I can choose to walk along a busy parkway. Comparing pedestrians to cars for this route, I can make it appear that there are very few pedestrians here and a huge number of cars. Or, I can choose to walk side streets through a quiet neighborhood. I would see 4-5 more pedestrians here (mostly moms with strollers, people walking dogs, etc.) and significantly less cars--so if comparing pedestrians to cars was the measurement, the same part of town has a completely different result--yet the two routes are just one block apart.
I could also choose to walk down a street that has several school stops. Boy would I be able to report a huge number of pedestrians then--plus cars are are relatively few since the locals know not to drive when it's "bus time" since traffic comes to a crawl when school buses are out. Or I could really skew the statistics by choosing to walking down Brockman Farm Road, which is essentially an abandoned road (it might even be officially abandoned by now). The only vehicle you see on it anymore is the occasional service truck, but since it's not used by cars you see lots of pedestrians--probably the most pedestrians of any of my routes.
Depending on which of the above routes I chose, I could easily interpret the observations to "prove" all sorts of things. For example, comparing the results from you, normie, michgc, and myself I can "prove" that there are significantly more pedestrians as you proceed west in northern Virginia. I don't happen to think that's true, but the data posted so far could be manipulated to prove that. That's why I don't think comparing pedestrians to cars is useful.
Most important: just because I found a flaw with one facet of this survey does not mean I disagree with the entire idea. To the contrary, I think the idea of reporting how many pedestrians you happen to see is useful and interesting. I don't think you can use these observations to compile statistics or "proof" of how frequently people walk. You cannot use them to "prove" an opinion is fact, but observations can still tell you a lot.
I'm sorry that you took offense at having a flaw in the idea of using cars as a measuring device pointed out. That's not an attack on you, it's a comment about your method for data analysis. If you don't want people to comment on methodology, don't ask them direct questions such as "What would you suggest as an alternative?"
You say that you were "reflecting aloud, challenging my own concepts, and trying to understand where the differences between mine and others came from."
That's great--but if you want to do that, you should expect that others will help you challenge those concepts by pointing out flaws in logic or methodology. Don't look at such comments as personal attacks, but rather an attempt to help you in your quest to understand.
You asked for my suggestion for an alternative. I'm sorry you didn't care for the suggestion I gave you. I still think my suggestion is the best option, but f you want another one, the only other one that I can (hesitantly) suggest is to ask volunteers to walk the exact same stretch that you mentioned in the OP, at different hours of the day--and to do so daily for a year.
At that point you might have enough to create a statistic that proves how frequently people walk in that particular stretch of road. I'm not sure if that would really be worth the effort, but that's the only approach I can think of that would (to some degree) prove such a thing and also involve collecting observations from your fellow posters on city data.
The thing is, an internet forum like this is just not a good place to compile statistics--it's a place to compile opinions, which are also extremely valuable, but are not the same as statistics. I have this thing about statistics. Many people like to manipulate flimsily constructed "statistics" to somehow prove their opinion is a fact, therefore I think it's important to point out potential flaws so that the end result be as accurate as possible. It's not an attack on you--not at all. It's a desire to see statistics that are less susceptible to manipulation.
Please do not take the word "manipulate" to be a personal accusation--it's not aimed at you. Anyone can easily manipulate statistics, especially when the methodology results in an assortment of variables that can alter the results.
For example, watch how easily I can manipulate the observations this thread has collected so far. Warning: this will probably get long and boring (if it hasn't already, LOL).
OK, for those who are interested and still reading, let's consider a detailed examination of my commute to work. I have a variety of routes I take--if I compare cars to pedestrians, each one can be used to manipulate the data in several different directions, depending on what I might find amusing. I can choose to walk along a busy parkway. Comparing pedestrians to cars for this route, I can make it appear that there are very few pedestrians here and a huge number of cars. Or, I can choose to walk side streets through a quiet neighborhood. I would see 4-5 more pedestrians here (mostly moms with strollers, people walking dogs, etc.) and significantly less cars--so if comparing pedestrians to cars was the measurement, the same part of town has a completely different result--yet the two routes are just one block apart.
I could also choose to walk down a street that has several school stops. Boy would I be able to report a huge number of pedestrians then--plus cars are are relatively few since the locals know not to drive when it's "bus time" since traffic comes to a crawl when school buses are out. Or I could really skew the statistics by choosing to walking down Brockman Farm Road, which is essentially an abandoned road (it might even be officially abandoned by now). The only vehicle you see on it anymore is the occasional service truck, but since it's not used by cars you see lots of pedestrians--probably the most pedestrians of any of my routes.
Depending on which of the above routes I chose, I could easily interpret the observations to "prove" all sorts of things. For example, comparing the results from you, normie, michgc, and myself I can "prove" that there are significantly more pedestrians as you proceed west in northern Virginia. I don't happen to think that's true, but the data posted so far could be manipulated to prove that. That's why I don't think comparing pedestrians to cars is useful.
Most important: just because I found a flaw with one facet of this survey does not mean I disagree with the entire idea. To the contrary, I think the idea of reporting how many pedestrians you happen to see is useful and interesting. I don't think you can use these observations to compile statistics or "proof" of how frequently people walk. You cannot use them to "prove" an opinion is fact, but observations can still tell you a lot.
1. I did not ask for alternatives for data collection - I am well aware that my method is not comparable to a scientific study. I meant an alternative way to normalize, and determine what is a "main drag".
2. I am not trying to prove anything. I am offering observations, which we do here all the time, on a range of issues. I made it a bit more quantitative than usual, because i thought saying "a lot" or "a few" was causing confusion. I did NOT mean to imply that this was a substitute for accurate counts of pedestrians. (BTW, I think the latter is a good idea, but its a govt responsibility)
3. despite statements above, I have found this discussion hostile, and my own posts repeatedly mischarecterized. I found it stated that I think there are NO walkers in the NoVa suburbs (something I never said, and do not believe) and that recreational walkers "dont count" (something I never said)
"Or, I can choose to walk side streets through a quiet neighborhood. I would see 4-5 more pedestrians here (mostly moms with strollers, people walking dogs, etc.) and significantly less cars--so if comparing pedestrians to cars was the measurement, the same part of town has a completely different result--yet the two routes are just one block apart. "
yes, thats what I was getting at in my discussion of why my normalization metric might not work for areas where most walking is recreational. That unfortunately was misread as a claim that recreational walkers "dont count"
I am curious how much recreational walking on neighborhood streets varies. I can present little evidence myself - my DW and I live convenient to a medium size park with trails, and to the Accotink trail - when we are walking at what seems to be the peak walking hour, around 6PM, or earlier on weekends, we tend to walk trails. We usually only do recreational walks on neighborhood streets after dark, when we prefer to avoid the trails. But thats usually after most recreational walkers have gone inside, I think.
Another interesting question might be the relative appeal of the nature trails vs neighborhood streets for recreational walkers. The LRT corridor happens to be close to one of the regions (the nation's?) premier metropolitan trails - the Cross County Trail. It may draw off people whom in some other areas might walk in their neighborhoods (other than the W&OD, im not familiar with the LC trails)
If you're interested in pedestrians along "main drags" you may not find that many. It's true that you may not find many people walking along Route 7--but check out the side streets. That's where you'll find many more pedestrians. In my own example, I can choose to walk to my job by taking the sidewalk along a parkway. But why would I do that when just one block north I can walk down a much more pleasant side street?
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.
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.
