Are freeways a complete failure in terms of safety and reliablilty?

[silverback1065]

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

[iBallasticwolf2]

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

[SidS1045]

Think about it. Controlled access highways were touted as being way safer than conventional highways. They were also touted as being a very reliable way to get people and goods from Point A to Point B.

They have failed miserably.

Despite supposedly being safer because so many of the potentially hazardous situations found on convention roads are eliminated, they have an extremely high accident rate, way higher than the conventional roads in my area. I feel like I am in way more danger on freeways than conventional roads. All you have to do to be safe on a freeway is pay attention to two things: The vehicle in front of you, and what's in the other lane if you are changing lanes. People obviously can't handle these very well.

Also, when accidents do happen, people cannot turn around and go another way like on a conventional highway.

There are no statistics to support your view.

In your own state of Texas, in 2014:

Total fatal crashes:  1737
Total fatalities:  1974

Interstates
Crashes:  206
Fatalities:  246

US & State Highways
Crashes:  827
Fatalities:  968

Farm to Market Roads
Crashes:  425
Fatalities:  467

County Roads
Crashes:  247
Fatalities:  257

City Streets
Crashes:  30
Fatalities:  34

Tollway
Crashes:  1
Fatalities:  1

Other Roads
Crashes:  1
Fatalities:  1

Source:  http://ftp.dot.state.tx.us/pub/txdot/trf/crash-statistics/2014/11.pdf

On city streets, there are very few fatalities, but no one is going very fast, so the chances of a fatal accident go down dramatically.  But of the road types where drivers are traveling faster than city speeds, interstate highways are most definitely the safest roads.

[hotdogPi]

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

[kkt]

When I moved to Washington, I had to take the written test but not the driving test.

Freeways generally have been very successful in terms of safety and reliability.  The accident rate per vehicle mile is much lower than on 2-lane or expressways.  It's true that the fatality rate per accident is higher, because of the higher speeds, but the fatality rate per vehicle mile is still lower than the 2-lane roads.

[bzakharin]

  Does anyone else think it's too easy to get a license, and that driver education is extremely lacking?

YES!   In Arizona, once the written test is passed, all one needs to do in order to get a license is practice driving around on surface streets, stopping at stop signs etc. The majority of my driving test was on residential streets. Never even had to look at a freeway. Luckily for me though I had plenty of freeway experience anyway.

So theoretically, someone can get a license with little to no freeway experience,(I'm not saying this is always the case, but I'm sure there are plenty) and I believe that is what causes some of the problems.

_{I didn't even have to parallel park, but I'm willing to overlook that one}

On my test I asked the teacher if I had to parallel park, he asked me if I wanted to and I said no, and he said "then we don't have to." I also only remember getting onto the interstate once and for about only 2 miles.

No interstates or controlled-access highways on NJ tests. In fact, if you can find a DMV sandwiched between limited access roads, you will never leave the facility during the entire test.  It's clearly a lot more important that you know exactly what the penalty is for an nth time DUI than actually knowing how to drive on a freeway. Parallel parking is a must, though.

[CtrlAltDel]

I’ve long thought it would be a good idea to have part of the driving test in a simulator. This way, they’d be able to test various situations that don’t come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can’t do out on the road, either because they’re dangerous or because they just don’t exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers’ abilities.

[Brandon]

It's if you move to another state, then you have to ditch the license you had for the state you last lived in. You must get a license for the state you currently live in by retaking the test.

That's only true for a handful of states.  If you move to Illinois, you do not need to retake any driving test of any sort.  Also true for Colorado, Louisiana, etc.

[Pete from Boston]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

Massachusetts doesn't have the budget for this.

[jakeroot]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

The school I went to in Washington State had a simulator like this.

Massachusetts doesn't have the budget for this.

Does the state run the driving schools?

[iBallasticwolf2]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

Brilliant Idea

[Pete from Boston]

I've long thought it would be a good idea to have part of the driving test in a simulator.

Massachusetts doesn't have the budget for this.

Does the state run the driving schools?

No, it gives the test at Registry of Motor Vehicles offices, using (inexplicably) specially assigned State Police officers.

Hiring civilians instead of police might free up some simulator money, but police rarely cede the most pedestrian of their tasks in this state.

[jeffandnicole]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

The problem with simulators is that they can be modified to make the test harder or easier.  And because of that you'll have people complaining that the computer was set to an expert level in order to make them fail the test.

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

I agree with testing stuff normally found, but not special road features. Since Michigan lefts are mostly found in Michigan, the chances that someone from the other 49 states will encounter one is fairly remote.  And then, they just revert to other common knowledge, such as reading signs to figure out what to do or not do.

And roundabouts, of course.

[silverback1065]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

The problem with simulators is that they can be modified to make the test harder or easier.  And because of that you'll have people complaining that the computer was set to an expert level in order to make them fail the test.

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

I agree with testing stuff normally found, but not special road features. Since Michigan lefts are mostly found in Michigan, the chances that someone from the other 49 states will encounter one is fairly remote.  And then, they just revert to other common knowledge, such as reading signs to figure out what to do or not do.

And roundabouts, of course.

The problem with that logic is most people don't even pay attention to signs at all to begin with. Which is a huge problem.

[1995hoo]

Regarding driver testing, one thing I think is unfortunate is that there seems to be a trend towards longer license validity periods coupled with online renewal. Here in Virginia, your license used to be valid five years, expiring on the last day of your birth month in years when your age ended in "0" or "5" (so your initial license might be valid less than five years). You had to get your vision tested every time you went to the DMV.

Then they kept the same five-year period but allowed you to renew online every other time unless you had two or more moving violations, in which case you had to renew in person because you had to retake the knowledge test (this part had already applied).

Now you can still renew online every other time (subject to the violations part), but licenses are now valid for eight years. That means the DMV checks your vision every 16 years. I am absolutely certain there are a lot of people who never get their eyes checked except when they go to the DMV. Your eyes can change an awful lot over a 16-year period (I know mine sure have, and my awareness of it is what spurred me to get them checked by a doctor, and then to start wearing eyeglasses).

[Rothman]

The problem with that logic is most people don't even pay attention to signs at all to begin with. Which is a huge problem.

Absolutely true.  I wouldn't even be able to conduct the research on this one due to my inherently strong bias -- I just don't understand people who don't read road signs.  I view such drivers as broken and in need of some serious rehabilitation or having their license revoked altogether until they're able to overcome whatever psychosis is occurring to keep them from understanding road signs.

[jeffandnicole]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

The problem with simulators is that they can be modified to make the test harder or easier.  And because of that you'll have people complaining that the computer was set to an expert level in order to make them fail the test.

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

I agree with testing stuff normally found, but not special road features. Since Michigan lefts are mostly found in Michigan, the chances that someone from the other 49 states will encounter one is fairly remote.  And then, they just revert to other common knowledge, such as reading signs to figure out what to do or not do.

And roundabouts, of course.

The problem with that logic is most people don't even pay attention to signs at all to begin with. Which is a huge problem.

False.

If "most people" don't pay attention to signs "at all", then we have a nation of people that fly thru Stop signs all the time.  That doesn't happen.

[silverback1065]

If people actually read signs, they would get lost way less often, and also make safer choices when driving because they would actually know what's going on ahead on the road

[silverback1065]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

The problem with simulators is that they can be modified to make the test harder or easier.  And because of that you'll have people complaining that the computer was set to an expert level in order to make them fail the test.

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

I agree with testing stuff normally found, but not special road features. Since Michigan lefts are mostly found in Michigan, the chances that someone from the other 49 states will encounter one is fairly remote.  And then, they just revert to other common knowledge, such as reading signs to figure out what to do or not do.

And roundabouts, of course.

The problem with that logic is most people don't even pay attention to signs at all to begin with. Which is a huge problem.

False.

If "most people" don't pay attention to signs "at all", then we have a nation of people that fly thru Stop signs all the time.  That doesn't happen.

The statement is really in reference to warning and information signs, not typical signs like stop signs and speed limit signs. If people read exit signs and warning signs life would be better for everyone. I don't think people understand most of these signs.
I person on the south side of Indy recently died because they didn't read or at least understand a sign explaining what the flashing yellow signal was for.

[Rothman]

I've long thought it would be a good idea to have part of the driving test in a simulator. This way, they'd be able to test various situations that don't come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can't do out on the road, either because they're dangerous or because they just don't exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers' abilities.

The problem with simulators is that they can be modified to make the test harder or easier.  And because of that you'll have people complaining that the computer was set to an expert level in order to make them fail the test.

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

I agree with testing stuff normally found, but not special road features. Since Michigan lefts are mostly found in Michigan, the chances that someone from the other 49 states will encounter one is fairly remote.  And then, they just revert to other common knowledge, such as reading signs to figure out what to do or not do.

And roundabouts, of course.

The problem with that logic is most people don't even pay attention to signs at all to begin with. Which is a huge problem.

False.

If "most people" don't pay attention to signs "at all", then we have a nation of people that fly thru Stop signs all the time.  That doesn't happen.

Still, the number of people that don't figure out where they're going until they're at a gore on a properly signed freeway is ridiculously high in my experience.

[Zeffy]

Still, the number of people that don't figure out where they're going until they're at a gore on a properly signed freeway is ridiculously high in my experience.

The most amusing part of when this happens is when they pass the gore, pull over, reverse into the striped area, and proceed to exit. Like, how many times did the signs tell you the exit was there? Oh right, you don't pay attention.

[jeffandnicole]

I’ve long thought it would be a good idea to have part of the driving test in a simulator. This way, they’d be able to test various situations that don’t come up in a humdrum run-of-the-mill road test. The simulation could put you in a skid and test whether you could get out of it. Or have you merge onto the expressway with someone in your blind spot. Things you can’t do out on the road, either because they’re dangerous or because they just don’t exist in that geographical area. That is, testing things more significant than where you hold your hands on the wheel and whether you stop at the stop sign for three mississippis or not would go a long way toward improving drivers’ abilities.

The problem with simulators is that they can be modified to make the test harder or easier.  And because of that you'll have people complaining that the computer was set to an expert level in order to make them fail the test.

the test should be focused on the rules of the road (i.e. who has the row) signs, signals (types and their function), types of roads, and roundabouts and other special road features like Michigan lefts and jughandles.

And roundabouts

And roundabouts

I agree with testing stuff normally found, but not special road features. Since Michigan lefts are mostly found in Michigan, the chances that someone from the other 49 states will encounter one is fairly remote.  And then, they just revert to other common knowledge, such as reading signs to figure out what to do or not do.

And roundabouts, of course.

The problem with that logic is most people don't even pay attention to signs at all to begin with. Which is a huge problem.

False.

If "most people" don't pay attention to signs "at all", then we have a nation of people that fly thru Stop signs all the time.  That doesn't happen.

The statement is really in reference to warning and information signs, not typical signs like stop signs and speed limit signs. If people read exit signs and warning signs life would be better for everyone. I don't think people understand most of these signs.
I person on the south side of Indy recently died because they didn't read or at least understand a sign explaining what the flashing yellow signal was for.

OK, fine, then we will have a nation of people that go straight instead of heeding the curve ahead sign.

As for the ONE person that died, you do understand that hundreds or thousands of others that day alone went thru that light without an issue, right?  And then we can say that's true every day.  If one person died at that intersection over the past year, that's a very low percentage of people (like, 0.000000001%) that didn't understand the sign.

Yes, not everyone may read every single sign out there, or comprehend every single sign.  And they still read Exit signs...they just try to get over too late, or figure out that's the exit they wanted too late. 

Trust me, no one here is a perfect driver.  No doubt you've almost missed your exit too once in a while.

[Rothman]

Nah.  I am a perfect driver.

[1995hoo]

Still, the number of people that don't figure out where they're going until they're at a gore on a properly signed freeway is ridiculously high in my experience.

The most amusing part of when this happens is when they pass the gore, pull over, reverse into the striped area, and proceed to exit. Like, how many times did the signs tell you the exit was there? Oh right, you don't pay attention.

Heh. Sunday afternoon I saw someone stop on the shoulder and then drive across the grass between the road and the exit ramp to go from I-295 in Maryland to the Wilson Bridge.

[cpzilliacus]

Not sure about your specific area, but in other areas, the numbers would go against your assertion.  Using 2013 Minnesota data for example, the crash rate on Interstate highways is the lowest of all the roadway classes, and the fatality rate is tiny compared with other roadway classes.

I agree with what Adam has written above, and assert it holds true across the U.S. and the rest of the world.

Freeway-class roads are generally the safest roads.  Note that there are some known unsafe limited-access roads, generally with older designs and not always built to freeway standards.

Three examples are I-278 in much of New York City, I-76 in Philadelphia (Schuylkill Expressway, sometimes called the Sure-Kill for that reason) and D.C. 295 in the District of Columbia.

All are relatively old, and have segments that were designed and engineered with features that would never gain approval today.