How do you handle probabilistic knowledge?

[Ivor the Engineer]

No events in life have a probability of 0 or 1 of occurring. Statistical information is rarely available about us as individuals, so we have to rely on averages of groups and assume we share many of the significant characteristics of a particular group if we wish to use the data in our decisions.

What behaviours (if any) do you engage in where you expect to beat the odds?

Is this an irrational way to behave?

 

[Mashuna]

Due to my crippling fear of any kind of injury, I used to stay in bed with the covers over my head. Then I heard that most accidents happen within one mile of your home, so I wander the streets 24x7, sleeping in doorways. That way, I'll never be hurt and will probably live forever.

 

[SezMe]

What behaviours (if any) do you engage in where you expect to beat the odds?

I have a crappy diet. Not enough fruits and vegatables. Too much meat.

Is this an irrational way to behave?

Yes.

 

[Ivor the Engineer]

Here's some info. on speed and accidents.

http://www.erso.eu/knowledge/content/20_speed/speed_is_a_central_issue_in_road_safety.htm

One particularly interesting bit:

Speeding: societal vs. individual consequences

The negative road safety outcomes of high speed are evident at an aggregate level. At the level of the individual driver, the risk of an accident is very small; at higher speeds the risk is higher, but still very small. Hence, an individual driver will hardly ever experience the safety consequences of excess speed. More or less the same applies for the environmental effects of speeding. These are also noticeable at an aggregate level, but hardly at the individual level (possibly with the exception of fuel consumption).

Contrary to the disadvantages, the advantages of higher speeds are experienced at the individual level. Individual advantages include just reaching traffic lights while still green, (subjectively) shorter journey times, thrill and enjoyment of speed or speeding.

This contradiction between societal and individual consequences makes persuading drivers of the value of speed management a difficult mission.

ETA: What has slowed me down is having a display of mpg on the dash. Whereas I would regularly travel at 80-90mph on motorways in my previous car with no such display, I now average 60-75mph.

 

[Ivor the Engineer]

Can people with more driving experience and/or higher performance (i.e. better braking) vehicles drive faster than average without increasing their risk above the group?

 

[Professor Yaffle]

Maybe some can, but it shouldn't be down to an individual to decide whether they are in that group. Many people think they are more skilled than they are.

http://www.apa.org/journals/features/psp7761121.pdf

 

[jdp]

Here's some info. on speed and accidents.

One particularly interesting bit:



ETA: What has slowed me down is having a display of mpg on the dash. Whereas I would regularly travel at 80-90mph on motorways in my previous car with no such display, I now average 60-75mph.

Wow, 80-90 is pretty excessive on most highways around me. I suppose the MPG display brings the relationship a bit closer to the wallet and few things change habits like that.

 

[Jeff Corey]

That is similar to the Lake Wobegon Effect. http://changingminds.org/explanations/theories/lake_woebegon.htm

 

[JoeEllison]

Maybe some can, but it shouldn't be down to an individual to decide whether they are in that group. Many people think they are more skilled than they are.

As far as driving is concerned, I'm pretty sure I'm pretty good at it. I drive the speed limit and avoid tailgating because I don't trust the rest of you wackos!

 

[Ivor the Engineer]

Wow, 80-90 is pretty excessive on most highways around me. I suppose the MPG display brings the relationship a bit closer to the wallet and few things change habits like that.

Yeah, it appeals to both my natural tightness and provides an alternative (and no doubt less risky) game to minimising journey time. I now regularly average 60mpg. I've yet to figure out if I save more fuel by free-wheeling down hills or using the engine to brake the acceleration.

 

[Ivor the Engineer]

Maybe some can, but it shouldn't be down to an individual to decide whether they are in that group. Many people think they are more skilled than they are.

http://www.apa.org/journals/features/psp7761121.pdf

Can having knowledge of our tendency to over-estimate our performance in comparison to others guard against doing it?

 

[aggle-rithm]

Odds are, we will all die some day. So, I recommend against life of any sort, as it always seems to end badly.

 

[Ivor the Engineer]

An interesting article in wiki on risk perception:

http://en.wikipedia.org/wiki/Risk_perception

A key early paper was written in 1969 by Chauncey Starr.[1] Starr used a revealed preference approach to find out what risks are considered acceptable by society. He assumed that society had reached equilibrium in its judgment of risks, so whatever risk levels actually existed in society were acceptable. His major finding was that people will accept risks 1,000 greater if they are voluntary (e.g. driving a car) than if they are involuntary (e.g. a nuclear disaster).

 

[blutoski]

Can having knowledge of our tendency to over-estimate our performance in comparison to others guard against doing it?

For some people, yes; for others, no.

I'm specifically thinking about Puthoff and Targ, who would write down all the ways they could think of that their paranormal experiment could be confounded by a cheating subject... and then do nothing to mitigate it.

They felt that by merely being 'aware of' the problem, they had eliminated it as a factor.

 

[Bob Klase]

Due to my crippling fear of any kind of injury, I used to stay in bed with the covers over my head. Then I heard that most accidents happen within one mile of your home, so I wander the streets 24x7, sleeping in doorways.

A few years ago I also heard that most accidents happen within one mile of your home, so I moved to a different house that was 3 miles away.

 

[Brendy]

No events in life have a probability of 0 or 1 of occurring. Statistical information is rarely available about us as individuals, so we have to rely on averages of groups and assume we share many of the significant characteristics of a particular group if we wish to use the data in our decisions.

What behaviours (if any) do you engage in where you expect to beat the odds?

Is this an irrational way to behave?

Poker, Craps, and Blackjack

I love gambling. Unfortunately only in Poker do I beat the odds and win. Well I don't really beat the odds. I get the same amount of flushes and full houses as everyone else, i just lose less when i lose and win more when i win.

 

[Modified]

As far as driving is concerned, I'm pretty sure I'm pretty good at it. I drive the speed limit and avoid tailgating because I don't trust the rest of you wackos!

It's much safer to try to match the speed of traffic than to drive the speed limit (assuming there is traffic). I would like to do that, but the usual traffic speed here is 15-20 mph over the limit, and I'm not comfortable knowing I could get a speeding ticket at any time, so I usually drive 5-10 over. I'm sure that increases the chance of my having an accident quite a bit, since cars are constantly tailgating and swerving around me. Driving at the speed limit would be much more dangerous, since there would be many more cars tailgating and swerving around, and many of those drivers would be steaming.

 

[sol invictus]

A few years ago I also heard that most accidents happen within one mile of your home, so I moved to a different house that was 3 miles away.

There's a sign on a road I sometimes take that says "dangerous curves". I always accelerate so as to spend as little time in the dangerous area as possible!

 

[Iconoclast08]

...Statistical information is rarely available about us as individuals, so we have to rely on averages of groups and assume we share many of the significant characteristics of a particular group if we wish to use the data in our decisions.

According to what I've read, extrapolating from the group (nomothetic) to the individual (idiographic) is one of the most frequently committed logical fallacies, although the sources I've read may not be the final word on this argument. I guess we have to use something as a guide. Regarding relying on statistics of central tendency, I always think about Stephen J. Gould's essay The Median Isn't the Message; these numbers alone can be way off for us individually, and averages can be heavily influenced by outliers (e.g., in skewed distributions).

Another factor to consider alongside the actual mean or median or mode is the variability about those values in the distribution as a whole. Puts a nice backdrop to the statistical picture that we can consider.

Lastly, carefully crafted and conducted single-n (case) studies can actually be more helpful than many people think when lots of them are done and when the proper methodology is employed (repeated measures, time-series designs; see Single Case Experimental Designs by Barlow, Nock, & Hersen). Although it will never be spot-on, if an individual constellation of variables very closely approximates your own, there may be some value to using it as a predictive tool in limited cases. This is especially true in psychotherapy research, where certain single-n study clusters have been shown to be excellent in gauging treatment efficacy and underlying mechanisms of action. Information does and can get lost when you lump heterogeneous groups of people together.

 

[Mashuna]

There's a sign on a road I sometimes take that says "dangerous curves". I always accelerate so as to spend as little time in the dangerous area as possible!

Which reminds me, I've got a Monica Bellucci film to watch tonight.