Saturday, March 28, 2009

Why we make mistakes by Joseph Hallinan

We are all victims of our own mistakes. We can dwell on this fact and our past errors, or you can pick up this book and discover why you are so bone headed. It seems that the root cause is that thing between your ears. It’s error prone by design. Unless you train it and stay alert, you’re going to mak misteaks, mani of them.

 

And you thought the English were smarter!

A few years ago in Wales, England, a mob of vigilantes attacked and vandalized the office of prominent children’s doctor. Why? Because according to the police they had confused the word ‘pediatrician’ with the word ‘pedophile’… Afterward the doctor gave an interview ‘I suppose I’m really a victim of ignorance.’ P1

 

Nothing like a fine glass of Dakota Cabernet

Our expectations can shape the way we see the world and often the way we act in it as well… ½ the people in a restaurant were told their complimentary glass of cabernet came from CA, the other was told it came from N. Dakota. Not only did the N Datoka group eat less of their meals, but they head for the door more quickly. P2

 

These effects occur largely outside of our consciousness; we’re biased – we just don’t know we’re biased. Some are so strong, that even when we know them, we find it hard to correct for them… Examining 33 studies of answer changing on standardized tests proves that on average that test takers are not hurt by changing their answers. And yet, even after students are told of these results, they still tend to stick with their 1st answers. P3

 

Within 1/10 of second after looking at a scene, we are usually able to extract its meaning. The price we pay for this rapid fire analysis is that we miss a lot of details. P3

 

When prices are set for multiple units (4 for $1.00 instead of $.25/each), sales increase 32%. P7

 

At normal viewing distances, the area of clear vision is only about the size of a quarter. The eye deals with this constraint by constantly darting about, moving and stopping 3 times per second. P12

 

Lean left

When people are forced to make turn at an intersection, right handers prefer turning right (lefties prefer turning left too). As a result, one should look to the left when searching for the shortest lines at stores, banks, and rides. P13

 

Experts and novices tend to look at things in different ways… Experts maintain a longer quite eye period – the amount of time needed to accurately program motor responses that occurs between the last glimpse of a target and the 1st twitch of our nervous system. Experts gaze at the ball much longer and rarely shift their sight. Less skilled individuals don’t stare at the ball very long, and tend to look away often. P14

 

Are you really paying attention?

As a stranger and a pedestrian talk as part of an experiment, they are rudely interrupted by 2 men who pass between them while carrying a door. The interruption is brief – 1 second. But during that 1 second, one of the men carrying the door switches places with the stranger. When the door is gone, he continues the conversation as if nothing had happened. Would the pedestrian notice that they are talking to someone new? The majority didn’t. Only 7 of 15 noticed! P15

 

Probably not

During a film, one actor was replaced by another, and the film was shown to 40 students. Only 1/3 noticed the change. P17

 

Maybe because you’re nearly blind

The eye has high resolution only at an angle of 2 degrees, or about the width of your thumb at arm’s length; beyond that, things get blurry. P18

 

When you look for things seldom seen, you seldom see it

Experimenters asked  volunteers to look at 1000s images. Each image was set against a busy background filled with other images.The volunteers were asked to report whether they saw a tool, like a wrench or a hammer. When the tool was present a lot – which was true ½ the time, the volunteers did a great job of spotting it. They were wrong only 7% of the time. But when the tool was rarely present (1 out of 100 images), their error rate soared to 30%. Why? They gave up. Observers have a quitting threshold – basically, the amount of time they will look for something before giving up. Typically observers slow down after making mistakes, and speed up after successes. Since observers looking for seldom seen items can successfully say no almost all the time and be right, they tend to speed up and drive down their quitting time… In fact, they abandoned their search in less than the average time required to find the target… What if your job were to find a gun? Or a tumor? People don’t want you to quit early – they want you to stay late… Routine mammograms reveal tumors only .3% of the time. Guns are even rarer; roughly 1 in a million passengers. Both occupations have considerable error rates. Radiologists miss 30% of the tumors. In one study 90% of lung tumors were missed for months! Airport screeners have a failure rate of 25%. At LAX, the TSA found that screeners missed 75% of bomb materials. P23

 

Why you can’t remember names

Researchers asked real people to study the biographies of fictional people. The biographies contained names, hometown, occupation, hobbies. So what did people remember about the fake people? Jobs were remembered 69%. Hobbies 68%. Hometowns 62%. First names only 31% and last names only 30%. Why should this be? The best guess is that names, in and of themselves, don’t mean much; they’re just arbitrary labels. Jobs, hobbies, and places, are often semantically richer – they mean something. P29

 

Or passwords

In one test, 30% forgot the passwords after 1 week. 65% were forgotten after 3 months. P4… 80% of IT help desk calls are for forgotten passwords. P34

 

Or phone numbers

A poll of 3000 people found that 25% couldn’t remember their own phone numbers. P34

 

Nice haircut! Do I know you?

When you’re trying to figure out whether you recognize someone, what do you look for? Studies have found that that the most consistent feature is… hair. Which is an interesting choice given that hair, of all our physical features, is one of the most easily altered. But hair it is. P37

 

If you want to remember a person’s face, try to make a number of difficult personal judgments about his face when you’re first meeting him. P38

 

Misidentifying one person for another is more common than many of us would like to believe – especially regarding witness testimony. Between 1989 and 2007, 201 prisoners in the US were freed through DNA analysis. Of these 77% had been mistakenly identified by eyewitnesses. P41

 

The ugly face of crime

Research indicates that criminals are, by and large, uglier than the rest of us… The long term consequences of being young and ugly were small but consistent . Unattractive individuals commit more crime in comparison to average looking ones, and very attractive individuals commit less crime in comparison to those who are average looking. Yet pretty faces are the ones we tend to recognize. P42

 

That time of the month is costly it seems

Researchers found that lap dancers earnings were strongly related to their menstrual cycles. During fertile periods for women not taking the pill, they earned $335 per shift. But during menstruation, earnings plunged 45% to only $185 per shift. Moreover, this pattern was remarkably consistent: all of the dancers made less money during their menstrual periods – whether on the pill or not. Those on the pill consistently earned about $80 less per shift… How did the guys know? A recent 2008 study found that women who are at the peak time of fertility may have changes in their voice that make them sound more attractive (higher pitched). This may occur because the larynx changes it size and shape in response to hormones. P47-8

 

Proof that snobbishness is real

Volunteers were asked to taste 5 wines that were labeled according to price: $5, $10, $35, $45, $90 per bottle. But researchers pulled a switcheroo. The $90 wine appeared twice – once in the $90 bottle, and again in the $10 bottle. The same for the $45 wine: it appeared in the $45 and the $5 bottles. But tasters never noticed; no matter what, they preferred the wine when it was in the more expensive bottle. And this was not just snobbery at work. Brain scans showed that higher priced wines generated more activity in an area of brain that responds to certain pleasurable experiences. What about the so called cheap wines? Well, then the brain registered less pleasure from the same wine. P51

 

On the SAT, researchers asked students how often they had switched their answer and got it wrong, and they consistently overestimated. When asked how often they stuck with their 1st answer and got the problem wrong, they underestimated. Students remember sticking with the 1st is a better strategy than it actually was. The paradox is even though the actual outcome suggests you should do more answer changing, your memories of it suggest the very opposite. P55

 

The numbers don’t add up

The median of lifetime sexual partners for men is 7; for women it is 4. But as with so much in life, the number depends on how rich we are. In wealthy countries it is 10, and in poor countries it is 6… Why there should be such a big gap is unknown. But researchers have found that men exaggerate their number of partners, and women downplay theirs.  p67

 

On average, everyone says they are better than average

One survey that 84% of doctors though their colleagues were influenced by gifts from drug companies. But as for themselves? Only 16% thought they were similarly influenced. P71

 

The coverup

When people take the pains to demonstrate that they are not corrupt in some way, they are actually more likely display exactly this corruption on subsequent tasks. P74

 

Keep your d**m eyes on the f******g road!

While driving, a single 2s glance away doubles the risk of an accident. So do multiple shorter glances that add up to 2s or more… A 2004 study showed that a driver takes about 86s on average to enter in data on a GPS, involving 20 to 30 glances away from the road. P84

 

Polka music actually works?

While wine shoppers perused the shelves, French music was played, and 40 bottles of French wine were sold. But when German music was played, sales of French wine plunged to just 20 bottles. Same was true of german wines; 22 bottles sold while German music was played, and only 8 when not… Of the 44 shoppers interviewed, only 6 said their choice was influenced by the music. P93

 

Why Safeway prices the way it does

Stores wishing to boost sales often use multiple unit pricing (4 for a $1.00 instead of $.25 each)… Such pricing resulted in a 32% increase in sales over single unit pricing. P105

 

Another way to boost sales is by using quantity limits – limit 10 per customer. The limit acts as an anchor. The higher the anchor the higher the sales. Only when the anchor gets absurdly high – say, 50,  does its affect tail off. P105

 

First mover advantage is real in some contexts

The key to anchoring is the first number, because people process information in the order presented. And the best place is first. Being listed first on the ballot can add 3% to the results by major party candidates… Making the first offer in a negotiation adds significant advantage, because it serves as the anchor for future discussion… Research shows that whichever party – buyer or seller – who made the 1st offer obtained a better outcome. P106

 

Why being young and inexperienced is a blessing

In April 2008, a 13 year old corrected NASA’s estimates that an asteroid would collide with the Earth. A few weeks before that, a 5th grader discovered an error at a Smithsonian Institution exhibit that had gone undetected for 27 years… As something becomes familiar, we tend to notice less, not more. We come to see things not as they are but as they ought to be, as we assume.  P113

 

And the 2005 best Halloween costume award goes to…

During Halloween 2005, the apparent suicide of a woman hanging from a tree went unreported for more than 14 hours – even though the body was plainly visible to neighbors and drivers. “They thought it was a Halloween decoration” explained the wife of the mayor. P114

 

The problem is that in remembering maps, we systematically distort them… Parisians when asked to map the city, consistently straightened the Siene. 92% understated the curvature. NYC cabdrivers do the same thing when asked to draw the streets of NYC… Further, people overestimate short distances, and underestimate long distances. And distances to a landmark are less than the distance from the landmark! Example: North Korea is to China, than China is to North Korea. P120

 

We constantly distort when we share information and retell stories 2.7 times on average. We omit, exaggerate or minimize info in 61% of our stories, but we admit doing this only 42% of the time. So quite a bit is unconscious. P130

 

Live fast, die poor

If you like to drive fast, the odds are you also like to trade stock a lot… For each additional speeding ticket this was correlated with an 11% increase in the turnover rate of driver’s stock portfolio. P134

 

Men also tend to trade stocks more often than women – 45% more often… And when aren’t married, the difference is even starker: single men trade 67% more often than single women. P135

 

Shoot first, and ask questions later

The US Army found that male soldiers tended to shoot people that they shouldn’t shoot (friendly fire), but female soldiers tended not to the shoot the people they should shoot (kill the enemy). P136

 

Living while being male is hazardous to your health

Men are 3 times as likely to be involved in an automobile fatality. They are also more likely to die from drowning or accidental poisoning. P138

 

But men don’t remember any of that in any case

Men tend to forget their mistakes more readily than women, and mistakes appear to dog women in ways that don’t bother men. Women indicate that failure affects their self esteem more than success. No such difference has been found in men. P140

 

When it comes to solving problems, a lack of confidence has been shown to affect not only the outcome we achieve but the approach we take… Self doubters are slower to abandon faulty strategies and less likely to come up with alternatives. P142

 

If parents are fearful they may keep their children close by, reducing the size of the home range. City kids may also not wander as far as country kids. But in general, 2 trends hold true: the home range for kids expands rapidly between ages of 6 and 9, and the range is universally larger for boys. P145

 

When researchers studied the subset of girls who traveled most freely and widely in their local areas, they found that these girls use about the same amount of detail in their maps as boys do. P146

 

When children are allowed to roam freely over an area, and closely explore rocks, trees, creeks, and the like, they seem to develop a deeper understanding of place. P146

 

It pays to be right I suppose

Almost everyone is overconfident – except the people who are depressed, and they tend to be realists. P149

 

And you could of bought $700 worth of donuts!

On average, gym members overpay by $700… 80% of monthly paid members would have been better off with a per visit plan… Why do they do it? The overestimate their self control, and confuse what they should do with what they will do. P152

 

When you don’t go the gym, the feedback is weak (you don’t get weak or fat immediately). And when feedback is weak, it is easy to ignore or distort. In the case of not going to the gym the signal is laziness. But we don’t like that signal, so we ignore it, and come up with rationalizations (traffic was bad, I’m feeling kinda sick, work was too hard, etc.) p159

 

Why are weather forecasters so well calibrated (accurate) when the rest of us aren’t (think economists especially)? The answer: corrective feedback… Forecasting started in 1920 in the Pecos Valley of New Mexico by a former schoolteacher cum railroad man cum grocery clerk, who began to provide farmers with a probability of rainfall so they could decide whether to cut and cure alfalfa on dry days. The farmers wanted to know if there was a good chance of rain so as not to cut that day… Eventually this system spread from Connecticut in 1954 to SF in 1956 to LA in 1957 and finally in 1965, nationwide… Analysts have found, using 150,000 forecasts over a 2 year period, that when forecasts predict a 30% chance of rain, that it rained almost exactly 30% of the time! P156

 

Betcha that you’re a self deluder

Initial success has a powerful influence over you. Students who guessed head and tails correctly during repeated coin flips became convinced that they were skilled enough at guessing, and that they could actually make successful prediction more than ½ of the time… 40% even felt that they could actually improve with practice. This is the illusion of control. P162

 

Horseracing handicappers were no more accurate with 40 pieces of information than with 5. But – and this was important – using more information did increase their confidence… from under 20% to over 30%. P165

 

Overconfidence quiz: http://www.tim-richardson.net/misc/estimation_quiz.html

Just for the record I got 8 out 10 correct – 80% confidence rating. I thought of course that I would get at 90% rating, so I failed. 99% of people fail to achieve this result.

 

Hooray for Cliffs Notes!

Students who were given 20 to 30 min learned more in that time from reading chapter summaries than reading the chapters. This was true if they were tested immediately after the reading, or tested 1 year later. P163

 

Mulligan

What percentage of 6 foot putts did PGA golfers on tour make? A former Amateur champion guessed over 80% Another veteran said ‘if you aren’t making 85%, you aren’t making any money. The actual average for the best golfers in the world is a measly 54.8%... Also the success rate in putting for par vs a birdie is far higher, 25% higher at a 5 ft distance. This suggests that psychological pressure affects even the best of golfers. p170

 

Chess grand masters, who typically have logged over 30,000 hours of play, can recall the positions of every piece nearly perfectly after a brief glimpse. Less experienced players, got the positions right only 50 to 70% of the time. What accounts for the difference? It is not that the grand masters have better memories, because if the board is scrambled in random positions, the grand masters have no better recall. The grand masters only had better memory when the pieces were in positions that made sense – or when they were part of a pattern that they recognized. P174

 

Witch doctors in our midst

Studies of autopsies have shown that doctors seriously misdiagnose fatal illnesses about 20% of the time… So millions of people are treated for the wrong disease. But the really shocking thing is this: the rate of misdiagnosis has not really changed since the 1950s according to JAMA. Compare that to aviation. The Navy’s accident rate decreased from 50 accidents per 100,000 hours flown in the 1950s to only 1.5 today… Civilian airlines have improved their safety records by over 65% in just the past 10 years alone… Why is this? Unlike pilots, doctors don’t go down with their planes. This difference gives pilots a powerful incentive to eliminate errors.  P193

 

But there’s more. Junior staff members should be allowed to question decisions made by senior staff members. When airline pilots were asked, nearly all of them – 97% - agreed with this statement. But only 55% of surgeons agreed… One survey of hand surgeons revealed that they operated on the wrong site (hand) at least once in their careers. P195

 

The golden state of mind

Every year the Harris poll asks Americans to name the state they would most like to live. And every year for the past five years the answer has been CA… But at last count over 2.2M Californians moved to another state, according to the census bureau. P201

 

The true golden goose

40% of rebate coupons are never redeemed. We overestimate our ability to fill out and mail them in, but factor the rebate into the purchase price... This misprediction is also why gift cards are such a bad idea… Unused gift cards together add up to an $8B per year boon for retailers, since it is pure money in the bank. P203

 

S**t happens; just deal with it

½ of the patients faced the possibility of having either colostomies reversed at some point in the future; for the other ½, the operation was permanent. Over a 6 month period, the patients were asked to rate their life satisfaction. If you’re like most people, you probably think those with the reversible colostomies would be happier, since we all dread finality… But it turns out those with permanent conditions were happier… Why? Hope impedes adaptation. If you’re stuck with something, you learn to live with it. And the sooner you learn to live with it, the happier you’ll be. P205

 

As counterintuitive as it may sound, when it comes to happiness, there’s a good deal of evidence that circumstances don’t matter as much as we might think. Researchers have found that social status, education, income, marital status, nor religious commitment account for no more than 3% of the variation in reported levels of happiness. Persons with severe disabilities ranging from quadrapelgia to blindness consistently report high levels of happiness. P206

 

Want to make fewer mistakes? Become a pessimist

The next time you have a major decision to make, ask yourself ‘What could go wrong?’ This may strike you as needlessly pessimistic, but thinking positively has limitations, among other things it can blind us to the pitfalls that lay camouflaged inside our ideas… If you get people to play devil’s advocate with themselves – asking what evidence is against this is – overconfidence is pretty close to eliminated. So give it a try. P214 Now, I’m too depressed and underconfident to try!

 

Only trust a statistician

The CIA has concluded that its’ Analysts should give little weight to anecdotal and personal case histories unless they are know to be statistically typical, and perhaps no weight at all if the aggregate data based upon a more valid sample can be obtained. So ask for averages not testimonials. P215

 

 

No comments: