In Grand Junction, Colorado a billboard showing images of President Obama in an unfavorable light was taken down after the owner apparently received threats of violence. Now to be clear about it, I think that billboard was over the top. It did nothing to advance reasonable dialogue and in fact probably aided those who support the president. Though I oppose Obama I think it was counterproductive. However that is not the real issue here. The issue is freedom of speech vs. what some people seem to perceive as a right not to be offended. In fact one opponent of the billboard said, "If it offends people, you do have a certain obligation to take it down...”
Even a little thought should show the fallacy of such a criterion. It implies that people have the right not to be offended. That makes the issue one of how people react rather than of the actions taken by the alleged offender. It gives the “offended” person complete control over what is not allowed. I could say that they way you comb your hair offends me so you better change that. Or I could claim to be offended by the color of your car. I may be offended by advertising for politicians I don't like. Note that those complaints are based on my reaction, not on anything the “guilty” party has done.
Imagine a world in which guilt is based not on the actions of the accused but on how people react to those actions. How might you manage your life so as to avoid trouble in such a world? You cannot, no matter what you do someone may claim it offends him. Even worse, you cannot know how they will react before you take the action. There will always be someone who can claim to be offended. That has the effect of an ex post facto law. You might do something you consider quite reasonable such as planting a vegetable garden. You might even ask your neighbors what they think of the idea and they might agree before you plant it. Then when it starts to grow a neighbor decides he really doesn't like the sight of corn and you should have planted flowers. He is offended and you are guilty of causing offense, even though you could not know beforehand he that would be offended.
In fact, taken to the extreme this could create the equivalent of a bill of attainder. Someone may decide that he doesn't like you and is offended by the sight of you. He is offended by your person, not your actions. You are guilty based on who you are, not on what you do.
This gets even worse. If you are accused of offending someone how can you defend yourself? There is no external evidence available. If the accuser claims you stole his grocery money the courts will ask for evidence. However what if you are accused of offending him what evidence might be considered? I could claim that you offend me and who is to say if I am telling the truth or lying? I might just be trying to get back at you or to gain some advantage for myself.
People have a right to their feelings and if they feel offended that is also their right. However they should not use that feeling to accuse others or as cause for legal action. They may complain about what someone did or said but their feelings are their own, unverifiable and not useable as a basis for action by anyone else.
There is yet more to this however. Some people seem to be professionally offended. They make a career out of it, either by constant complaining to get what they want or sometimes even by making money from being offended. Jesse Jackson is one that I suspect falls in this category. If there is even a chance a black person has been discriminated against he likes to show up to offer his advice and stir up demonstrators. He seems to do that based on the assumption that racism is involved, ignoring other possible causes for the perceived offense.
For example Jackson (and many others) was professionally offended in 2006 when three Duke Lacrosse players were accused of raping a black woman. He jumped in with complaints about spoiled white boys abusing minorities. I don't know how much his Rainbow/PUSH coalition collected by fund raising as a result but in that case his being offended was the result of a rush to judgment. It turned out that the accused were not guilty. Two of them had solid alibis and the “victim's” story had more holes than a chunk of Swiss cheese. Worse, as an indication of how professionally offended some people were, the taxi driver who provided the alibi was excoriated for telling the truth – and he was a black man. Jackson and others were offended before learning the facts. The result was a suspension of the entire team, heightened racial tension, and several young men having their reputations sullied, all without factual justification. Had the “offended” simply waited for the facts that could have been prevented.
Of course not everybody who is professionally offended gets money for it, at least not directly. Some only get sympathy, others get preferential treatment in hiring, school admission etc. That leads to discrimination, extra costs and other problems for society.
We should look at facts, evidence, and sound logic. And we should never base action only on someone being offended. In fact we should ignore the professionally offended until such time as they present evidence of bad action by their accused offenders.
Showing posts with label facts. Show all posts
Showing posts with label facts. Show all posts
Monday, October 18, 2010
Monday, August 24, 2009
And the Facts Mean? (Part 4)
The EEOC has set an 80% standard for hiring and promotion. If a company hires 50% of the white people who apply, it must hire at least 80% of that or 40% of the minorities who apply. Failure to do so is considered prima facie evidence of discrimination. That rule might be useful in determining which companies bear investigation, but it is a terrible method of determining who is actually discriminating.
I've already discussed the fact that different groups have different characteristics. Some minorities, especially in the inner cities, have a culture that discourages education. Why should a company be forced to hire any certain number of them if the job requires cognitive skills? However there are also statistical problems with that rule.
Suppose for example 10% of the people living in a town are black and 90% white. A company in that town has four people doing a certain job. Now 10% of those four employees would be 0.4 people and 80% of that is 0.32 people. It is rather difficult to hire a fraction of a person. If they have one Black among the four employees, then 25% of the people there would be black, far in excess of the 10% of the population that Blacks represent.
Even if the company has 40 people doing that job, statisticians would tell us that it is unreasonable to expect exactly four of them to be black and 36 white. In fact statistically it would not be at all surprising if there were anywhere from zero to eight Blacks among those 40 workers. If you have the time, you can demonstrate this to yourself if you can find a few marbles of different colors. Put 9 of one color and one of a different color in a box and shake it up. If you don't have marbles you can use something else, maybe coins with a coin from a particular year representing the minority. You could even put a dot of fingernail polish on the minority coin.
This experiment is to simulate hiring at random, with the marbles representing the applicants. The nine majority color marbles represent the 90% white population in the town and the odd marble (or coin of a particular date) represents the 10% black population. Now without looking, randomly draw one marble from the box and record what color it is. Replace the marble, shake the box and draw again. Repeat this 40 times to simulate hiring 40 random people. How many times did you draw the odd marble? Probably not exactly four times.
Now repeat the above experiment at least ten times, each time recording how many times you drew the odd marble. Most likely once or twice you will draw the odd marble only once or not at all. Once or twice you will draw it 6 to 8 times out of the 40. Most of the time you will draw it three, four or five times out of the 40 tries. However even only drawing it three times fails the EEOC test of 80%, that means you only “hired the minority” 75% of the times that the minority “applied” for your job opening. Would you like to be exposed to charges of discrimination for the times you drew zero, one, two or even three minority marbles?
If you took the time to do this experiment what you have seen is normal statistical fluctuations. You can see something similar by tossing a coin ten times as I described in a previous blog. The most likely result is five heads and five tails. However there are many other possible results. Each of those other results is less likely than the five heads and five tails, but there are more of those possibilities so together they outweigh the probability of equal heads and tails. In fact the probability of getting exactly five heads and five tails is less than 25%.*
“But wait,” you say. “Don’t large companies hire more than 40 people, even more than 400?” Yes they do, but each work area usually has a much smaller number of people. Furthermore, even for larger numbers we would expect some deviation from the most expected value. The larger the number in the group, the less deviation we would expect percentage-wise, but there will be some. On top of that, as we go up the corporate ladder to more lucrative jobs, the number of employees gets smaller. A company may have 10,000 employees but will probably only have fifty or so in top management and fewer than ten at the vice presidential level or higher.
To say that a company is guilty of discrimination on the basis of statistics alone is to ignore those statistical fluctuations, as well as the fact that we cannot expect people of different backgrounds to have identical qualifications. Such a declaration of guilt ignores science. While statistical differences may raise questions, we should insist on actual evidence of biased actions before claiming that anyone or any company is guilty.
Now on somewhat of a personal note: I’ve been publishing five blogs per week and have enjoyed it. However there are also other things I would like to do, including complete a novel I started some time ago. To free up time for those projects I plan to cut back, initially to only three blogs per week. They will probably to be published on Monday, Wednesday and Friday. I’ll see how that goes and hopefully I won’t have to cut down to twice per week.
*For the mathematically inclined, the probability of getting exactly m heads out of n coin tosses is
P = (0.5)^n [n!/{(n-m)! m!}]
If you like my blog, please tell others.
If you don’t like it, please tell me.
I've already discussed the fact that different groups have different characteristics. Some minorities, especially in the inner cities, have a culture that discourages education. Why should a company be forced to hire any certain number of them if the job requires cognitive skills? However there are also statistical problems with that rule.
Suppose for example 10% of the people living in a town are black and 90% white. A company in that town has four people doing a certain job. Now 10% of those four employees would be 0.4 people and 80% of that is 0.32 people. It is rather difficult to hire a fraction of a person. If they have one Black among the four employees, then 25% of the people there would be black, far in excess of the 10% of the population that Blacks represent.
Even if the company has 40 people doing that job, statisticians would tell us that it is unreasonable to expect exactly four of them to be black and 36 white. In fact statistically it would not be at all surprising if there were anywhere from zero to eight Blacks among those 40 workers. If you have the time, you can demonstrate this to yourself if you can find a few marbles of different colors. Put 9 of one color and one of a different color in a box and shake it up. If you don't have marbles you can use something else, maybe coins with a coin from a particular year representing the minority. You could even put a dot of fingernail polish on the minority coin.
This experiment is to simulate hiring at random, with the marbles representing the applicants. The nine majority color marbles represent the 90% white population in the town and the odd marble (or coin of a particular date) represents the 10% black population. Now without looking, randomly draw one marble from the box and record what color it is. Replace the marble, shake the box and draw again. Repeat this 40 times to simulate hiring 40 random people. How many times did you draw the odd marble? Probably not exactly four times.
Now repeat the above experiment at least ten times, each time recording how many times you drew the odd marble. Most likely once or twice you will draw the odd marble only once or not at all. Once or twice you will draw it 6 to 8 times out of the 40. Most of the time you will draw it three, four or five times out of the 40 tries. However even only drawing it three times fails the EEOC test of 80%, that means you only “hired the minority” 75% of the times that the minority “applied” for your job opening. Would you like to be exposed to charges of discrimination for the times you drew zero, one, two or even three minority marbles?
If you took the time to do this experiment what you have seen is normal statistical fluctuations. You can see something similar by tossing a coin ten times as I described in a previous blog. The most likely result is five heads and five tails. However there are many other possible results. Each of those other results is less likely than the five heads and five tails, but there are more of those possibilities so together they outweigh the probability of equal heads and tails. In fact the probability of getting exactly five heads and five tails is less than 25%.*
“But wait,” you say. “Don’t large companies hire more than 40 people, even more than 400?” Yes they do, but each work area usually has a much smaller number of people. Furthermore, even for larger numbers we would expect some deviation from the most expected value. The larger the number in the group, the less deviation we would expect percentage-wise, but there will be some. On top of that, as we go up the corporate ladder to more lucrative jobs, the number of employees gets smaller. A company may have 10,000 employees but will probably only have fifty or so in top management and fewer than ten at the vice presidential level or higher.
To say that a company is guilty of discrimination on the basis of statistics alone is to ignore those statistical fluctuations, as well as the fact that we cannot expect people of different backgrounds to have identical qualifications. Such a declaration of guilt ignores science. While statistical differences may raise questions, we should insist on actual evidence of biased actions before claiming that anyone or any company is guilty.
Now on somewhat of a personal note: I’ve been publishing five blogs per week and have enjoyed it. However there are also other things I would like to do, including complete a novel I started some time ago. To free up time for those projects I plan to cut back, initially to only three blogs per week. They will probably to be published on Monday, Wednesday and Friday. I’ll see how that goes and hopefully I won’t have to cut down to twice per week.
*For the mathematically inclined, the probability of getting exactly m heads out of n coin tosses is
P = (0.5)^n [n!/{(n-m)! m!}]
If you like my blog, please tell others.
If you don’t like it, please tell me.
Thursday, August 20, 2009
And the Facts Mean? (Part 2)
Confession time: I belong to a group that has government approval but not a single black member. Furthermore I recently attended an international conference of people from similar groups and saw not even one black person there. If someone looked only at the statistics they might claim that we are discriminating, but I can assure you we are not. In fact I cannot remember a black person ever applying for membership. That group is my rescue team and the immediate cause of that racial imbalance is that we recruit only among competent mountaineers. However for some reason there are few Blacks involved in mountaineering. In fact I've yet to meet any though I did talk with a woman who once saw two Blacks climbing Washington's Mount Adams.
Even lacking firm statistics, it is clear that Blacks are seriously underrepresented in the mountaineering community and in mountain rescue. The area where I live has plenty of black residents, so why are there hardly any black climbers? I don't have any answers though I am very curious. It cannot be lack of inborn talent; black athletes have long demonstrated great talent. Nor can it be discrimination; mountains and cliffs neither know nor care about the skin color of people climbing them. I suspect the reason is a combination of culture and finances but that is a guess on my part, not backed up with any research.
Though not a big issue, the racial imbalance among mountaineers does illustrate the fact that statistical differences between groups are not necessarily caused by discrimination. We don't need to know what the reason is in order to know that some suggested reasons are wrong. I needn't know why there are few black mountaineers to know that discrimination is not the cause.
Now I like statistics, I think they can be fun and useful. However acting on statistics alone usually means bypassing vital information. If our mountain rescue team were accused of discrimination, the statistics might support the charge. If we were forced to include a certain number of black people, we would probably out of business. I’m not sure we could find them and even if we did, they would likely lack the mountaineering experience necessary to safe and effective rescue work. They would become a danger to us, to our subjects, and to themselves. The effect would be bad for all concerned, including the occasional black hiker we are called to help.
Statistics tends to lump people together in large groups. That is both its strength and its weakness. It is a strength because that allows us to make sense of large amounts of data – if we are very careful how we use that data. However statistics cannot tell us much about individual cases. It may tell us that young inner city black men are more likely to be criminals than are men of middle age living in the suburbs, be they black or white. However we should not use that information to convict any particular young black man from the inner city, there are also upstanding citizens among that group.
Statistics alone cannot tell us why some groups have different characteristics than others. Blacks may be underrepresented in mountaineering, but we have to dig deeper than the numbers if we want to learn why. Likewise Blacks may be over-represented among NFL running backs and underrepresented among quarterbacks in that league. The numbers won’t tell us why. However it is highly unlikely that discrimination is the reason for the NFL statistics. Football fans want winners and any team owner who put skin color ahead of ability to win games would soon face a bankruptcy judge. Team owners seek the players who will help them win. Even if those owners happen to be biased, they cannot afford to discriminate on the basis of anything but ability.
I’ve not heard of anybody wanting to charge NFL or NBA teams with discrimination, even though player racial statistics clearly do not reflect the average racial make-up of the U.S. population. That is wise, those players are hired for their ability, which is as it should be. Unfortunately that is not the case in other businesses. If a manufacturer or retailer fails to hire the statistically expected number of women or Blacks, and hire them into the statistically expected positions, that employer will probably be accused of discrimination. Furthermore, in violation of standard judicial procedure in this country, the burden of proof will be on the employer, not the accuser.
The Equal Employment Opportunity Commission has actually codified this over-reliance on statistics alone. Though the rules do show some understanding of variations from the statistical norm, that allowance is quite rigid, it fails to consider sample size or intergroup variations. Worse yet, they do not consider the fact that there are very real differences between groups. A factory located between an inner city and the suburbs is expected to hire people from the inner city into the same positions and in approximately the same proportions as employees from the suburbs. That ignores the fact that people from the inner city are almost certainly not as cognitively skilled as those from the suburbs.
Of course there is a way the factory owner can avoid that problem. He can move operations to a place where most people have the skills he needs, like somewhere a long way from the inner city. In fact if he wants to avoid bankruptcy he may be forced to do that. It is difficult to compete if you are forced to hire incompetent people. That means that the inner city people he might have hired into low paying jobs will now have no jobs at all. The rule intended to help those inner city people will in fact harm them.
The fact is that there are differences between groups. In many cities, the inner city black culture militates against doing well in school. It should be no surprise that the products of that culture lack the cognitive skills needed in many jobs. Research has also shown that women’s brains are different from men’s brains. That doesn’t mean one is smarter than the other, but it may well help explain differences in what they like to do and how they do it. It should be no surprise that women prefer different types of work than do men. That is no excuse foe discrimination against the inner city Black who does have the skills for a job, or a woman who happens to enjoy running a jackhammer. However it is a reason not to expect all groups to be equally represented in all occupations.
Unfortunately too many powerful people assume that any deviation from the most expected value automatically means discrimination. Nothing could be farther from the truth. However the fact that those people are sometimes in a position to force others to act on their false beliefs harms our economy.
Next I think I’ll have a little fun with the NFL running back and quarterback issue before I return to some statistical stuff.
If you like my blog, please tell others.
If you don’t like it, please tell me.
Even lacking firm statistics, it is clear that Blacks are seriously underrepresented in the mountaineering community and in mountain rescue. The area where I live has plenty of black residents, so why are there hardly any black climbers? I don't have any answers though I am very curious. It cannot be lack of inborn talent; black athletes have long demonstrated great talent. Nor can it be discrimination; mountains and cliffs neither know nor care about the skin color of people climbing them. I suspect the reason is a combination of culture and finances but that is a guess on my part, not backed up with any research.
Though not a big issue, the racial imbalance among mountaineers does illustrate the fact that statistical differences between groups are not necessarily caused by discrimination. We don't need to know what the reason is in order to know that some suggested reasons are wrong. I needn't know why there are few black mountaineers to know that discrimination is not the cause.
Now I like statistics, I think they can be fun and useful. However acting on statistics alone usually means bypassing vital information. If our mountain rescue team were accused of discrimination, the statistics might support the charge. If we were forced to include a certain number of black people, we would probably out of business. I’m not sure we could find them and even if we did, they would likely lack the mountaineering experience necessary to safe and effective rescue work. They would become a danger to us, to our subjects, and to themselves. The effect would be bad for all concerned, including the occasional black hiker we are called to help.
Statistics tends to lump people together in large groups. That is both its strength and its weakness. It is a strength because that allows us to make sense of large amounts of data – if we are very careful how we use that data. However statistics cannot tell us much about individual cases. It may tell us that young inner city black men are more likely to be criminals than are men of middle age living in the suburbs, be they black or white. However we should not use that information to convict any particular young black man from the inner city, there are also upstanding citizens among that group.
Statistics alone cannot tell us why some groups have different characteristics than others. Blacks may be underrepresented in mountaineering, but we have to dig deeper than the numbers if we want to learn why. Likewise Blacks may be over-represented among NFL running backs and underrepresented among quarterbacks in that league. The numbers won’t tell us why. However it is highly unlikely that discrimination is the reason for the NFL statistics. Football fans want winners and any team owner who put skin color ahead of ability to win games would soon face a bankruptcy judge. Team owners seek the players who will help them win. Even if those owners happen to be biased, they cannot afford to discriminate on the basis of anything but ability.
I’ve not heard of anybody wanting to charge NFL or NBA teams with discrimination, even though player racial statistics clearly do not reflect the average racial make-up of the U.S. population. That is wise, those players are hired for their ability, which is as it should be. Unfortunately that is not the case in other businesses. If a manufacturer or retailer fails to hire the statistically expected number of women or Blacks, and hire them into the statistically expected positions, that employer will probably be accused of discrimination. Furthermore, in violation of standard judicial procedure in this country, the burden of proof will be on the employer, not the accuser.
The Equal Employment Opportunity Commission has actually codified this over-reliance on statistics alone. Though the rules do show some understanding of variations from the statistical norm, that allowance is quite rigid, it fails to consider sample size or intergroup variations. Worse yet, they do not consider the fact that there are very real differences between groups. A factory located between an inner city and the suburbs is expected to hire people from the inner city into the same positions and in approximately the same proportions as employees from the suburbs. That ignores the fact that people from the inner city are almost certainly not as cognitively skilled as those from the suburbs.
Of course there is a way the factory owner can avoid that problem. He can move operations to a place where most people have the skills he needs, like somewhere a long way from the inner city. In fact if he wants to avoid bankruptcy he may be forced to do that. It is difficult to compete if you are forced to hire incompetent people. That means that the inner city people he might have hired into low paying jobs will now have no jobs at all. The rule intended to help those inner city people will in fact harm them.
The fact is that there are differences between groups. In many cities, the inner city black culture militates against doing well in school. It should be no surprise that the products of that culture lack the cognitive skills needed in many jobs. Research has also shown that women’s brains are different from men’s brains. That doesn’t mean one is smarter than the other, but it may well help explain differences in what they like to do and how they do it. It should be no surprise that women prefer different types of work than do men. That is no excuse foe discrimination against the inner city Black who does have the skills for a job, or a woman who happens to enjoy running a jackhammer. However it is a reason not to expect all groups to be equally represented in all occupations.
Unfortunately too many powerful people assume that any deviation from the most expected value automatically means discrimination. Nothing could be farther from the truth. However the fact that those people are sometimes in a position to force others to act on their false beliefs harms our economy.
Next I think I’ll have a little fun with the NFL running back and quarterback issue before I return to some statistical stuff.
If you like my blog, please tell others.
If you don’t like it, please tell me.
Wednesday, August 19, 2009
And the Facts Mean? (Part 1)
A while back my friend Randy loaned me a book. I returned it to him after reading only a couple of chapters. Can you guess why?
I suspect most will guess that I didn't like the book. Actually I did like it.
Your next guess is likely to be that he asked for it back, or I didn't have time to read it or something along those lines. All reasonable guesses that explain the facts – but again all wrong.
In fact I liked the book so much that I went out and bought my own copy.*
There are many other cases in our lives when the most obvious conclusion can be wrong, often flagrantly wrong. We seldom have perfect data, nor is our logic always flawless. That is just part of the human condition and we have to live with it. There are two basic ways of living with this imperfection: (1) We can recognize that we might be wrong and check our conclusions against new facts when we get them, or (2) we can assume that we are correct and refuse to consider evidence to the contrary.
Refusing to recognize the facts sets us up for major problems. What if the question is more important than why I returned the book to my friend? Maybe something like the wisdom of an investment you made? If information becomes available that the company you invested in now has unexpected problems, it would be wise to re-visit that decision. Most of us understand this and are willing to change our thinking if we get new information.
Unfortunately there are some whose intellectual arrogance causes them to stick with bad decisions, even when those decisions become costly. Worse, many of those people are third party decision-makers who do not pay the price for their mistakes. Instead their errors cause problems for others. Most of them are government employees who can impose the costs of their decisions on others. Since they do not themselves suffer the consequences of their bad decisions they have little incentive to correct their thinking.
For example, the Equal Employment Opportunity Commission charged Sears, Roebuck and Company with sex discrimination. The case dragged on for years, from 1973 to 1986. In the end an appeals court dismissed the suit, citing the “EEOC's failure to present testimony of any witnesses who claimed that they had been victims of discrimination by Sears.” In spite of not being able to find a single woman who thought she had been discriminated against, those government employees pursued a case against the innocent.
The EEOC had one fact, namely that there was a statistical disparity among Sears' employees. From that they jumped to the conclusion that Sears was discriminating. EEOC decision-makers failed to consider if there might be other reasons for that disparity. Discrimination was one possible explanation of the fact, but not the only one.
And who paid for that government action? First of course was Sears, its customers and stockholders. Those customers and stockholders were the sole source of the money needed for legal defense. Second, taxpayers had to pony up the money to prosecute the suit. Taxpayers, stockholders, and customers were all losers in that action, even though Sears “won” the case.
Perhaps a more important question is who did not pay for that action? The answer is the decision-makers who pursued a case on only statistical evidence (except insofar as they were taxpayers). Those bureaucrats continued to draw their salaries and receive their benefits, all tax paid. They were third-party decision-makers, suffering no consequences for the costs they inflicted on others.
There are of course many such cases in this country. The EEOC and others regard statistics alone as proof. In fact Alice Kessler-Harris, who testified for the EEOC in the Sears case, baldly states that she “did not agree that women's lack of 'interest' could absolve a company of charges of discrimination.” That even though she also “did think that there was some as yet undefined difference between men and women.” I find that an amazing position. She agrees that men and women are different and might have different interests. Yet she expects the statistics to show identity of outcome unless there is discrimination!
The EEOC and others have a severe form of intellectual arrogance that allows them to ignore facts and alternative explanations and assign any statistical difference to discrimination. A company in private business with such a distorted vision would soon fall prey to its competitors and probably go out of business. Unfortunately bureaucrats are insulated from reality and can continue to force their version of reality on the rest of us. That will not change until the voters demand that their government require real evidence, not just statistical differences, before charging companies with discrimination.
Next time I plan to discuss some of the reasons statistics alone can be a problem.
*The book was Vindicating the Founders by Thomas G. West.
If you like my blog, please tell others.
If you don't like it, please tell me.
I suspect most will guess that I didn't like the book. Actually I did like it.
Your next guess is likely to be that he asked for it back, or I didn't have time to read it or something along those lines. All reasonable guesses that explain the facts – but again all wrong.
In fact I liked the book so much that I went out and bought my own copy.*
There are many other cases in our lives when the most obvious conclusion can be wrong, often flagrantly wrong. We seldom have perfect data, nor is our logic always flawless. That is just part of the human condition and we have to live with it. There are two basic ways of living with this imperfection: (1) We can recognize that we might be wrong and check our conclusions against new facts when we get them, or (2) we can assume that we are correct and refuse to consider evidence to the contrary.
Refusing to recognize the facts sets us up for major problems. What if the question is more important than why I returned the book to my friend? Maybe something like the wisdom of an investment you made? If information becomes available that the company you invested in now has unexpected problems, it would be wise to re-visit that decision. Most of us understand this and are willing to change our thinking if we get new information.
Unfortunately there are some whose intellectual arrogance causes them to stick with bad decisions, even when those decisions become costly. Worse, many of those people are third party decision-makers who do not pay the price for their mistakes. Instead their errors cause problems for others. Most of them are government employees who can impose the costs of their decisions on others. Since they do not themselves suffer the consequences of their bad decisions they have little incentive to correct their thinking.
For example, the Equal Employment Opportunity Commission charged Sears, Roebuck and Company with sex discrimination. The case dragged on for years, from 1973 to 1986. In the end an appeals court dismissed the suit, citing the “EEOC's failure to present testimony of any witnesses who claimed that they had been victims of discrimination by Sears.” In spite of not being able to find a single woman who thought she had been discriminated against, those government employees pursued a case against the innocent.
The EEOC had one fact, namely that there was a statistical disparity among Sears' employees. From that they jumped to the conclusion that Sears was discriminating. EEOC decision-makers failed to consider if there might be other reasons for that disparity. Discrimination was one possible explanation of the fact, but not the only one.
And who paid for that government action? First of course was Sears, its customers and stockholders. Those customers and stockholders were the sole source of the money needed for legal defense. Second, taxpayers had to pony up the money to prosecute the suit. Taxpayers, stockholders, and customers were all losers in that action, even though Sears “won” the case.
Perhaps a more important question is who did not pay for that action? The answer is the decision-makers who pursued a case on only statistical evidence (except insofar as they were taxpayers). Those bureaucrats continued to draw their salaries and receive their benefits, all tax paid. They were third-party decision-makers, suffering no consequences for the costs they inflicted on others.
There are of course many such cases in this country. The EEOC and others regard statistics alone as proof. In fact Alice Kessler-Harris, who testified for the EEOC in the Sears case, baldly states that she “did not agree that women's lack of 'interest' could absolve a company of charges of discrimination.” That even though she also “did think that there was some as yet undefined difference between men and women.” I find that an amazing position. She agrees that men and women are different and might have different interests. Yet she expects the statistics to show identity of outcome unless there is discrimination!
The EEOC and others have a severe form of intellectual arrogance that allows them to ignore facts and alternative explanations and assign any statistical difference to discrimination. A company in private business with such a distorted vision would soon fall prey to its competitors and probably go out of business. Unfortunately bureaucrats are insulated from reality and can continue to force their version of reality on the rest of us. That will not change until the voters demand that their government require real evidence, not just statistical differences, before charging companies with discrimination.
Next time I plan to discuss some of the reasons statistics alone can be a problem.
*The book was Vindicating the Founders by Thomas G. West.
If you like my blog, please tell others.
If you don't like it, please tell me.
Thursday, June 25, 2009
Facts? What Facts?
If you like my blog, please tell others.
If you don’t like it, please tell me.
"It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Attributed to Mark Twain)
Your neighbor tells you how a certain over-the-counter product greatly improved his health. He had only a couple of mild colds last winter instead of the four or five bad ones he usually gets. Or you read a similar recommendation in somebody’s blog. They claim it’s wonderful and you really should try it. Do you rush right out and buy the stuff? I would wary, for several reasons:
a. The proponents of the product might be seeing just the placebo effect. People tend to get better when they think a medication is helping, even if that medication was in fact just a sugar pill.
b. Some bloggers are being paid to push certain products – and at present they are not required disclose that fact. This can even enhance the placebo effect since the blogger wants it to work so that he can legitimately get paid. He may not be deliberately dishonest, but he can easily fool himself.
c. Your informant may occasionally be deliberately dishonest. That is especially true if he is being paid to advertise that product.
d. People recover from most diseases on their own. The medication may have had nothing to do with the recovery.
e. Random chance plays a big part in things like sickness, as do changes in our bodies. Your neighbor may have just not been exposed to as many colds as usual, or his aging, more experienced body may have already learned to deal with the viruses he happened to meet last winter.
So how do you decide if you should try the medication or not? Or if you should accept some other product recommendation? There is no easy way, but some things can help. The first step is gathering information, that’s kind of obvious. The problem is, what information do you need and how do you know what to trust? The following questions are helpful. They are based on suggestions from the Center for Evaluative Clinical Studies (CECS) of the Dartmouth Medical School:
First, exactly what is the assertion? What does it claim and not claim? It is all too easy to unconsciously extrapolate to other situations where that assertion does not necessarily apply.
Second, would you care if the assertion were true? If you live in Hawaii you probably won’t care much about a product claiming to make a home heating system more efficient.
Third, who stands to benefit? The blogger or other advertiser is likely to be biased in favor of the product, consciously or unconsciously. Likewise the manufacturer or salesperson.
Forth, how good is the evidence? Does it come from multiple, independent studies? How good are those studies?
Now that last question can be difficult for a layperson to answer. Most people lack the background and training to evaluate such studies. However again we can build on CECS suggestions. We should ask ourselves:
What are the key elements of any study? What population? (A study showing a medication safe and effective in adults may not be meaningful to children, for example.) What exposure? What was the outcome in terms of both nature of effect and size of that effect?
Are the data relevant? Did researchers use the right study population and exposure? (Studies exposing rats to many times the equivalent dose human users might take should be considered suspect.) Is the effect large enough to be both statistically significant and meaningful to you?
Are the data valid? Is there a properly chosen comparison group? Were measurements done properly? Might the effect be due to chance?
Finally, are there other studies with similar findings to support the assertion? Are there studies that contradict the assertion?
That’s a lot of questions but they are all useful. If the stakes are low you might not want to pursue answers to all of them. However it is worthwhile to understand each question and know how it applies.
Of course most people don’t know how to evaluate statistical significance. That’s a big subject but I’ll try to make a start on it next time.
If you don’t like it, please tell me.
"It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Attributed to Mark Twain)
Your neighbor tells you how a certain over-the-counter product greatly improved his health. He had only a couple of mild colds last winter instead of the four or five bad ones he usually gets. Or you read a similar recommendation in somebody’s blog. They claim it’s wonderful and you really should try it. Do you rush right out and buy the stuff? I would wary, for several reasons:
a. The proponents of the product might be seeing just the placebo effect. People tend to get better when they think a medication is helping, even if that medication was in fact just a sugar pill.
b. Some bloggers are being paid to push certain products – and at present they are not required disclose that fact. This can even enhance the placebo effect since the blogger wants it to work so that he can legitimately get paid. He may not be deliberately dishonest, but he can easily fool himself.
c. Your informant may occasionally be deliberately dishonest. That is especially true if he is being paid to advertise that product.
d. People recover from most diseases on their own. The medication may have had nothing to do with the recovery.
e. Random chance plays a big part in things like sickness, as do changes in our bodies. Your neighbor may have just not been exposed to as many colds as usual, or his aging, more experienced body may have already learned to deal with the viruses he happened to meet last winter.
So how do you decide if you should try the medication or not? Or if you should accept some other product recommendation? There is no easy way, but some things can help. The first step is gathering information, that’s kind of obvious. The problem is, what information do you need and how do you know what to trust? The following questions are helpful. They are based on suggestions from the Center for Evaluative Clinical Studies (CECS) of the Dartmouth Medical School:
First, exactly what is the assertion? What does it claim and not claim? It is all too easy to unconsciously extrapolate to other situations where that assertion does not necessarily apply.
Second, would you care if the assertion were true? If you live in Hawaii you probably won’t care much about a product claiming to make a home heating system more efficient.
Third, who stands to benefit? The blogger or other advertiser is likely to be biased in favor of the product, consciously or unconsciously. Likewise the manufacturer or salesperson.
Forth, how good is the evidence? Does it come from multiple, independent studies? How good are those studies?
Now that last question can be difficult for a layperson to answer. Most people lack the background and training to evaluate such studies. However again we can build on CECS suggestions. We should ask ourselves:
What are the key elements of any study? What population? (A study showing a medication safe and effective in adults may not be meaningful to children, for example.) What exposure? What was the outcome in terms of both nature of effect and size of that effect?
Are the data relevant? Did researchers use the right study population and exposure? (Studies exposing rats to many times the equivalent dose human users might take should be considered suspect.) Is the effect large enough to be both statistically significant and meaningful to you?
Are the data valid? Is there a properly chosen comparison group? Were measurements done properly? Might the effect be due to chance?
Finally, are there other studies with similar findings to support the assertion? Are there studies that contradict the assertion?
That’s a lot of questions but they are all useful. If the stakes are low you might not want to pursue answers to all of them. However it is worthwhile to understand each question and know how it applies.
Of course most people don’t know how to evaluate statistical significance. That’s a big subject but I’ll try to make a start on it next time.
Subscribe to:
Posts (Atom)
