Saturday, September 18, 2004

Adoption

Some years ago, while getting ready to check out from Motel 6, I watched a bit of a movie starring Glenn Close. This film--probably entitled Immediate Family--judging from the Internet--dealt with a family seeking to complete an adoption. One line went: "Birth is nature's way. Adoption is God's way."

That was a very insightful perspective. It even touched the heart of this agnostic!


Tuesday, September 14, 2004

Dress sizes

For my consumer behavior class, I have assigned Why We Buy by Paco Underhill, the anthropologist turned retail consultant. In one chapter, he addresses the issue of making it easier for men to shop for clothing as gifts for their wives or girlfriends. Underhill suggests that it might be helpful for stores to offer women the opportunity to have their measurements on file so that men can shop knowing that the needed information is available.

This seems like a good idea. I realized that, even as a marketing professor, I would not know much about how dresses are measured. I recall references to dress sizes in three dimensions (specified simply in order, with the expectation that one would know what each dimension was). I decided to look up which dimensions are involved, and to my surpirse, height does not appear to be one of them!

I was surprised.

Saturday, September 11, 2004

Euro Cents Across Europe

An Associated Press story (see http://seattletimes.nwsource.com/html/nationworld/2002032765_nopenny11.html) today reports that retailers in many European countries are moving toward discarding the Euro one and two cent pieces. The practice of rounding to the nearest five cent piece is becoming increasingly common--except in Germany. Apparently, although some regions of Germany tend to be quite affluent, the desire for precision outweights the hassle of dealing with these small coints. This is despite the reality that the cost of manufacturing the one and two cent coins is higher than their value.

The story cites a poll from across the EU that 61% of respondents want to eliminate the one cent piece. Enthusiasm for rounding does, however, have its limits--only 55% favor doing away with the two cent piece.

A Gallup poll from the U.S. is cited, finding that "more than half" of respondents here favor doing away with the penny through rounding to the nearest nickel. Yet, it appears that "only about a quarter" of Americans actually spend their pennies, with most accumulating.


Friday, September 10, 2004

Election polls

With public opinion polls showing a tight race again in this year's Presidential election, it might be useful to take a look at how polls are conducted.

A poll conducted by Newsweek from September 2-3 shows 52% for Bush and 42% for Kerry among "likely voters." This poll has a "margin of error" 0f +/- 3%. It might be noted that this is actually one of the larger differences we have seen for a while. Had we been closer to a 50-50 split, it would have been more difficult to determine who was ahead. Even at 52% vs. 48%, we could not say with much certainty who was actually ahead.

Why such a large margin? The above survey was based on a survey of 1,008 respondents. With so many people, why can't we make more precise determinations?

Computing the margin of error is based on the idea of a "confidence interval" within statistics. If we tossed a coin ten times and got six heads, we would not have much reason to doubt that the coin was unbiased. On the other hand, if we threw 10,000 and got 60% heads, we would worry more.

To determine a confidence interval, we start by determining the "standard deviation" of the percentages for a given sample size. First, we need to make a slight modification since 52% and 42% figures do not add up t0 100%--people could vote for a third candidate or still be undecided. Therefore, we will base our calculations on those who are--and those who are not--supporting one of the candidates. For Bush, we have 52% for (p) and 48% "not for" (1-p).

The formula for the standard deviation of a "proportion"--or percentage--is

(p(1-p)/n)^0.5.

That is, the square root of the quantity p times 1-p over n.

To put in actual numbers:

(0.52(0.48)/1,008)^0.5 = 0.0157

To be 95% certain—an arbitrarily selected standard favored by many statisticians in the U.S.--that our confidence interval will cover the true proportion, we must go 1.96 standard deviations on either side of the sample proportion. (You would have to look up this number in a table). We thus have

1.96(0.0157) = 0.0308.

The 95% confidence interval for those supporting Bush is therefore 0.52 +/- 0.0308 or from 48.92% to 55.08%. Repeating the calculations for Kerry, his confidence interval turns out to be 38.95% to 45.05%.

It is interesting to observe that the margin of error is so large precisely because the two candidates are so close to each other. Note that, in the numerator of the formula for the standard deviation, the figures p and (1-p) figure prominently. 0.5(0.5) = 0.25, which is much larger than the (0.99) (0.01) = 0.0099 figures we would have if we were determining the margin of error for a 99% to 1% race. From an intuitive point of view, looking at any one voter, it is difficult to tell how he or she will vote in a 50-50 race. In a 99-1 race, not only is the sample outcome overwhelming, but on top of this, the margin of error will be much less because there is little doubt how each individual voter will behave.

In principle, we could reduce the margin of error proportionately by increasing the sample size. If we had 3,000 respondents instead of 1,008, the margin of error would be only about 1%. Sampling such larger groups would be more expensive, and because of the time sensitive nature of polling data, this increase could also result in delays in finishing the poll.

An important issue is conducting opinion polls is the question of who should be surveyed. The above survey is based on "likely voters." If the sampling base is all eligible--or at least all registered--voters, Kerry's figures appear to be higher. From the point of view of the representation of people in a democracy, that may be a more appropriate figure. However, it is a reality that only those voters who actually turn out will decide the election outcome. Thus, the estimate based on "likely voters" is likely to represent a more accurate prediction of actual election results if the election were held at the time of the election. It should also be noted that despite the best efforts, it is not possible in practice to identify a perfect sampling base of either all eligible voters or those most likely. Some individuals that should have been included will be omitted from any practical survey, and some that are not eligible will be included by mistake. Professional pollsters, however, do the best they can do minimize such biases.

Thursday, September 09, 2004

The genetics of autism

A recently released study hypothesizes that autism may be caused by displacement in sequence of genetic material. The details are rather complicated, and the model apparently requires at least two sequence displacements to work.

This may be a good opportunity to reflect on the genetics of autism. I was inspired to create a model that explains the different ways that genetics may lead to the manifestation of autism. This model can be viewed at http://www.larsperner.com/pdf/autism_genetics.pdf .

Impact of Genetics. Overall, genetics appear to have a strong influence on autism. Studies of concordance among identical twins--or the extent to which both will show autism if one does--have reached different results. One study puts the odds at around 90% while others suggest a rate more like 50%. This question is complicated by two factors--criteria for diagnosis and environmental impacts.

If one uses a strict definition of "autistic disorder," chances that two identical twins will share this diagnosis are smaller than if more flexible criteria are used. The fact that only one receives a diagnosis does not mean that the other has not inherited characteristics of autism. Studies show that even if an identical twin of a person with autism does not meet the criteria for a diagnosis, he or she is likely to show symptoms that may not reach the threshold for diagnosis. For example, the twin may have Asperger's Syndrome, PDD-NOS, or other characteristics that may fall short of warranting an outright diagnosis.

The shared impact of environment is an additional complicating factor. Ideal studies of identical twins study individuals who have been raised apart, but in autism, it is difficult to find sufficient numbers of people who are both identical twins and raised apart. To some extent, one can control for the environment by comparing observations to "fraternal" (or, to avoid being sexist, "sororal") twins. However, it is likely that identical twins will have their perceived identities greatly influenced by having a genetically identical other. This may be less important in studying conditions that do not appear to be related to personality, but with the ambiguity and subjective nature of autism, this will be a more significant problem here. It should also be pointed out that identical twins may actually have experienced significant differences in their environment, starring with their respective positions in the womb.

A One-Gene Model. First in the model, we see the possibility that one single gene could, in principle, be enough to cause autism. We do not know of such a gene, and several genes could individually result in autism.

Multiple-Gene Vulnerability Model. Another model, which Dr. Geraldine Dawson of the University of Washington discussed in her keynote address this summer at the conference of the Autism Society of America, suggests that many genes may be involved in autism. An individual may need to reach a certain threshold--that is, have at least a critical number of genetic conditions--before any symptoms are shown. Again, it is likely that a large number of genes are involved and that no single one of these genes is necessarily shared by all people on the autistic spectrum. For example, if, say, thirty genes are involved, one may, in the simplest model, need to have any five. It could be that some genes are more important, so that if certain "power genes" are affected, only three might be needed.

“Stressor” Genes in Combination With Autism Genes. Autism often inflicts a great deal of stress on the body. It is possible that certain genes that are not directly associated with autism may, reduce the body’s ability to cope with a smaller number of autism would not have been enough to reach a threshold for the manifestation of autism, reduce the body’s ability to “manage” complications from genes that could, by themselves, have been handled.

Again, for example, if stressor genes are present, it might take only three autism genes—rather than five normally required—for autism to “kick in.” Genetic susceptibility to the accumulation of heavy metals in the blood and intolerance of glutens (typically found among grain based foods) and/or casein (typically found in milk-based foods) are observed

Environmental Vulnerability in Combination With Genes. Another possibility is that the person may suffer from one or more genetic problems that, by themselves, would not lead to autism. However, when combined with autism genes, autism might result. Problems associated with autism often inflict a great deal of stress on the body, so if another "stressor" gene is present, the body may have fewer resources left to compensate for autism genes that could otherwise have been "managed." For example, even if genes that result in intolerance to glutens (found in certain kinds of breads) and casein (found in certain milk products) are present, this may leave the body more vulnerable to influence by more "direct" autism genes. Vulnerability to heavy metal accumulation could have a similar impact in combination with autism genes.

In a related situation, environmental complications--such as exposure to toxins, malnutrition, fever, or other illnesses--may stress the body in much the same way as non-autism genes. Again, fewer autism vulnerability genes may be needed to reach the needed threshold.

Gene Sequence Disruptions. A special case of a genetic cause of autism is disruptions in gene sequences as discussed in the beginning of this post. Here, the genetic information may be the same as in a non-affected individual, but the sequence of the genetic does may have been corrupted. This is roughly analogous to information that has been entered into a book or spreadsheet out of order. The effects of genetics appear to be expressed through the creation of proteins that are determined by the genetic code, and a small alteration in sequences may cause considerable reverberations throughout the protein outcome.

"Innocent" But Correlated Genes. One way to "hunt" for autism genes is to study the frequency with which various genes occur among affected individuals relative to rest of the population. Genes that are shared in high numbers are more likely to be implicated. The vastness of the human genome and the large costs of collecting complete data for a large number of people often makes it difficult to use this kind of "brute" force. A shortcut, however, involves a comparison of genetic conditions that have an identifiable physical manifestation. For example, if more people on the autistic spectrum were left handed, we might suspect a common gene. If we know the location of a condition that is observed frequently among people on the autistic spectrum, we can examine the frequencies of such genes in autistic and non-autistic groups. It turns out, however, that the correlation of genetics and traits does not always mean that the same genes are involved in two conditions. Occasionally, it will be reported in the news that a "gene marker" has been found for a particular condition. A gene marker is not a specifically identified gene, but rather a hypothesized gene that is believed to be located within a certain region on a specific gene. The reason why this result comes about has to do with the way that chromosomes are mixed in the reproduction process. Among chromosomes, physical fit between the chromosomes of each parent will be better in some areas than in others. Therefore, genes located near each other are more likely to be inherited together. Here, then, we have an illusory correlation. Autism and certain genetic traits may occur together, but the respective genetic traits do not necessarily cause--or even influence--the expression of autism.

Genetic Protection. One additional complication is that there is apparently a certain amount of "backup" material--or "spare" genes--that can be activated if certain genetic material that would otherwise be used is missing. (It can, however, also be a problem if both the default and the "backup" genes are "turned on.") To the extent that this backup material is consistently available across individuals, irrespective of other genetic differences, this would merely result in a gene being less important in causing autism. However, when this protection "kicks in" less consistently, more "noise" is introduced, obscuring the effects of individual genes.

Genetics Within a Broader Context. For another perspective on the dynamics of autism, see my integrative model at http://www.larsperner.com/pdf/overall_autism_model.pdf .

Conclusion. Autism is a perplexing phenomenon. Individual aspects of the condition are highly confusing by themselves, and many paradoxes have been observed. Genetics--even for relatively simple traits--is also a very challenging field, so it is not surprising that, when the two are considered together, drawing conclusions is difficult. With the mechanics of genetics being somewhat outside by area of knowledge, I cannot offer any credible prediciton on how soon--or whether--we breaktroughs in this area may be reached.

Wednesday, September 08, 2004

Astronomy and fiscal responsibility

The crash of the Genesis solar explorer capsule reminds me of a rather touchy issue that has caused me to agonize for many years. Personally, I am very interested in astronomy, but I wonder if it is fair to society as a whole--or to the world--to spend so much money on somethng that is, ultimately, mostly of theoretical interest for those of us who are curious.

Space exploration unquestionably results in technology that can be transferred to applications. Although I understand that it is a controversial question whether the first Moon explorations are genuinely responsible for the advent of teflon, I suspect that other materials science and aviation contributions can be significant. I am less sure that these come about in a cost effective manner.

As suggested by a prior post, I am particularly interested in planets in other solar systems. I can remember the days before we knew that any existed, and I will be particularly interested if we ever find more Earth-like planets. Ultimately, with other budgetary priorities--and with the environmental consequences of the space launches involved--I am wondering if this is a too expensive indulgence.

Tuesday, September 07, 2004

Another kind of role model

During his Presidency, it might be argued that, at least in some respects, Bill Clinton did not consistently serve as a good role model.

Now, Bill Clinton's recent heart surgery may have a significant influence on the behavior of many people who might otherwise go with undiagnosed heart problems. The former President, before the surgery, actually encouraged others to be checked, and because of the attention given to his case, will be more likely to be examined themselves. Apparently, when Fonzie obtained a library card in one episode of Happy Days, libraries experienced a dramatic increase in applications. It is less clear, however, how long the effects of Clinton's role modeling will last. The effect on people actually experiencing symptoms will probably last longer than it will for those who are "merely" in a high risk category.

Saturday, September 04, 2004

Super Marketing Man

I don't drink, I don't smoke, and I don't do drugs (tall people are naturally high, anyway). I do, however, have one vice: Internet domain names. Every now and then, I register a new one.

Yesterday, a colleague made an off-hand reference to the "Marketing Man." This inspired me to check if MarketingMan.com was available. It wasn't, and it probably went way back in the old days (i.e., around 1999 or 2000).

Today, on an off-chance, it occurred to me to check if SuperMarketingMan.com was available. It actually was! Now I have to decide what to post on that site. For once, I don't feel guilty about acquiring this one. It describes me rather well! Maybe I should write an essay on my contributions to the intellectual development of marketing. Perhaps an autobiography would be more suitable. What do y'all think?


Thursday, September 02, 2004

The chicken egg and the turkey egg

I wonder which one came first.


Smaller extra-Solar System planets

A newspaper article today reported that several planets of a size similar to those of Uranus and Neptune have now been found to orbit stars outside our solar system. Previously, only Saturn size planets had been detected.

For some reason, I am fascinated by the idea of planets orbiting stars outside our system. Planets discovered so far tend to be to big, and to orbit too closely to their parent stars, to be able to support organic life, but it is predicted that we will eventually be able to detect earth-sized planets.

One astronomer predicts that the Milky Way contains as many as two billion planets. I have no idea of the merits of this claim, but if this figure is true, I wish I owned a dollar for each one!

Demographics

A colleague of mine asked the students in his MBA level international marketing course about the current size of the U.S. population. Estimates ranged from 110 million to 1.5 billion.